mirror-github-bspeice/fractorium
1
0
Fork 0
mirror of https://bitbucket.org/mfeemster/fractorium.git synced 2025-02-01 18:40:12 -05:00
Code Issues Actions Packages Projects Releases Wiki Activity
04e72c27de
fractorium/Source/Ember/Variations05.h

3265 lines
109 KiB
C
Raw Normal View History

Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
#pragma once
#include "Variation.h"
namespace EmberNs
{
/// <summary>
/// bubble2.
/// </summary>
template <typename T>
class EMBER_API Bubble2Variation : public ParametricVariation<T>
{
public:
Bubble2Variation(T weight = 1.0) : ParametricVariation<T>("bubble2", VAR_BUBBLE2, weight, true)
{
Init();
}
PARVARCOPY(Bubble2Variation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T t = T(0.25) * (helper.m_PrecalcSumSquares + SQR(helper.In.z)) + 1;
T r = m_Weight / t;
helper.Out.x = helper.In.x * r * m_X;
helper.Out.y = helper.In.y * r * m_Y;
if (helper.In.z >= 0)
helper.Out.z = m_Weight * (helper.In.z + m_Z);
else
helper.Out.z = m_Weight * (helper.In.z - m_Z);
helper.Out.z += helper.In.z * r * m_Z;//The += is intentional.
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string z = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t t = (real_t)(0.25) * (precalcSumSquares + SQR(vIn.z)) + 1;\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t r = xform->m_VariationWeights[" << varIndex << "] / t;\n"
<< "\n"
<< "\t\tvOut.x = vIn.x * r * " << x << ";\n"
<< "\t\tvOut.y = vIn.y * r * " << y << ";\n"
<< "\n"
<< "\t\tif (vIn.z >= 0)\n"
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * (vIn.z + " << z << ");\n"
<< "\t\telse\n"
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * (vIn.z - " << z << ");\n"
<< "\n"
<< "\t\tvOut.z += vIn.z * r * " << z << ";\n"
<< "\t}\n";
return ss.str();
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "bubble2_x", 1));//Original used a prefix of bubble_, which is incompatible with Ember's design.
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "bubble2_y", 1));
m_Params.push_back(ParamWithName<T>(&m_Z, prefix + "bubble2_z"));
}
private:
T m_X;
T m_Y;
T m_Z;
};
/// <summary>
/// CircleLinear.
/// </summary>
template <typename T>
class EMBER_API CircleLinearVariation : public ParametricVariation<T>
{
public:
CircleLinearVariation(T weight = 1.0) : ParametricVariation<T>("CircleLinear", VAR_CIRCLELINEAR, weight)
{
Init();
}
PARVARCOPY(CircleLinearVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
int m = int(Floor<T>(T(0.5) * helper.In.x / m_Sc));
int n = int(Floor<T>(T(0.5) * helper.In.y / m_Sc));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T x = helper.In.x - (m * 2 + 1) * m_Sc;
T y = helper.In.y - (n * 2 + 1) * m_Sc;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
T u = Zeps(Hypot(x, y));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T v = (T(0.3) + T(0.7) * DiscreteNoise2(m + 10, n + 3)) * m_Sc;
T z1 = DiscreteNoise2(int(m + m_Seed), n);
if ((z1 < m_Dens1) && (u < v))
{
if (m_Reverse > 0)
{
if (z1 < m_Dens1 * m_Dens2)
{
x *= m_K;
y *= m_K;
}
else
{
T z = v / u * (1 - m_K) + m_K;
x *= z;
y *= z;
}
}
else
{
if (z1 > m_Dens1 * m_Dens2)
{
x *= m_K;
y *= m_K;
}
else
{
T z = v / u * (1 - m_K) + m_K;
x *= z;
y *= z;
}
}
}
helper.Out.x = m_Weight * (x + (m * 2 + 1) * m_Sc);
helper.Out.y = m_Weight * (y + (n * 2 + 1) * m_Sc);
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string sc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string k = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string dens1 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string dens2 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string reverse = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string seed = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tint m = (int)floor((real_t)(0.5) * vIn.x / " << sc << ");\n"
<< "\t\tint n = (int)floor((real_t)(0.5) * vIn.y / " << sc << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t x = vIn.x - (m * 2 + 1) * " << sc << ";\n"
<< "\t\treal_t y = vIn.y - (n * 2 + 1) * " << sc << ";\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\treal_t u = Zeps(Hypot(x, y));\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t v = ((real_t)(0.3) + (real_t)(0.7) * CircleLinearDiscreteNoise2(m + 10, n + 3)) * " << sc << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t z1 = CircleLinearDiscreteNoise2((int)(m + " << seed << "), n);\n"
<< "\n"
<< "\t\tif ((z1 < " << dens1 << ") && (u < v))\n"
<< "\t\t{\n"
<< "\t\t if (" << reverse << " > 0)\n"
<< "\t\t {\n"
<< "\t\t if (z1 < " << dens1 << " * " << dens2 << ")\n"
<< "\t\t {\n"
<< "\t\t x *= " << k << ";\n"
<< "\t\t y *= " << k << ";\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
<< "\t\t real_t z = v / u * (1 - " << k << ") + " << k << ";\n"
<< "\n"
<< "\t\t x *= z;\n"
<< "\t\t y *= z;\n"
<< "\t\t }\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
<< "\t\t if (z1 > " << dens1 << " * " << dens2 << ")\n"
<< "\t\t {\n"
<< "\t\t x *= " << k << ";\n"
<< "\t\t y *= " << k << ";\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
<< "\t\t real_t z = v / u * (1 - " << k << ") + " << k << ";\n"
<< "\n"
<< "\t\t x *= z;\n"
<< "\t\t y *= z;\n"
<< "\t\t }\n"
<< "\t\t }\n"
<< "\t\t}\n"
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * (x + (m * 2 + 1) * " << sc << ");\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * (y + (n * 2 + 1) * " << sc << ");\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLFuncsString() const
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
return
"real_t CircleLinearDiscreteNoise2(int x, int y)\n"
"{\n"
" const real_t im = 2147483647;\n"
" const real_t am = 1 / im;\n"
"\n"
" int n = x + y * 57;\n"
" n = (n << 13) ^ n;\n"
" return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;\n"
"}\n"
"\n";
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_Sc, prefix + "CircleLinear_Sc", 1, REAL_NONZERO));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_K, prefix + "CircleLinear_K", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_Dens1, prefix + "CircleLinear_Dens1", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_Dens2, prefix + "CircleLinear_Dens2", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_Reverse, prefix + "CircleLinear_Reverse", 1));
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "CircleLinear_X", 10));
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "CircleLinear_Y", 10));
m_Params.push_back(ParamWithName<T>(&m_Seed, prefix + "CircleLinear_Seed", 0, INTEGER));
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
T DiscreteNoise2(int x, int y)
{
const T im = T(2147483647);
const T am = (1 / im);
int n = x + y * 57;
n = (n << 13) ^ n;
return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;
}
T m_Sc;
T m_K;
T m_Dens1;
T m_Dens2;
T m_Reverse;
T m_X;
T m_Y;
T m_Seed;
};
/// <summary>
/// CircleRand.
/// The original would loop infinitely as x and y approached zero, so put a check for a max of 10 iters.
/// </summary>
template <typename T>
class EMBER_API CircleRandVariation : public ParametricVariation<T>
{
public:
CircleRandVariation(T weight = 1.0) : ParametricVariation<T>("CircleRand", VAR_CIRCLERAND, weight)
{
Init();
}
PARVARCOPY(CircleRandVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
intmax_t m, n, iters = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T x, y, u;
do
{
x = m_X * (1 - 2 * rand.Frand01<T>());
y = m_Y * (1 - 2 * rand.Frand01<T>());
m = Floor<T>(T(0.5) * x / m_Sc);
n = Floor<T>(T(0.5) * y / m_Sc);
x -= (m * 2 + 1) * m_Sc;
y -= (n * 2 + 1) * m_Sc;
u = Hypot(x, y);
if (++iters > 10)
break;
}
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
while ((DiscreteNoise2(int(m + m_Seed), int(n)) > m_Dens) || (u > (T(0.3) + T(0.7) * DiscreteNoise2(int(m + 10), int(n + 3))) * m_Sc));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = m_Weight * (x + (m * 2 + 1) * m_Sc);
helper.Out.y = m_Weight * (y + (n * 2 + 1) * m_Sc);
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string sc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string dens = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string seed = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\tint m, n, iters = 0;\n"
<< "\t\treal_t x, y, u;\n"
<< "\n"
<< "\t\tdo\n"
<< "\t\t{\n"
<< "\t\t x = " << x << " * (1 - 2 * MwcNext01(mwc));\n"
<< "\t\t y = " << y << " * (1 - 2 * MwcNext01(mwc));\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t m = (int)floor((real_t)(0.5) * x / " << sc << ");\n"
<< "\t\t n = (int)floor((real_t)(0.5) * y / " << sc << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t x = x - (m * 2 + 1) * " << sc << ";\n"
<< "\t\t y = y - (n * 2 + 1) * " << sc << ";\n"
<< "\t\t u = Hypot(x, y);\n"
<< "\n"
<< "\t\t if (++iters > 10)\n"
<< "\t\t break;\n"
<< "\t\t}\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\twhile ((CircleRandDiscreteNoise2((int)(m + " << seed << "), n) > " << dens << ") || (u > ((real_t)(0.3) + (real_t)(0.7) * CircleRandDiscreteNoise2(m + 10, n + 3)) * " << sc << "));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * (x + (m * 2 + 1) * " << sc << ");\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * (y + (n * 2 + 1) * " << sc << ");\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLFuncsString() const
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
return
"real_t CircleRandDiscreteNoise2(int x, int y)\n"
"{\n"
" const real_t im = 2147483647;\n"
" const real_t am = 1 / im;\n"
"\n"
" int n = x + y * 57;\n"
" n = (n << 13) ^ n;\n"
" return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;\n"
"}\n"
"\n";
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_Sc, prefix + "CircleRand_Sc", 1, REAL_NONZERO));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_Dens, prefix + "CircleRand_Dens", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "CircleRand_X", 10));
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "CircleRand_Y", 10));
m_Params.push_back(ParamWithName<T>(&m_Seed, prefix + "CircleRand_Seed", 0, INTEGER));
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
T DiscreteNoise2(int x, int y)
{
const T im = T(2147483647);
const T am = (1 / im);
int n = x + y * 57;
n = (n << 13) ^ n;
return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;
}
T m_Sc;
T m_Dens;
T m_X;
T m_Y;
T m_Seed;
};
/// <summary>
/// CircleTrans1.
/// The original would loop infinitely as x and y approached zero, so put a check for a max of 10 iters.
/// </summary>
template <typename T>
class EMBER_API CircleTrans1Variation : public ParametricVariation<T>
{
public:
CircleTrans1Variation(T weight = 1.0) : ParametricVariation<T>("CircleTrans1", VAR_CIRCLETRANS1, weight)
{
Init();
}
PARVARCOPY(CircleTrans1Variation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T ux, uy, u, x, y;
Trans(m_X, m_Y, helper.In.x, helper.In.y, &ux, &uy);
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
intmax_t m = Floor<T>(T(0.5) * ux / m_Sc);
intmax_t n = Floor<T>(T(0.5) * uy / m_Sc);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
x = ux - (m * 2 + 1) * m_Sc;
y = uy - (n * 2 + 1) * m_Sc;
u = Hypot(x, y);
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
if ((DiscreteNoise2(int(m + m_Seed), int(n)) > m_Dens) || (u > (T(0.3) + T(0.7) * DiscreteNoise2(int(m + 10), int(n + 3))) * m_Sc))
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ux = ux;
uy = uy;
}
else
{
CircleR(&ux, &uy, rand);
}
helper.Out.x = m_Weight * ux;
helper.Out.y = m_Weight * uy;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string sc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string dens = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string seed = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\treal_t ux, uy, u, x, y;\n"
<< "\n"
<< "\t\tCircleTrans1Trans(" << x << ", " << y << ", vIn.x, vIn.y, &ux, &uy);\n"
<< "\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tint m = (int)floor((real_t)(0.5) * ux / " << sc << ");\n"
<< "\t\tint n = (int)floor((real_t)(0.5) * uy / " << sc << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tx = ux - (m * 2 + 1) * " << sc << ";\n"
<< "\t\ty = uy - (n * 2 + 1) * " << sc << ";\n"
<< "\t\tu = Hypot(x, y);\n"
<< "\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tif ((CircleTrans1DiscreteNoise2((int)(m + " << seed << "), n) > " << dens << ") || (u > ((real_t)(0.3) + (real_t)(0.7) * CircleTrans1DiscreteNoise2(m + 10, n + 3)) * " << sc << "))\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t{\n"
<< "\t\t ux = ux;\n"
<< "\t\t uy = uy;\n"
<< "\t\t}\n"
<< "\t\telse\n"
<< "\t\t{\n"
<< "\t\t CircleTrans1CircleR(" << x << ", " << y << ", " << sc << ", " << seed << ", " << dens << ", &ux, &uy, mwc);\n"
<< "\t\t}\n"
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * ux;\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * uy;\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLFuncsString() const
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
return
"real_t CircleTrans1DiscreteNoise2(int x, int y)\n"
"{\n"
" const real_t im = 2147483647;\n"
" const real_t am = 1 / im;\n"
"\n"
" int n = x + y * 57;\n"
" n = (n << 13) ^ n;\n"
" return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;\n"
"}\n"
"\n"
"void CircleTrans1Trans(real_t a, real_t b, real_t x, real_t y, real_t* x1, real_t* y1)\n"
"{\n"
This is commit of Simon Detheridge's pull request "osx-opencl" with a few modifications. -If we change OpenCLWrapper to capture build log errors, we don't need to manually concat them in RendererCL because the overridden ErrorReport() function already concatenates the errors from both classes. -Do not define T in OpenCL programs. We already have real_t to handle this. -Do keep the casting to real_t. However this should not be necessary because there is a command line option to do this automatically which we already use: -cl-single-precision-constant. The only reason we do this is because the Apple OpenCL compiler does not follow the standard and obviously ignores this option. Absolutely awful. -Fix a few improper casts in the CircleTrans1 and GlynnSim1 variations. -Add an automated OpenCL program build tester to EmberTester, as well as a cast checker.
2015-03-22 15:46:10 -04:00
" *x1 = (x - a) * (real_t)(0.5) + a;\n"
" *y1 = (y - b) * (real_t)(0.5) + b;\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
"}\n"
"\n"
"void CircleTrans1CircleR(real_t mx, real_t my, real_t sc, real_t seed, real_t dens, real_t* ux, real_t* vy, uint2* mwc)\n"
"{\n"
" int m, n, iters = 0;\n"
" real_t x, y, alpha, u;\n"
"\n"
" do\n"
" {\n"
" x = fabs(mx) * (1 - 2 * MwcNext01(mwc));\n"
" y = fabs(my) * (1 - 2 * MwcNext01(mwc));\n"
This is commit of Simon Detheridge's pull request "osx-opencl" with a few modifications. -If we change OpenCLWrapper to capture build log errors, we don't need to manually concat them in RendererCL because the overridden ErrorReport() function already concatenates the errors from both classes. -Do not define T in OpenCL programs. We already have real_t to handle this. -Do keep the casting to real_t. However this should not be necessary because there is a command line option to do this automatically which we already use: -cl-single-precision-constant. The only reason we do this is because the Apple OpenCL compiler does not follow the standard and obviously ignores this option. Absolutely awful. -Fix a few improper casts in the CircleTrans1 and GlynnSim1 variations. -Add an automated OpenCL program build tester to EmberTester, as well as a cast checker.
2015-03-22 15:46:10 -04:00
" m = (int)floor((real_t)(0.5) * x / sc);\n"
" n = (int)floor((real_t)(0.5) * y / sc);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
" alpha = M_2PI * MwcNext01(mwc);\n"
This is commit of Simon Detheridge's pull request "osx-opencl" with a few modifications. -If we change OpenCLWrapper to capture build log errors, we don't need to manually concat them in RendererCL because the overridden ErrorReport() function already concatenates the errors from both classes. -Do not define T in OpenCL programs. We already have real_t to handle this. -Do keep the casting to real_t. However this should not be necessary because there is a command line option to do this automatically which we already use: -cl-single-precision-constant. The only reason we do this is because the Apple OpenCL compiler does not follow the standard and obviously ignores this option. Absolutely awful. -Fix a few improper casts in the CircleTrans1 and GlynnSim1 variations. -Add an automated OpenCL program build tester to EmberTester, as well as a cast checker.
2015-03-22 15:46:10 -04:00
" u = (real_t)(0.3) + (real_t)(0.7) * CircleTrans1DiscreteNoise2(m + 10, n + 3);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
" x = u * cos(alpha);\n"
" y = u * sin(alpha);\n"
"\n"
" if (++iters > 10)\n"
" break;\n"
" }\n"
" while (CircleTrans1DiscreteNoise2((int)(m + seed), n) > dens);\n"
"\n"
" *ux = x + (m * 2 + 1) * sc;\n"
" *vy = y + (n * 2 + 1) * sc;\n"
"}\n"
"\n";
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_Sc, prefix + "CircleTrans1_Sc", 1, REAL_NONZERO));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_Dens, prefix + "CircleTrans1_Dens", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "CircleTrans1_X", 10));
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "CircleTrans1_Y", 10));
m_Params.push_back(ParamWithName<T>(&m_Seed, prefix + "CircleTrans1_Seed", 0, INTEGER));
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
T DiscreteNoise2(int x, int y)
{
const T im = T(2147483647);
const T am = (1 / im);
int n = x + y * 57;
n = (n << 13) ^ n;
return ((n * (n * n * 15731 + 789221) + 1376312589) & 0x7fffffff) * am;
}
void Trans(T a, T b, T x, T y, T* x1, T* y1)
{
*x1 = (x - a) * T(0.5) + a;
*y1 = (y - b) * T(0.5) + b;
}
void CircleR(T* ux, T* vy, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand)
{
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
intmax_t m, n, iters = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T x, y, alpha, u;
do
{
x = fabs(m_X) * (1 - 2 * rand.Frand01<T>());
y = fabs(m_Y) * (1 - 2 * rand.Frand01<T>());
m = Floor<T>(T(0.5) * x / m_Sc);
n = Floor<T>(T(0.5) * y / m_Sc);
alpha = M_2PI * rand.Frand01<T>();
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
u = T(0.3) + T(0.7) * DiscreteNoise2(int(m + 10), int(n + 3));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
x = u * cos(alpha);
y = u * sin(alpha);
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
if (++iters > 10)
break;
}
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
while (DiscreteNoise2(int(m + m_Seed), int(n)) > m_Dens);
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
*ux = x + (m * 2 + 1) * m_Sc;
*vy = y + (n * 2 + 1) * m_Sc;
}
T m_Sc;
T m_Dens;
T m_X;
T m_Y;
T m_Seed;
};
/// <summary>
/// cubic3D.
/// </summary>
template <typename T>
class EMBER_API Cubic3DVariation : public ParametricVariation<T>
{
public:
Cubic3DVariation(T weight = 1.0) : ParametricVariation<T>("cubic3D", VAR_CUBIC3D, weight)
{
Init();
}
PARVARCOPY(Cubic3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
int useNode = rand.Rand() & 7;//Faster than % 8.
T exnze, wynze, znxy;
T lattd = m_Weight * T(0.5);
T px, py, pz;
exnze = 1 - (m_SmoothStyle * (1 - (cos(atan2(helper.In.x, helper.In.z)))));
wynze = 1 - (m_SmoothStyle * (1 - (sin(atan2(helper.In.y, helper.In.z)))));
if (m_SmoothStyle > 1)
znxy = 1 - (m_SmoothStyle * (1 - ((exnze + wynze) / 2 * m_SmoothStyle)));
else
znxy = 1 - (m_SmoothStyle * (1 - ((exnze + wynze) * T(0.5))));
if (m_VarType == VARTYPE_PRE)
{
px = helper.In.x;
py = helper.In.y;
pz = helper.In.z;
}
else
{
px = outPoint.m_X;
py = outPoint.m_Y;
pz = outPoint.m_Z;
}
switch (useNode)
{
case 0 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) + lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) + lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) + lattd;
break;
case 1 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) + lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) - lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) + lattd;
break;
case 2 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) + lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) + lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) - lattd;
break;
case 3 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) + lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) - lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) - lattd;
break;
case 4 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) - lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) + lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) + lattd;
break;
case 5 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) - lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) - lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) + lattd;
break;
case 6 :
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) - lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) + lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) - lattd;
break;
case 7 :
More C++11
2014-09-10 01:41:26 -04:00
default:
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = ((px - (m_Smooth * (1 - m_Fill) * px * exnze)) + (helper.In.x * m_Smooth * m_Fill * exnze)) - lattd;
helper.Out.y = ((py - (m_Smooth * (1 - m_Fill) * py * wynze)) + (helper.In.y * m_Smooth * m_Fill * wynze)) - lattd;
helper.Out.z = ((pz - (m_Smooth * (1 - m_Fill) * pz * znxy)) + (helper.In.z * m_Smooth * m_Fill * znxy)) - lattd;
break;
}
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string xpand = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string style = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string fill = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string smooth = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string smoothStyle = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\tint useNode = MwcNext(mwc) & 7;\n"
<< "\t\treal_t exnze, wynze, znxy;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t lattd = xform->m_VariationWeights[" << varIndex << "] * (real_t)(0.5);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t px, py, pz;\n"
<< "\n"
<< "\t\texnze = 1 - (" << smoothStyle << " * (1 - (cos(atan2(vIn.x, vIn.z)))));\n"
<< "\t\twynze = 1 - (" << smoothStyle << " * (1 - (sin(atan2(vIn.y, vIn.z)))));\n"
<< "\n"
<< "\t\tif (" << smoothStyle << " > 1)\n"
<< "\t\t znxy = 1 - (" << smoothStyle << " * (1 - ((exnze + wynze) / 2 * " << smoothStyle << ")));\n"
<< "\t\telse\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t znxy = 1 - (" << smoothStyle << " * (1 - ((exnze + wynze) * (real_t)(0.5))));\n";
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
if (m_VarType == VARTYPE_PRE)
{
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
ss <<
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
"\t\tpx = vIn.x;\n"
"\t\tpy = vIn.y;\n"
"\t\tpz = vIn.z;\n";
}
else
{
ss <<
"\t\tpx = outPoint->m_X;\n"
"\t\tpy = outPoint->m_Y;\n"
"\t\tpz = outPoint->m_Z;\n";
}
ss <<
"\t\tswitch (useNode)\n"
"\t\t{\n"
"\t\t case 0 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) + lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) + lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) + lattd;\n"
"\t\t break;\n"
"\t\t case 1 :\n"
"\t\t vOut.x = ((px - (" << smooth << " *(1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) + lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " *(1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) - lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " *(1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) + lattd;\n"
"\t\t break;\n"
"\t\t case 2 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) + lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) + lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) - lattd;\n"
"\t\t break;\n"
"\t\t case 3 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) + lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) - lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) - lattd;\n"
"\t\t break;\n"
"\t\t case 4 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) - lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) + lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) + lattd;\n"
"\t\t break;\n"
"\t\t case 5 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) - lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) - lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) + lattd;\n"
"\t\t break;\n"
"\t\t case 6 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) - lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) + lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) - lattd;\n"
"\t\t break;\n"
"\t\t case 7 :\n"
"\t\t vOut.x = ((px - (" << smooth << " * (1 - " << fill << ") * px * exnze)) + (vIn.x * " << smooth << " * " << fill << " * exnze)) - lattd;\n"
"\t\t vOut.y = ((py - (" << smooth << " * (1 - " << fill << ") * py * wynze)) + (vIn.y * " << smooth << " * " << fill << " * wynze)) - lattd;\n"
"\t\t vOut.z = ((pz - (" << smooth << " * (1 - " << fill << ") * pz * znxy)) + (vIn.z * " << smooth << " * " << fill << " * znxy)) - lattd;\n"
"\t\t break;\n"
"\t\t}\n"
"\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
if (fabs(m_Xpand) <= 1)
m_Fill = m_Xpand * T(0.5);
else
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
m_Fill = std::sqrt(m_Xpand) * T(0.5);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
if (fabs(m_Weight) <= T(0.5))
m_Smooth = m_Weight * 2;//Causes full effect above m_Weight = 0.5.
else
m_Smooth = 1;
if (fabs(m_Style) <= 1)
{
m_SmoothStyle = m_Style;
}
else
{
if (m_Style > 1)
m_SmoothStyle = 1 + (m_Style - 1) * T(0.25);
else
m_SmoothStyle = (m_Style + 1) * T(0.25) - 1;
}
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Xpand, prefix + "cubic3D_xpand", T(0.25)));
m_Params.push_back(ParamWithName<T>(&m_Style, prefix + "cubic3D_style"));
m_Params.push_back(ParamWithName<T>(true, &m_Fill, prefix + "cubic3D_fill"));//Precalc.
m_Params.push_back(ParamWithName<T>(true, &m_Smooth, prefix + "cubic3D_smooth"));
m_Params.push_back(ParamWithName<T>(true, &m_SmoothStyle, prefix + "cubic3D_smooth_style"));
}
private:
T m_Xpand;
T m_Style;
T m_Fill;//Precalc.
T m_Smooth;
T m_SmoothStyle;
};
/// <summary>
/// cubicLattice_3D.
/// </summary>
template <typename T>
class EMBER_API CubicLattice3DVariation : public ParametricVariation<T>
{
public:
CubicLattice3DVariation(T weight = 1.0) : ParametricVariation<T>("cubicLattice_3D", VAR_CUBIC_LATTICE3D, weight)
{
Init();
}
PARVARCOPY(CubicLattice3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
int useNode = rand.Rand() & 7;//Faster than % 8.
T exnze, wynze, znxy, px, py, pz, lattd = m_Weight;
if (m_Style == 2)
{
exnze = cos(atan2(helper.In.x, helper.In.z));
wynze = sin(atan2(helper.In.y, helper.In.z));
znxy = (exnze + wynze) * T(0.5);
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
}
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
else
{
exnze = 1;
wynze = 1;
znxy = 1;
}
if (m_VarType == VARTYPE_PRE)
{
px = helper.In.x;
py = helper.In.y;
pz = helper.In.z;
}
else
{
px = outPoint.m_X;
py = outPoint.m_Y;
pz = outPoint.m_Z;
}
T pxtx = px + helper.In.x;
T pyty = py + helper.In.y;
T pztz = pz + helper.In.z;
switch (useNode)
{
case 0 :
helper.Out.x = pxtx * m_Fill * exnze + lattd;
helper.Out.y = pyty * m_Fill * wynze + lattd;
helper.Out.z = pztz * m_Fill * znxy + lattd;
break;
case 1 :
helper.Out.x = pxtx * m_Fill * exnze + lattd;
helper.Out.y = pyty * m_Fill * wynze - lattd;
helper.Out.z = pztz * m_Fill * znxy + lattd;
break;
case 2 :
helper.Out.x = pxtx * m_Fill * exnze + lattd;
helper.Out.y = pyty * m_Fill * wynze + lattd;
helper.Out.z = pztz * m_Fill * znxy - lattd;
break;
case 3 :
helper.Out.x = pxtx * m_Fill * exnze + lattd;
helper.Out.y = pyty * m_Fill * wynze - lattd;
helper.Out.z = pztz * m_Fill * znxy - lattd;
break;
case 4 :
helper.Out.x = pxtx * m_Fill * exnze - lattd;
helper.Out.y = pyty * m_Fill * wynze + lattd;
helper.Out.z = pztz * m_Fill * znxy + lattd;
break;
case 5 :
helper.Out.x = pxtx * m_Fill * exnze - lattd;
helper.Out.y = pyty * m_Fill * wynze - lattd;
helper.Out.z = pztz * m_Fill * znxy + lattd;
break;
case 6 :
helper.Out.x = pxtx * m_Fill * exnze - lattd;
helper.Out.y = pyty * m_Fill * wynze + lattd;
helper.Out.z = pztz * m_Fill * znxy - lattd;
break;
case 7 :
More C++11
2014-09-10 01:41:26 -04:00
default:
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = pxtx * m_Fill * exnze - lattd;
helper.Out.y = pyty * m_Fill * wynze - lattd;
helper.Out.z = pztz * m_Fill * znxy - lattd;
break;
}
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string xpand = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string style = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string fill = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\tint useNode = MwcNext(mwc) & 7;\n"
<< "\t\treal_t exnze, wynze, znxy, px, py, pz, lattd = xform->m_VariationWeights[" << varIndex << "];\n"
<< "\n"
<< "\t\tif (" << style << " == 2)\n"
<< "\t\t{\n"
<< "\t\t exnze = cos(atan2(vIn.x, vIn.z));\n"
<< "\t\t wynze = sin(atan2(vIn.y, vIn.z));\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t znxy = (exnze + wynze) * (real_t)(0.5);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t}\n"
<< "\t\telse\n"
<< "\t\t{\n"
<< "\t\t exnze = 1;\n"
<< "\t\t wynze = 1;\n"
<< "\t\t znxy = 1;\n"
<< "\t\t}\n";
if (m_VarType == VARTYPE_PRE)
{
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
ss <<
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
"\t\tpx = vIn.x;\n"
"\t\tpy = vIn.y;\n"
"\t\tpz = vIn.z;\n";
}
else
{
ss <<
"\t\tpx = outPoint->m_X;\n"
"\t\tpy = outPoint->m_Y;\n"
"\t\tpz = outPoint->m_Z;\n";
}
ss << "\t\treal_t pxtx = px + vIn.x;\n"
<< "\t\treal_t pyty = py + vIn.y;\n"
<< "\t\treal_t pztz = pz + vIn.z;\n"
<< "\n"
<< "\t\tswitch (useNode)\n"
<< "\t\t{\n"
<< "\t\t case 0 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze + lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze + lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy + lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 1 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze + lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze - lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy + lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 2 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze + lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze + lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy - lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 3 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze + lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze - lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy - lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 4 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze - lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze + lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy + lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 5 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze - lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze - lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy + lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 6 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze - lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze + lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy - lattd;\n"
<< "\t\t break;\n"
<< "\t\t case 7 :\n"
<< "\t\t vOut.x = pxtx * " << fill << " * exnze - lattd;\n"
<< "\t\t vOut.y = pyty * " << fill << " * wynze - lattd;\n"
<< "\t\t vOut.z = pztz * " << fill << " * znxy - lattd;\n"
<< "\t\t break;\n"
<< "\t\t}\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
if (fabs(m_Xpand) <= 1)
m_Fill = m_Xpand * T(0.5);
else
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
m_Fill = std::sqrt(m_Xpand) * T(0.5);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Xpand, prefix + "cubicLattice_3D_xpand", T(0.2)));//Original used a prefix of cubic3D_, which is incompatible with Ember's design.
m_Params.push_back(ParamWithName<T>(&m_Style, prefix + "cubicLattice_3D_style", 1, INTEGER, 1, 2));
m_Params.push_back(ParamWithName<T>(true, &m_Fill, prefix + "cubicLattice_3D_fill"));//Precalc.
}
private:
T m_Xpand;
T m_Style;
T m_Fill;//Precalc.
};
/// <summary>
/// foci_3D.
/// </summary>
template <typename T>
class EMBER_API Foci3DVariation : public Variation<T>
{
public:
Foci3DVariation(T weight = 1.0) : Variation<T>("foci_3D", VAR_FOCI3D, weight, false, false, false, false, true) { }
VARCOPY(Foci3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T expx = exp(helper.In.x) * T(0.5);
T expnx = T(0.25) / expx;
T boot = helper.In.z == 0 ? helper.m_PrecalcAtanyx : helper.In.z;
T tmp = m_Weight / (expx + expnx - (cos(helper.In.y) * cos(boot)));
helper.Out.x = (expx - expnx) * tmp;
helper.Out.y = sin(helper.In.y) * tmp;
helper.Out.z = sin(boot) * tmp;
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t expx = exp(vIn.x) * (real_t)(0.5);\n"
<< "\t\treal_t expnx = (real_t)(0.25) / expx;\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t boot = vIn.z == 0 ? precalcAtanyx : vIn.z;\n"
<< "\t\treal_t tmp = xform->m_VariationWeights[" << varIndex << "] / (expx + expnx - (cos(vIn.y) * cos(boot)));\n"
<< "\n"
<< "\t\tvOut.x = (expx - expnx) * tmp;\n"
<< "\t\tvOut.y = sin(vIn.y) * tmp;\n"
<< "\t\tvOut.z = sin(boot) * tmp;\n"
<< "\t}\n";
return ss.str();
}
};
/// <summary>
/// ho.
/// </summary>
template <typename T>
class EMBER_API HoVariation : public ParametricVariation<T>
{
public:
HoVariation(T weight = 1.0) : ParametricVariation<T>("ho", VAR_HO, weight)
{
Init();
}
PARVARCOPY(HoVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T uu = SQR(helper.In.x);
T vv = SQR(helper.In.y);
T ww = SQR(helper.In.z);
T atOmegaX = atan2(vv, ww);
T atOmegaY = atan2(uu, ww);
T su = sin(helper.In.x);
T cu = cos(helper.In.x);
T sv = sin(helper.In.y);
T cv = cos(helper.In.y);
T cucv = cu * cv;
T sucv = su * cv;
0.4.0.9 Beta 07/27/2014 0.4.0.9 Beta 07/27/2014 --User Changes Properly set tab order on all controls. Calculate and report iters/second in the final render dialog. Immediately draw yellow dot on xform mouse down on previously unselected xform. --Bug Fixes Fix GlynnSim1, GlynnSim2, GlynnSim3 and juliaNab by ensuring the first argument to pow() is >= 0. Ensure OpenCL platform and device combo boxes in the final render dialog expand as needed. --Code Changes Make VariationTreeSpinbox take its parent VariationTreeWidgetItem as a constructor argument. This makes SetupVariationTree() and VariationSpinBoxValueChanged() more efficient. Make Interference2 and ho use fabs().
2014-07-28 01:25:38 -04:00
T x = pow(fabs(cucv), m_XPow) + (cucv * m_XPow) + (T(0.25) * atOmegaX);//Must fabs first argument to pow, because negative values will return NaN.
T y = pow(fabs(sucv), m_YPow) + (sucv * m_YPow) + (T(0.25) * atOmegaY);//Original did not do this and would frequently return bad values.
T z = pow(fabs(sv), m_ZPow) + sv * m_ZPow;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = m_Weight * x;
helper.Out.y = m_Weight * y;
helper.Out.z = m_Weight * z;
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string xpow = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ypow = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string zpow = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\treal_t uu = SQR(vIn.x);\n"
<< "\t\treal_t vv = SQR(vIn.y);\n"
<< "\t\treal_t ww = SQR(vIn.z);\n"
<< "\t\treal_t atOmegaX = atan2(vv, ww);\n"
<< "\t\treal_t atOmegaY = atan2(uu, ww);\n"
<< "\t\treal_t su = sin(vIn.x);\n"
<< "\t\treal_t cu = cos(vIn.x);\n"
<< "\t\treal_t sv = sin(vIn.y);\n"
<< "\t\treal_t cv = cos(vIn.y);\n"
<< "\t\treal_t cucv = cu * cv;\n"
<< "\t\treal_t sucv = su * cv;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t x = pow(fabs(cucv), " << xpow << ") + (cucv * " << xpow << ") + ((real_t)(0.25) * atOmegaX);\n"
<< "\t\treal_t y = pow(fabs(sucv), " << ypow << ") + (sucv * " << ypow << ") + ((real_t)(0.25) * atOmegaY);\n"
0.4.0.9 Beta 07/27/2014 0.4.0.9 Beta 07/27/2014 --User Changes Properly set tab order on all controls. Calculate and report iters/second in the final render dialog. Immediately draw yellow dot on xform mouse down on previously unselected xform. --Bug Fixes Fix GlynnSim1, GlynnSim2, GlynnSim3 and juliaNab by ensuring the first argument to pow() is >= 0. Ensure OpenCL platform and device combo boxes in the final render dialog expand as needed. --Code Changes Make VariationTreeSpinbox take its parent VariationTreeWidgetItem as a constructor argument. This makes SetupVariationTree() and VariationSpinBoxValueChanged() more efficient. Make Interference2 and ho use fabs().
2014-07-28 01:25:38 -04:00
<< "\t\treal_t z = pow(fabs(sv), " << zpow << ") + sv * " << zpow << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * x;\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * y;\n"
<< "\t\tvOut.z = xform->m_VariationWeights[" << varIndex << "] * z;\n"
<< "\t}\n";
return ss.str();
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_XPow, prefix + "ho_xpow", 3));
m_Params.push_back(ParamWithName<T>(&m_YPow, prefix + "ho_ypow", 3));
m_Params.push_back(ParamWithName<T>(&m_ZPow, prefix + "ho_zpow", 3));
}
private:
T m_XPow;
T m_YPow;
T m_ZPow;
};
/// <summary>
/// Julia3Dq.
/// </summary>
template <typename T>
class EMBER_API Julia3DqVariation : public ParametricVariation<T>
{
public:
Julia3DqVariation(T weight = 1.0) : ParametricVariation<T>("julia3Dq", VAR_JULIA3DQ, weight, true, true, false, false, true)
{
Init();
}
PARVARCOPY(Julia3DqVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T temp = helper.m_PrecalcAtanyx * m_InvPower + rand.Rand() * m_InvPower2pi;
T sina = sin(temp);
T cosa = cos(temp);
T z = helper.In.z * m_AbsInvPower;
T r2d = helper.m_PrecalcSumSquares;
T r = m_Weight * pow(r2d + SQR(z), m_HalfInvPower);
T rsss = r * helper.m_PrecalcSqrtSumSquares;
helper.Out.x = rsss * cosa;
helper.Out.y = rsss * sina;
helper.Out.z = r * z;
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string power = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string divisor = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string invPower = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string absInvPower = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string halfInvPower = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string invPower2pi = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\treal_t temp = precalcAtanyx * " << invPower << " + MwcNext(mwc) * " << invPower2pi << ";\n"
<< "\t\treal_t sina = sin(temp);\n"
<< "\t\treal_t cosa = cos(temp);\n"
<< "\t\treal_t z = vIn.z * " << absInvPower << ";\n"
<< "\t\treal_t r2d = precalcSumSquares;\n"
<< "\t\treal_t r = xform->m_VariationWeights[" << varIndex << "] * pow(r2d + SQR(z), " << halfInvPower << ");\n"
<< "\t\treal_t rsss = r * precalcSqrtSumSquares;\n"
<< "\n"
<< "\t\tvOut.x = rsss * cosa;\n"
<< "\t\tvOut.y = rsss * sina;\n"
<< "\t\tvOut.z = r * z;\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_InvPower = m_Divisor / m_Power;
m_AbsInvPower = fabs(m_InvPower);
m_HalfInvPower = T(0.5) * m_InvPower - T(0.5);
m_InvPower2pi = M_2PI / m_Power;
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Power, prefix + "julia3Dq_power", 3, INTEGER_NONZERO));
m_Params.push_back(ParamWithName<T>(&m_Divisor, prefix + "julia3Dq_divisor", 2, INTEGER_NONZERO));
m_Params.push_back(ParamWithName<T>(true, &m_InvPower, prefix + "julia3Dq_inv_power"));//Precalc.
m_Params.push_back(ParamWithName<T>(true, &m_AbsInvPower, prefix + "julia3Dq_abs_inv_power"));
m_Params.push_back(ParamWithName<T>(true, &m_HalfInvPower, prefix + "julia3Dq_half_inv_power"));
m_Params.push_back(ParamWithName<T>(true, &m_InvPower2pi, prefix + "julia3Dq_inv_power_2pi"));
}
private:
T m_Power;
T m_Divisor;
T m_InvPower;//Precalc.
T m_AbsInvPower;
T m_HalfInvPower;
T m_InvPower2pi;
};
/// <summary>
/// line.
/// </summary>
template <typename T>
class EMBER_API LineVariation : public ParametricVariation<T>
{
public:
LineVariation(T weight = 1.0) : ParametricVariation<T>("line", VAR_LINE, weight)
{
Init();
}
PARVARCOPY(LineVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T r = rand.Frand01<T>() * m_Weight;
helper.Out.x = m_Ux * r;
helper.Out.y = m_Uy * r;
helper.Out.z = m_Uz * r;
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string delta = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string phi = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ux = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string uy = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string uz = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\treal_t r = MwcNext01(mwc) * xform->m_VariationWeights[" << varIndex << "];\n"
<< "\n"
<< "\t\tvOut.x = " << ux << " * r;\n"
<< "\t\tvOut.y = " << uy << " * r;\n"
<< "\t\tvOut.z = " << uz << " * r;\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
//Unit vector of the line.
m_Ux = cos(m_Delta * T(M_PI)) * cos(m_Phi * T(M_PI));
m_Uy = sin(m_Delta * T(M_PI)) * cos(m_Phi * T(M_PI));
m_Uz = sin(m_Phi * T(M_PI));
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
T r = std::sqrt(SQR(m_Ux) + SQR(m_Uy) + SQR(m_Uz));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
//Normalize.
m_Ux /= r;
m_Uy /= r;
m_Uz /= r;
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Delta, prefix + "line_delta"));
m_Params.push_back(ParamWithName<T>(&m_Phi, prefix + "line_phi"));
m_Params.push_back(ParamWithName<T>(true, &m_Ux, prefix + "line_ux"));//Precalc.
m_Params.push_back(ParamWithName<T>(true, &m_Uy, prefix + "line_uy"));
m_Params.push_back(ParamWithName<T>(true, &m_Uz, prefix + "line_uz"));
}
private:
T m_Delta;
T m_Phi;
T m_Ux;//Precalc.
T m_Uy;
T m_Uz;
};
/// <summary>
/// loonie_3D.
/// </summary>
template <typename T>
class EMBER_API Loonie3DVariation : public ParametricVariation<T>
{
public:
Loonie3DVariation(T weight = 1.0) : ParametricVariation<T>("loonie_3D", VAR_LOONIE3D, weight, true, false, false, false, true)
{
Init();
}
PARVARCOPY(Loonie3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T kikr = helper.m_PrecalcAtanyx;
T efTez = helper.In.z == 0 ? kikr : helper.In.z;
T r2 = helper.m_PrecalcSumSquares + SQR(efTez);
if (r2 < m_Vv)
{
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
T r = m_Weight * std::sqrt(m_Vv / r2 - 1);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = r * helper.In.x;
helper.Out.y = r * helper.In.y;
helper.Out.z = r * efTez * T(0.5);
}
else
{
helper.Out.x = m_Weight * helper.In.x;
helper.Out.y = m_Weight * helper.In.y;
helper.Out.z = m_Weight * efTez * T(0.5);
}
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string vv = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\treal_t kikr = precalcAtanyx;\n"
<< "\t\treal_t efTez = vIn.z == 0 ? kikr : vIn.z;\n"
<< "\t\treal_t r2 = precalcSumSquares + SQR(efTez);\n"
<< "\n"
<< "\t\tif (r2 < " << vv << ")\n"
<< "\t\t{\n"
<< "\t\t real_t r = xform->m_VariationWeights[" << varIndex << "] * sqrt(" << vv << " / r2 - 1);\n"
<< "\n"
<< "\t\t vOut.x = r * vIn.x;\n"
<< "\t\t vOut.y = r * vIn.y;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.z = r * efTez * (real_t)(0.5);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t}\n"
<< "\t\telse\n"
<< "\t\t{\n"
<< "\t\t vOut.x = xform->m_VariationWeights[" << varIndex << "] * vIn.x;\n"
<< "\t\t vOut.y = xform->m_VariationWeights[" << varIndex << "] * vIn.y;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * efTez * (real_t)(0.5);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t}\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_Vv = SQR(m_Weight);
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(true, &m_Vv, prefix + "loonie_3D_vv"));//Precalcs only, no params.
}
private:
T m_Vv;//Precalcs only, no params.
};
/// <summary>
/// mcarpet.
/// </summary>
template <typename T>
class EMBER_API McarpetVariation : public ParametricVariation<T>
{
public:
McarpetVariation(T weight = 1.0) : ParametricVariation<T>("mcarpet", VAR_MCARPET, weight, true)
{
Init();
}
PARVARCOPY(McarpetVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T t = helper.m_PrecalcSumSquares * T(0.25) + 1;
T r = m_Weight / t;
helper.Out.x = helper.In.x * r * m_X;
helper.Out.y = helper.In.y * r * m_Y;
helper.Out.x += (1 - (m_Twist * SQR(helper.In.x)) + helper.In.y) * m_Weight;//The += is intentional.
helper.Out.y += m_Tilt * helper.In.x * m_Weight;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string twist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string tilt = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t t = precalcSumSquares * (real_t)(0.25) + 1;\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t r = xform->m_VariationWeights[" << varIndex << "] / t;\n"
<< "\n"
<< "\t\tvOut.x = vIn.x * r * " << x << ";\n"
<< "\t\tvOut.y = vIn.y * r * " << y << ";\n"
<< "\t\tvOut.x += (1 - (" << twist << " * SQR(vIn.x)) + vIn.y) * xform->m_VariationWeights[" << varIndex << "];\n"
<< "\t\tvOut.y += " << tilt << " * vIn.x * xform->m_VariationWeights[" << varIndex << "];\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "mcarpet_x"));
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "mcarpet_y"));
m_Params.push_back(ParamWithName<T>(&m_Twist, prefix + "mcarpet_twist"));
m_Params.push_back(ParamWithName<T>(&m_Tilt, prefix + "mcarpet_tilt"));
}
private:
T m_X;
T m_Y;
T m_Twist;
T m_Tilt;
};
/// <summary>
/// waves2_3D.
/// Original used a precalc for the input points, but it doesn't
/// work with Ember's design, so it gets calculated on every iter
/// which is slightly slower.
/// </summary>
template <typename T>
class EMBER_API Waves23DVariation : public ParametricVariation<T>
{
public:
Waves23DVariation(T weight = 1.0) : ParametricVariation<T>("waves2_3D", VAR_WAVES23D, weight)
{
Init();
}
PARVARCOPY(Waves23DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T avgxy = (helper.In.x + helper.In.y) * T(0.5);
helper.Out.x = m_Weight * (helper.In.x + m_Scale * sin(helper.In.y * m_Freq));
helper.Out.y = m_Weight * (helper.In.y + m_Scale * sin(helper.In.x * m_Freq));
helper.Out.z = m_Weight * (helper.In.z + m_Scale * sin(avgxy * m_Freq));//Averages the XY to get Z.
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string freq = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string scale = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t avgxy = (vIn.x + vIn.y) * (real_t)(0.5);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * (vIn.x + " << scale << " * sin(vIn.y * " << freq << "));\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * (vIn.y + " << scale << " * sin(vIn.x * " << freq << "));\n"
<< "\t\tvOut.z = xform->m_VariationWeights[" << varIndex << "] * (vIn.z + " << scale << " * sin(avgxy * " << freq << "));\n"
<< "\t}\n";
return ss.str();
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Freq, prefix + "waves2_3D_freq", 2));
m_Params.push_back(ParamWithName<T>(&m_Scale, prefix + "waves2_3D_scale", 1));
}
private:
T m_Freq;
T m_Scale;
};
/// <summary>
/// Pie3D.
/// </summary>
template <typename T>
class EMBER_API Pie3DVariation : public ParametricVariation<T>
{
public:
Pie3DVariation(T weight = 1.0) : ParametricVariation<T>("pie3D", VAR_PIE3D, weight)
{
Init();
}
PARVARCOPY(Pie3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
int sl = int(rand.Frand01<T>() * m_Slices + T(0.5));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T a = m_Rotation + M_2PI * (sl + rand.Frand01<T>() * m_Thickness) / m_Slices;
T r = m_Weight * rand.Frand01<T>();
helper.Out.x = r * cos(a);
helper.Out.y = r * sin(a);
helper.Out.z = m_Weight * sin(r);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string slices = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string rotation = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string thickness = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tint sl = (int)(MwcNext01(mwc) * " << slices << " + (real_t)(0.5));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t a = " << rotation << " + M_2PI * (sl + MwcNext01(mwc) * " << thickness << ") / " << slices << ";\n"
<< "\t\treal_t r = xform->m_VariationWeights[" << varIndex << "] * MwcNext01(mwc);\n"
<< "\n"
<< "\t\tvOut.x = r * cos(a);\n"
<< "\t\tvOut.y = r * sin(a);\n"
<< "\t\tvOut.z = xform->m_VariationWeights[" << varIndex << "] * sin(r);\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Random(QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
m_Params[0].Set(10 * rand.Frand01<T>());//Slices.
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params[1].Set(M_2PI * rand.Frand11<T>());//Rotation.
m_Thickness = rand.Frand01<T>();
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Slices, prefix + "pie3D_slices", 6, INTEGER_NONZERO, 1));
m_Params.push_back(ParamWithName<T>(&m_Rotation, prefix + "pie3D_rotation", T(0.5), REAL_CYCLIC, 0, M_2PI));
m_Params.push_back(ParamWithName<T>(&m_Thickness, prefix + "pie3D_thickness", T(0.5), REAL, 0, 1));
}
private:
T m_Slices;
T m_Rotation;
T m_Thickness;
};
/// <summary>
/// popcorn2_3D.
/// </summary>
template <typename T>
class EMBER_API Popcorn23DVariation : public ParametricVariation<T>
{
public:
Popcorn23DVariation(T weight = 1.0) : ParametricVariation<T>("popcorn2_3D", VAR_POPCORN23D, weight, false, false, false, false, true)
{
Init();
}
PARVARCOPY(Popcorn23DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T otherZ, tempPZ = 0;
T tempTZ = helper.In.z == 0 ? m_Vv * m_SinTanC * helper.m_PrecalcAtanyx : helper.In.z;
if (m_VarType == VARTYPE_PRE)
otherZ = helper.In.z;
else
otherZ = outPoint.m_Z;
if (otherZ == 0)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
tempPZ = m_Vv * m_SinTanC * helper.m_PrecalcAtanyx;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
-tan(float) was crashing under VS2013 with large numbers so replace with SafeTan<T>(). -Put about dialog in the center of the screen. A Qt upgrade somehow moved it to the side.
2014-09-08 00:05:27 -04:00
helper.Out.x = m_HalfWeight * (helper.In.x + m_X * sin(SafeTan<T>(m_C * helper.In.y)));
helper.Out.y = m_HalfWeight * (helper.In.y + m_Y * sin(SafeTan<T>(m_C * helper.In.x)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.z = tempPZ + m_Vv * (m_Z * m_SinTanC * tempTZ);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
More C++11
2014-09-10 01:41:26 -04:00
int i = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string z = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string c = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string stc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string hw = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string vv = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\treal_t otherZ, tempPZ = 0;\n"
<< "\t\treal_t tempTZ = vIn.z == 0 ? " << vv << " * " << stc << " * precalcAtanyx : vIn.z;\n";
if (m_VarType == VARTYPE_PRE)
ss << "\t\totherZ = vIn.z;\n";
else
ss << "\t\totherZ = outPoint->m_Z;\n";
ss << "\t\tif (otherZ == 0)\n"
<< "\t\t tempPZ = " << vv << " * " << stc << " * precalcAtanyx;\n"
<< "\n"
<< "\t\tvOut.x = " << hw << " * (vIn.x + " << x << " * sin(tan(" << c << " * vIn.y)));\n"
<< "\t\tvOut.y = " << hw << " * (vIn.y + " << y << " * sin(tan(" << c << " * vIn.x)));\n"
<< "\t\tvOut.z = tempPZ + " << vv << " * (" << z << " * " << stc << " * tempTZ);\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
-tan(float) was crashing under VS2013 with large numbers so replace with SafeTan<T>(). -Put about dialog in the center of the screen. A Qt upgrade somehow moved it to the side.
2014-09-08 00:05:27 -04:00
m_SinTanC = sin(SafeTan<T>(m_C));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_HalfWeight = m_Weight * T(0.5);
if (fabs(m_Weight) <= 1)
m_Vv = fabs(m_Weight) * m_Weight;//Sqr(m_Weight) value retaining sign.
else
m_Vv = m_Weight;
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Random(QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_X = T(0.2) + rand.Frand01<T>();
m_Y = T(0.2) * rand.Frand01<T>();
m_Z = T(0.2) * rand.Frand01<T>();
m_C = 5 * rand.Frand01<T>();
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "popcorn2_3D_x", T(0.1)));
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "popcorn2_3D_y", T(0.1)));
m_Params.push_back(ParamWithName<T>(&m_Z, prefix + "popcorn2_3D_z", T(0.1)));
m_Params.push_back(ParamWithName<T>(&m_C, prefix + "popcorn2_3D_c", 3));
m_Params.push_back(ParamWithName<T>(true, &m_SinTanC, prefix + "popcorn2_3D_sintanc"));
m_Params.push_back(ParamWithName<T>(true, &m_HalfWeight, prefix + "popcorn2_3D_half_weight"));
m_Params.push_back(ParamWithName<T>(true, &m_Vv, prefix + "popcorn2_3D_vv"));
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
T m_X;
T m_Y;
T m_Z;
T m_C;
T m_SinTanC;//Precalcs.
T m_HalfWeight;
T m_Vv;
};
/// <summary>
/// sinusoidal3d.
/// </summary>
template <typename T>
class EMBER_API Sinusoidal3DVariation : public Variation<T>
{
public:
Sinusoidal3DVariation(T weight = 1.0) : Variation<T>("sinusoidal3D", VAR_SINUSOIDAL3D, weight) { }
VARCOPY(Sinusoidal3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
helper.Out.x = m_Weight * sin(helper.In.x);
helper.Out.y = m_Weight * sin(helper.In.y);
helper.Out.z = m_Weight * (atan2(SQR(helper.In.x), SQR(helper.In.y)) * cos(helper.In.z));
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * sin(vIn.x);\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * sin(vIn.y);\n"
<< "\t\tvOut.z = xform->m_VariationWeights[" << varIndex << "] * (atan2(SQR(vIn.x), SQR(vIn.y)) * cos(vIn.z));\n"
<< "\t}\n";
return ss.str();
}
};
/// <summary>
/// scry_3D.
/// </summary>
template <typename T>
class EMBER_API Scry3DVariation : public ParametricVariation<T>
{
public:
Scry3DVariation(T weight = 1.0) : ParametricVariation<T>("scry_3D", VAR_SCRY3D, weight, true, false, false, false, true)
{
Init();
}
PARVARCOPY(Scry3DVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T t = helper.m_PrecalcSumSquares + SQR(helper.In.z);
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
T r = 1 / (std::sqrt(t) * (t + m_InvWeight));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T z = helper.In.z == 0 ? helper.m_PrecalcAtanyx : helper.In.z;
helper.Out.x = helper.In.x * r;
helper.Out.y = helper.In.y * r;
helper.Out.z = z * r;
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
More C++11
2014-09-10 01:41:26 -04:00
int i = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string invWeight = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\treal_t t = precalcSumSquares + SQR(vIn.z);\n"
<< "\t\treal_t r = 1 / (sqrt(t) * (t + " << invWeight << "));\n"
<< "\t\treal_t z = vIn.z == 0 ? precalcAtanyx : vIn.z;\n"
<< "\n"
<< "\t\tvOut.x = vIn.x * r;\n"
<< "\t\tvOut.y = vIn.y * r;\n"
<< "\t\tvOut.z = z * r;\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_InvWeight = 1 / Zeps(m_Weight);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(true, &m_InvWeight, prefix + "scry_3D_inv_weight"));//Precalcs only, no params.
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
T m_InvWeight;//Precalcs only, no params.
};
/// <summary>
/// shredlin.
/// </summary>
template <typename T>
class EMBER_API ShredlinVariation : public ParametricVariation<T>
{
public:
ShredlinVariation(T weight = 1.0) : ParametricVariation<T>("shredlin", VAR_SHRED_LIN, weight)
{
Init();
}
PARVARCOPY(ShredlinVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
const int xpos = helper.In.x < 0;
const int ypos = helper.In.y < 0;
const T xrng = helper.In.x / m_XDistance;
const T yrng = helper.In.y / m_YDistance;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
helper.Out.x = m_Xw * ((xrng - int(xrng)) * m_XWidth + int(xrng) + (T(0.5) - xpos) * m_1mX);
helper.Out.y = m_Yw * ((yrng - int(yrng)) * m_YWidth + int(yrng) + (T(0.5) - ypos) * m_1mY);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.z = m_Weight * helper.In.z;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
//outPoint.m_X = 0;
//outPoint.m_Y = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string xdist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string xwidth = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ydist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ywidth = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string xw = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string yw = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string onemx = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string onemy = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
<< "\t\tconst int xpos = vIn.x < 0;\n"
<< "\t\tconst int ypos = vIn.y < 0;\n"
<< "\t\tconst real_t xrng = vIn.x / " << xdist << ";\n"
<< "\t\tconst real_t yrng = vIn.y / " << ydist << ";\n"
<< "\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tvOut.x = " << xw << " * ((xrng - (int)xrng) * " << xwidth << " + (int)xrng + ((real_t)(0.5) - xpos) * " << onemx << ");\n"
<< "\t\tvOut.y = " << yw << " * ((yrng - (int)yrng) * " << ywidth << " + (int)yrng + ((real_t)(0.5) - ypos) * " << onemy << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\tvOut.z = xform->m_VariationWeights[" << varIndex << "] * vIn.z;\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_Xw = m_Weight * m_XDistance;
m_Yw = m_Weight * m_YDistance;
m_1mX = 1 - m_XWidth;
m_1mY = 1 - m_YWidth;
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_XDistance, prefix + "shredlin_xdistance", 1, REAL_NONZERO));
m_Params.push_back(ParamWithName<T>(&m_XWidth, prefix + "shredlin_xwidth", T(0.5), REAL, -1, 1));
m_Params.push_back(ParamWithName<T>(&m_YDistance, prefix + "shredlin_ydistance", 1, REAL_NONZERO));
m_Params.push_back(ParamWithName<T>(&m_YWidth, prefix + "shredlin_ywidth", T(0.5), REAL, -1, 1));
m_Params.push_back(ParamWithName<T>(true, &m_Xw, prefix + "shredlin_xw"));
m_Params.push_back(ParamWithName<T>(true, &m_Yw, prefix + "shredlin_yw"));
m_Params.push_back(ParamWithName<T>(true, &m_1mX, prefix + "shredlin_1mx"));
m_Params.push_back(ParamWithName<T>(true, &m_1mY, prefix + "shredlin_1my"));
}
private:
T m_XDistance;
T m_XWidth;
T m_YDistance;
T m_YWidth;
T m_Xw;//Precalc.
T m_Yw;
T m_1mX;
T m_1mY;
};
/// <summary>
/// splitbrdr.
/// </summary>
template <typename T>
class EMBER_API SplitBrdrVariation : public ParametricVariation<T>
{
public:
SplitBrdrVariation(T weight = 1.0) : ParametricVariation<T>("SplitBrdr", VAR_SPLIT_BRDR, weight, true)
{
Init();
}
PARVARCOPY(SplitBrdrVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T b = m_Weight / (helper.m_PrecalcSumSquares * T(0.25) + 1);
T roundX = Rint(helper.In.x);
T roundY = Rint(helper.In.y);
T offsetX = helper.In.x - roundX;
T offsetY = helper.In.y - roundY;
helper.Out.x = helper.In.x * b;
helper.Out.y = helper.In.y * b;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
if (rand.Frand01<T>() >= T(0.75))
{
helper.Out.x += m_Weight * (offsetX * T(0.5) + roundX);
helper.Out.y += m_Weight * (offsetY * T(0.5) + roundY);
}
else
{
if (fabs(offsetX) >= fabs(offsetY))
{
if (offsetX >= 0)
{
helper.Out.x += m_Weight * (offsetX * T(0.5) + roundX + m_X);
helper.Out.y += m_Weight * (offsetY * T(0.5) + roundY + m_Y * offsetY / offsetX);
}
else
{
helper.Out.x += m_Weight * (offsetX * T(0.5) + roundX - m_Y);
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
helper.Out.y += m_Weight * (offsetY * T(0.5) + roundY - m_Y * offsetY / offsetX);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
}
else
{
if (offsetY >= 0)
{
helper.Out.y += m_Weight * (offsetY * T(0.5) + roundY + m_Y);
helper.Out.x += m_Weight * (offsetX * T(0.5) + roundX + offsetX / offsetY * m_Y);
}
else
{
helper.Out.y += m_Weight * (offsetY * T(0.5) + roundY - m_Y);
helper.Out.x += m_Weight * (offsetX * T(0.5) + roundX - offsetX / offsetY * m_X);
}
}
}
helper.Out.x += helper.In.x * m_Px;
helper.Out.y += helper.In.y * m_Py;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
string x = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string px = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string py = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\treal_t b = xform->m_VariationWeights[" << varIndex << "] / (precalcSumSquares * (real_t)(0.25) + 1);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t roundX = rint(vIn.x);\n"
<< "\t\treal_t roundY = rint(vIn.y);\n"
<< "\t\treal_t offsetX = vIn.x - roundX;\n"
<< "\t\treal_t offsetY = vIn.y - roundY;\n"
<< "\n"
<< "\t\tvOut.x = vIn.x * b;\n"
<< "\t\tvOut.y = vIn.y * b;\n"
<< "\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tif (MwcNext01(mwc) >= (real_t)(0.75))\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t{\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.x += xform->m_VariationWeights[" << varIndex << "] * (offsetX * (real_t)(0.5) + roundX);\n"
<< "\t\t vOut.y += xform->m_VariationWeights[" << varIndex << "] * (offsetY * (real_t)(0.5) + roundY);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t}\n"
<< "\t\telse\n"
<< "\t\t{\n"
<< "\t\t if (fabs(offsetX) >= fabs(offsetY))\n"
<< "\t\t {\n"
<< "\t\t if (offsetX >= 0)\n"
<< "\t\t {\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.x += xform->m_VariationWeights[" << varIndex << "] * (offsetX * (real_t)(0.5) + roundX + " << x << ");\n"
<< "\t\t vOut.y += xform->m_VariationWeights[" << varIndex << "] * (offsetY * (real_t)(0.5) + roundY + " << y << " * offsetY / offsetX);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.x += xform->m_VariationWeights[" << varIndex << "] * (offsetX * (real_t)(0.5) + roundX - " << y << ");\n"
<< "\t\t vOut.y += xform->m_VariationWeights[" << varIndex << "] * (offsetY * (real_t)(0.5) + roundY - " << y << " * offsetY / offsetX);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
<< "\t\t if (offsetY >= 0)\n"
<< "\t\t {\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.y += xform->m_VariationWeights[" << varIndex << "] * (offsetY * (real_t)(0.5) + roundY + " << y << ");\n"
<< "\t\t vOut.x += xform->m_VariationWeights[" << varIndex << "] * (offsetX * (real_t)(0.5) + roundX + offsetX / offsetY * " << y << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t vOut.y += xform->m_VariationWeights[" << varIndex << "] * (offsetY * (real_t)(0.5) + roundY - " << y << ");\n"
<< "\t\t vOut.x += xform->m_VariationWeights[" << varIndex << "] * (offsetX * (real_t)(0.5) + roundX - offsetX / offsetY * " << x << ");\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t }\n"
<< "\t\t}\n"
<< "\n"
<< "\t\tvOut.x += vIn.x * " << px << ";\n"
<< "\t\tvOut.y += vIn.y * " << py << ";\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_X, prefix + "SplitBrdr_x", T(0.25)));//Original used a prefix of splitb_, which is incompatible with Ember's design.
m_Params.push_back(ParamWithName<T>(&m_Y, prefix + "SplitBrdr_y", T(0.25)));
m_Params.push_back(ParamWithName<T>(&m_Px, prefix + "SplitBrdr_px"));
m_Params.push_back(ParamWithName<T>(&m_Py, prefix + "SplitBrdr_py"));
}
private:
T m_X;
T m_Y;
T m_Px;
T m_Py;
};
/// <summary>
/// wdisc.
/// </summary>
template <typename T>
class EMBER_API WdiscVariation : public Variation<T>
{
public:
WdiscVariation(T weight = 1.0) : Variation<T>("wdisc", VAR_WDISC, weight, true, true, false, false, true) { }
VARCOPY(WdiscVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T a = T(M_PI) / (helper.m_PrecalcSqrtSumSquares + 1);
T r = helper.m_PrecalcAtanyx * T(M_1_PI);
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
if (r > 0)
a = T(M_PI) - a;
helper.Out.x = m_Weight * r * cos(a);
helper.Out.y = m_Weight * r * sin(a);
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\treal_t a = M_PI / (precalcSqrtSumSquares + 1);\n"
<< "\t\treal_t r = precalcAtanyx * M_1_PI;\n"
<< "\n"
<< "\t\tif (r > 0)\n"
<< "\t\t a = M_PI - a;\n"
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * r * cos(a);\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * r * sin(a);\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
};
/// <summary>
/// falloff.
/// </summary>
template <typename T>
class EMBER_API FalloffVariation : public ParametricVariation<T>
{
public:
FalloffVariation(T weight = 1.0) : ParametricVariation<T>("falloff", VAR_FALLOFF, weight, false, false, false, false, true)
{
Init();
}
PARVARCOPY(FalloffVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
const T ax = rand.Frand<T>(T(-0.5), T(0.5));
const T ay = rand.Frand<T>(T(-0.5), T(0.5));
const T az = rand.Frand<T>(T(-0.5), T(0.5));
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
const T r = std::sqrt(Sqr(helper.In.x - m_X0) + Sqr(helper.In.y - m_Y0) + Sqr(helper.In.z - m_Z0));
Remove ReadMe.txt from all project files. Add Curves.h, and CurvesGraphicsView.h/cpp to support bezier color curves. Add Curves member to Ember. Add curves capability to EmberCL. Remove some unused variables in the kernel created in RendererCL::CreateFinalAccumKernelString(). Use glm namespace for vec classes if GLM_VERSION >= 96, else use glm::detail. As a result of using glm namespace, all instances of min and max had to be qualified with std:: Split ComputeCamera into that and ComputeQuality(). Reduce the amount of ComputeCamera() and MakeDmap() calls on each incremental iter that doesn't use temporal samples. Fix clamping bug with DE filter widths. Provide functions to return the kernels from RendererCL to assist with diagnostics and debugging. Prevent extra newline in EmberRender when only rendering a single image. Add the ability to delete an ember at a given index in EmberFile. Allow deleting/focusing ember in library tab with delete and enter keys. Reorder some code in Fractorium.h to match the tabs order. Add and call ClearFinalImages() to clear buffers in controller to fix bug where previous CPU render would be shown for a split second when switching from OpenCL back to CPU. Refactor ember library pointer syncing to a function SyncPointers(). Add the ability to save ember Xmls to an unique automatically generated name after the first time the user has specified a name.
2015-03-21 18:27:37 -04:00
const T rc = ((m_Invert != 0 ? std::max<T>(1 - r, 0) : std::max<T>(r, 0)) - m_MinDist) * m_InternalScatter;//Original called a macro named min, which internally performed max.
const T rs = std::max<T>(rc, 0);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T sigma, phi, rad, sigmas, sigmac, phis, phic;
T scale, denom;
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
switch (int(m_Type))
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
case 0://Linear.
helper.Out.x = m_Weight * (helper.In.x + m_MulX * ax * rs);
helper.Out.y = m_Weight * (helper.In.y + m_MulY * ay * rs);
helper.Out.z = m_Weight * (helper.In.z + m_MulZ * az * rs);
break;
case 1://Radial.
sigma = asin(r == 0 ? 0 : helper.In.z / r) + m_MulZ * az * rs;
phi = helper.m_PrecalcAtanyx + m_MulY * ay * rs;
rad = r + m_MulX * ax * rs;
sigmas = sin(sigma);
sigmac = cos(sigma);
phis = sin(phi);
phic = cos(phi);
helper.Out.x = m_Weight * (rad * sigmac * phic);
helper.Out.y = m_Weight * (rad * sigmac * phis);
helper.Out.z = m_Weight * (rad * sigmas);
break;
case 2://Box.
More C++11
2014-09-10 01:41:26 -04:00
default:
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
scale = Clamp<T>(rs, 0, T(0.9)) + T(0.1);
denom = 1 / scale;
0.4.0.9 Beta 07/27/2014 0.4.0.9 Beta 07/27/2014 --User Changes Properly set tab order on all controls. Calculate and report iters/second in the final render dialog. Immediately draw yellow dot on xform mouse down on previously unselected xform. --Bug Fixes Fix GlynnSim1, GlynnSim2, GlynnSim3 and juliaNab by ensuring the first argument to pow() is >= 0. Ensure OpenCL platform and device combo boxes in the final render dialog expand as needed. --Code Changes Make VariationTreeSpinbox take its parent VariationTreeWidgetItem as a constructor argument. This makes SetupVariationTree() and VariationSpinBoxValueChanged() more efficient. Make Interference2 and ho use fabs().
2014-07-28 01:25:38 -04:00
helper.Out.x = m_Weight * Lerp<T>(helper.In.x, floor(helper.In.x * denom) + scale * ax, m_MulX * rs) + m_MulX * pow(ax, m_BoxPow) * rs * denom;//m_BoxPow should be an integer value held in T,
helper.Out.y = m_Weight * Lerp<T>(helper.In.y, floor(helper.In.y * denom) + scale * ay, m_MulY * rs) + m_MulY * pow(ay, m_BoxPow) * rs * denom;//so fabs() shouldn't be necessary.
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.z = m_Weight * Lerp<T>(helper.In.z, floor(helper.In.z * denom) + scale * az, m_MulZ * rs) + m_MulZ * pow(az, m_BoxPow) * rs * denom;
break;
}
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string scatter = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string minDist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulX = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulY = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulZ = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string x0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string z0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string invert = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string type = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string boxPow = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string internalScatter = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
<< "\t\tconst real_t ax = MwcNext0505(mwc);\n"
<< "\t\tconst real_t ay = MwcNext0505(mwc);\n"
<< "\t\tconst real_t az = MwcNext0505(mwc);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\tconst real_t r = sqrt(Sqr(vIn.x - " << x0 << ") + Sqr(vIn.y - " << y0 << ") + Sqr(vIn.z - " << z0 << "));\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tconst real_t rc = ((" << invert << " != 0 ? max(1 - r, (real_t)(0.0)) : max(r, (real_t)(0.0))) - " << minDist << ") * " << internalScatter << ";\n"
<< "\t\tconst real_t rs = max(rc, (real_t)(0.0));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\treal_t sigma, phi, rad, sigmas, sigmac, phis, phic;\n"
<< "\t\treal_t scale, denom;\n"
<< "\n"
<< "\t\tswitch ((int)" << type << ")\n"
<< "\t\t{\n"
<< "\t\t case 0:\n"
<< "\t\t vOut.x = xform->m_VariationWeights[" << varIndex << "] * (vIn.x + " << mulX << " * ax * rs);\n"
<< "\t\t vOut.y = xform->m_VariationWeights[" << varIndex << "] * (vIn.y + " << mulY << " * ay * rs);\n"
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * (vIn.z + " << mulZ << " * az * rs);\n"
<< "\t\t break;\n"
<< "\t\t case 1:\n"
<< "\t\t sigma = asin(r == 0 ? 0 : vIn.z / r) + " << mulZ << " * az * rs;\n"
<< "\t\t phi = precalcAtanyx + " << mulY << " * ay * rs;\n"
<< "\t\t rad = r + " << mulX << " * ax * rs;\n"
<< "\n"
<< "\t\t sigmas = sin(sigma);\n"
<< "\t\t sigmac = cos(sigma);\n"
<< "\t\t phis = sin(phi);\n"
<< "\t\t phic = cos(phi);\n"
<< "\n"
<< "\t\t vOut.x = xform->m_VariationWeights[" << varIndex << "] * (rad * sigmac * phic);\n"
<< "\t\t vOut.y = xform->m_VariationWeights[" << varIndex << "] * (rad * sigmac * phis);\n"
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * (rad * sigmas);\n"
<< "\t\t break;\n"
<< "\t\t case 2:\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t scale = Clamp(rs, 0, (real_t)(0.9)) + (real_t)(0.1);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t denom = 1 / scale;\n"
<< "\t\t vOut.x = xform->m_VariationWeights[" << varIndex << "] * Lerp(vIn.x, floor(vIn.x * denom) + scale * ax, " << mulX << " * rs) + " << mulX << " * pow(ax, " << boxPow << ") * rs * denom;\n"
<< "\t\t vOut.y = xform->m_VariationWeights[" << varIndex << "] * Lerp(vIn.y, floor(vIn.y * denom) + scale * ay, " << mulY << " * rs) + " << mulY << " * pow(ay, " << boxPow << ") * rs * denom;\n"
<< "\t\t vOut.z = xform->m_VariationWeights[" << varIndex << "] * Lerp(vIn.z, floor(vIn.z * denom) + scale * az, " << mulZ << " * rs) + " << mulZ << " * pow(az, " << boxPow << ") * rs * denom;\n"
<< "\t\t break;\n"
<< "\t\t}\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_InternalScatter = T(0.04) * m_Scatter;
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_Scatter, prefix + "falloff_scatter", 1, REAL, EPS, TMAX));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_MinDist, prefix + "falloff_mindist", T(0.5), REAL, 0, TMAX));
m_Params.push_back(ParamWithName<T>(&m_MulX, prefix + "falloff_mul_x", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulY, prefix + "falloff_mul_y", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulZ, prefix + "falloff_mul_z", 0, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_X0, prefix + "falloff_x0"));
m_Params.push_back(ParamWithName<T>(&m_Y0, prefix + "falloff_y0"));
m_Params.push_back(ParamWithName<T>(&m_Z0, prefix + "falloff_z0"));
m_Params.push_back(ParamWithName<T>(&m_Invert, prefix + "falloff_invert", 0, INTEGER, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_Type, prefix + "falloff_type", 0, INTEGER, 0, 2));
m_Params.push_back(ParamWithName<T>(&m_BoxPow, prefix + "falloff_boxpow", 2, INTEGER, 2, 32));//Original defaulted this to 0 which directly contradicts the specified range of 2-32.
m_Params.push_back(ParamWithName<T>(true, &m_InternalScatter, prefix + "falloff_internal_scatter"));
}
private:
T m_Scatter;
T m_MinDist;
T m_MulX;
T m_MulY;
T m_MulZ;
T m_X0;
T m_Y0;
T m_Z0;
T m_Invert;
T m_Type;
T m_BoxPow;
T m_InternalScatter;
};
/// <summary>
/// falloff2.
/// </summary>
template <typename T>
class EMBER_API Falloff2Variation : public ParametricVariation<T>
{
public:
Falloff2Variation(T weight = 1.0) : ParametricVariation<T>("falloff2", VAR_FALLOFF2, weight, true, false, false, false, true)
{
Init();
}
PARVARCOPY(Falloff2Variation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
const v4T random(rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)));
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
const T distA = std::sqrt(Sqr(helper.In.x - m_X0) + Sqr(helper.In.y - m_Y0) + Sqr(helper.In.z - m_Z0));
Remove ReadMe.txt from all project files. Add Curves.h, and CurvesGraphicsView.h/cpp to support bezier color curves. Add Curves member to Ember. Add curves capability to EmberCL. Remove some unused variables in the kernel created in RendererCL::CreateFinalAccumKernelString(). Use glm namespace for vec classes if GLM_VERSION >= 96, else use glm::detail. As a result of using glm namespace, all instances of min and max had to be qualified with std:: Split ComputeCamera into that and ComputeQuality(). Reduce the amount of ComputeCamera() and MakeDmap() calls on each incremental iter that doesn't use temporal samples. Fix clamping bug with DE filter widths. Provide functions to return the kernels from RendererCL to assist with diagnostics and debugging. Prevent extra newline in EmberRender when only rendering a single image. Add the ability to delete an ember at a given index in EmberFile. Allow deleting/focusing ember in library tab with delete and enter keys. Reorder some code in Fractorium.h to match the tabs order. Add and call ClearFinalImages() to clear buffers in controller to fix bug where previous CPU render would be shown for a split second when switching from OpenCL back to CPU. Refactor ember library pointer syncing to a function SyncPointers(). Add the ability to save ember Xmls to an unique automatically generated name after the first time the user has specified a name.
2015-03-21 18:27:37 -04:00
const T distB = m_Invert != 0 ? std::max<T>(1 - distA, 0) : std::max<T>(distA, 0);//Original called a macro named min, which internally performed max.
const T dist = std::max<T>((distB - m_MinDist) * m_RMax, 0);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
switch (int(m_Type))
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
case 0://Linear.
{
helper.Out.x = helper.In.x + m_MulX * random.x * dist;
helper.Out.y = helper.In.y + m_MulY * random.y * dist;
helper.Out.z = helper.In.z + m_MulZ * random.z * dist;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + m_MulC * random.w * dist, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
case 1://Radial.
if (helper.In.x == 0 && helper.In.y == 0 && helper.In.z == 0)
{
helper.Out.x = helper.In.x;
helper.Out.y = helper.In.y;
helper.Out.z = helper.In.z;
}
else
{
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
const T rIn = std::sqrt(helper.m_PrecalcSumSquares + SQR(helper.In.z));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
const T sigma = asin(helper.In.z / rIn) + m_MulZ * random.z * dist;
const T phi = helper.m_PrecalcAtanyx + m_MulY * random.y * dist;
const T r = rIn + m_MulX * random.x * dist;
const T sigmas = sin(sigma);
const T sigmac = cos(sigma);
const T phis = sin(phi);
const T phic = cos(phi);
helper.Out.x = r * sigmac * phic;
helper.Out.y = r * sigmac * phis;
helper.Out.z = r * sigmas;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + m_MulC * random.w * dist, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
case 2://Gaussian.
More C++11
2014-09-10 01:41:26 -04:00
default:
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
const T sigma = dist * random.y * M_2PI;
const T phi = dist * random.z * T(M_PI);
const T rad = dist * random.x;
const T sigmas = sin(sigma);
const T sigmac = cos(sigma);
const T phis = sin(phi);
const T phic = cos(phi);
helper.Out.x = helper.In.x + m_MulX * rad * sigmac * phic;
helper.Out.y = helper.In.y + m_MulY * rad * sigmac * phis;
helper.Out.z = helper.In.z + m_MulZ * rad * sigmas;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + m_MulC * random.w * dist, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
}
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
More C++11
2014-09-10 01:41:26 -04:00
int i = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string scatter = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string minDist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulX = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulY = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulZ = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulC = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string x0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string y0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string z0 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string invert = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string type = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string rMax = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
<< "\t\tconst real_t randx = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randy = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randz = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randc = MwcNext0505(mwc);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\tconst real_t distA = sqrt(Sqr(vIn.x - " << x0 << ") + Sqr(vIn.y - " << y0 << ") + Sqr(vIn.z - " << z0 << "));\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tconst real_t distB = " << invert << " != 0 ? max(1 - distA, (real_t)(0.0)) : max(distA, (real_t)(0.0));\n"
<< "\t\tconst real_t dist = max((distB - " << minDist << ") * " << rMax<< ", (real_t)(0.0));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tswitch ((int)" << type << ")\n"
<< "\t\t{\n"
<< "\t\t case 0:\n"
<< "\t\t vOut.x = vIn.x + " << mulX << " * randx * dist;\n"
<< "\t\t vOut.y = vIn.y + " << mulY << " * randy * dist;\n"
<< "\t\t vOut.z = vIn.z + " << mulZ << " * randz * dist;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + " << mulC << " * randc * dist, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t break;\n"
<< "\t\t case 1:\n"
<< "\t\t if (vIn.x == 0 && vIn.y == 0 && vIn.z == 0)\n"
<< "\t\t {\n"
<< "\t\t vOut.x = vIn.x;\n"
<< "\t\t vOut.y = vIn.y;\n"
<< "\t\t vOut.z = vIn.z;\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
<< "\t\t real_t rIn = sqrt(precalcSumSquares + SQR(vIn.z));\n"
<< "\t\t real_t sigma = asin(vIn.z / rIn) + " << mulZ << " * randz * dist;\n"
<< "\t\t real_t phi = precalcAtanyx + " << mulY << " * randy * dist;\n"
<< "\t\t real_t r = rIn + " << mulX << " * randx * dist;\n"
<< "\t\t real_t sigmas = sin(sigma);\n"
<< "\t\t real_t sigmac = cos(sigma);\n"
<< "\t\t real_t phis = sin(phi);\n"
<< "\t\t real_t phic = cos(phi);\n"
<< "\n"
<< "\t\t vOut.x = r * sigmac * phic;\n"
<< "\t\t vOut.y = r * sigmac * phis;\n"
<< "\t\t vOut.z = r * sigmas;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + " << mulC << " * randc * dist, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t break;\n"
<< "\t\t case 2:\n"
Fix falloff2 Fix falloff2
2014-07-08 22:33:31 -04:00
<< "\t\t {\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t real_t sigma = dist * randy * M_2PI;\n"
<< "\t\t real_t phi = dist * randz * M_PI;\n"
<< "\t\t real_t rad = dist * randx;\n"
<< "\t\t real_t sigmas = sin(sigma);\n"
<< "\t\t real_t sigmac = cos(sigma);\n"
<< "\t\t real_t phis = sin(phi);\n"
<< "\t\t real_t phic = cos(phi);\n"
<< "\n"
<< "\t\t vOut.x = vIn.x + " << mulX << " * rad * sigmac * phic;\n"
<< "\t\t vOut.y = vIn.y + " << mulY << " * rad * sigmac * phis;\n"
<< "\t\t vOut.z = vIn.z + " << mulZ << " * rad * sigmas;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + " << mulC << " * randc * dist, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t break;\n"
Fix falloff2 Fix falloff2
2014-07-08 22:33:31 -04:00
<< "\t\t }\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t}\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_RMax = T(0.04) * m_Scatter;
}
protected:
void Init()
{
string prefix = Prefix();
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.clear();
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_Scatter, prefix + "falloff2_scatter", 1, REAL, EPS, TMAX));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_MinDist, prefix + "falloff2_mindist", T(0.5), REAL, 0, TMAX));
m_Params.push_back(ParamWithName<T>(&m_MulX, prefix + "falloff2_mul_x", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulY, prefix + "falloff2_mul_y", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulZ, prefix + "falloff2_mul_z", 0, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulC, prefix + "falloff2_mul_c", 0, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_X0, prefix + "falloff2_x0"));
m_Params.push_back(ParamWithName<T>(&m_Y0, prefix + "falloff2_y0"));
m_Params.push_back(ParamWithName<T>(&m_Z0, prefix + "falloff2_z0"));
m_Params.push_back(ParamWithName<T>(&m_Invert, prefix + "falloff2_invert", 0, INTEGER, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_Type, prefix + "falloff2_type", 0, INTEGER, 0, 2));
m_Params.push_back(ParamWithName<T>(true, &m_RMax, prefix + "falloff2_rmax"));
}
private:
T m_Scatter;
T m_MinDist;
T m_MulX;
T m_MulY;
T m_MulZ;
T m_MulC;
T m_X0;
T m_Y0;
T m_Z0;
T m_Invert;
T m_Type;
T m_RMax;
};
/// <summary>
/// falloff3.
/// </summary>
template <typename T>
class EMBER_API Falloff3Variation : public ParametricVariation<T>
{
public:
Falloff3Variation(T weight = 1.0) : ParametricVariation<T>("falloff3", VAR_FALLOFF3, weight, true, false, false, false, true)
{
Init();
}
PARVARCOPY(Falloff3Variation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
const v4T random(rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)), rand.Frand<T>(T(-0.5), T(0.5)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T radius;
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
switch (int(m_BlurShape))
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
case 0://Circle.
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
radius = std::sqrt(Sqr(helper.In.x - m_CenterX) + Sqr(helper.In.y - m_CenterY) + Sqr(helper.In.z - m_CenterZ));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
break;
case 1://Square.
More C++11
2014-09-10 01:41:26 -04:00
default:
Remove ReadMe.txt from all project files. Add Curves.h, and CurvesGraphicsView.h/cpp to support bezier color curves. Add Curves member to Ember. Add curves capability to EmberCL. Remove some unused variables in the kernel created in RendererCL::CreateFinalAccumKernelString(). Use glm namespace for vec classes if GLM_VERSION >= 96, else use glm::detail. As a result of using glm namespace, all instances of min and max had to be qualified with std:: Split ComputeCamera into that and ComputeQuality(). Reduce the amount of ComputeCamera() and MakeDmap() calls on each incremental iter that doesn't use temporal samples. Fix clamping bug with DE filter widths. Provide functions to return the kernels from RendererCL to assist with diagnostics and debugging. Prevent extra newline in EmberRender when only rendering a single image. Add the ability to delete an ember at a given index in EmberFile. Allow deleting/focusing ember in library tab with delete and enter keys. Reorder some code in Fractorium.h to match the tabs order. Add and call ClearFinalImages() to clear buffers in controller to fix bug where previous CPU render would be shown for a split second when switching from OpenCL back to CPU. Refactor ember library pointer syncing to a function SyncPointers(). Add the ability to save ember Xmls to an unique automatically generated name after the first time the user has specified a name.
2015-03-21 18:27:37 -04:00
radius = std::max(fabs(helper.In.x - m_CenterX), std::max(fabs(helper.In.y - m_CenterY), (fabs(helper.In.z - m_CenterZ))));//Original called a macro named min, which internally performed max.
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
break;
}
Remove ReadMe.txt from all project files. Add Curves.h, and CurvesGraphicsView.h/cpp to support bezier color curves. Add Curves member to Ember. Add curves capability to EmberCL. Remove some unused variables in the kernel created in RendererCL::CreateFinalAccumKernelString(). Use glm namespace for vec classes if GLM_VERSION >= 96, else use glm::detail. As a result of using glm namespace, all instances of min and max had to be qualified with std:: Split ComputeCamera into that and ComputeQuality(). Reduce the amount of ComputeCamera() and MakeDmap() calls on each incremental iter that doesn't use temporal samples. Fix clamping bug with DE filter widths. Provide functions to return the kernels from RendererCL to assist with diagnostics and debugging. Prevent extra newline in EmberRender when only rendering a single image. Add the ability to delete an ember at a given index in EmberFile. Allow deleting/focusing ember in library tab with delete and enter keys. Reorder some code in Fractorium.h to match the tabs order. Add and call ClearFinalImages() to clear buffers in controller to fix bug where previous CPU render would be shown for a split second when switching from OpenCL back to CPU. Refactor ember library pointer syncing to a function SyncPointers(). Add the ability to save ember Xmls to an unique automatically generated name after the first time the user has specified a name.
2015-03-21 18:27:37 -04:00
const T dist = std::max<T>(((m_InvertDistance != 0 ? std::max<T>(1 - radius, 0) : std::max<T>(radius, 0)) - m_MinDistance) * m_RMax, 0);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
switch (int(m_BlurType))
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
case 0://Gaussian.
{
const T sigma = dist * random.y * M_2PI;
const T phi = dist * random.z * T(M_PI);
const T rad = dist * random.x;
const T sigmas = sin(sigma);
const T sigmac = cos(sigma);
const T phis = sin(phi);
const T phic = cos(phi);
helper.Out.x = helper.In.x + m_MulX * rad * sigmac * phic;
helper.Out.y = helper.In.y + m_MulY * rad * sigmac * phis;
helper.Out.z = helper.In.z + m_MulZ * rad * sigmas;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + m_MulC * random.w * dist, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
case 1://Radial.
if (helper.In.x == 0 && helper.In.y == 0 && helper.In.z == 0)
{
helper.Out.x = helper.In.x;
helper.Out.y = helper.In.y;
helper.Out.z = helper.In.z;
}
else
{
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
const T rIn = std::sqrt(helper.m_PrecalcSumSquares + SQR(helper.In.z));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
const T sigma = asin(helper.In.z / rIn) + m_MulZ * random.z * dist;
const T phi = helper.m_PrecalcAtanyx + m_MulY * random.y * dist;
const T r = rIn + m_MulX * random.x * dist;
const T sigmas = sin(sigma);
const T sigmac = cos(sigma);
const T phis = sin(phi);
const T phic = cos(phi);
helper.Out.x = r * sigmac * phic;
helper.Out.y = r * sigmac * phis;
helper.Out.z = r * sigmas;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + m_MulC * random.w * dist, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
case 2://Log.
More C++11
2014-09-10 01:41:26 -04:00
default:
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
const T coeff = m_RMax <= EPS ? dist : dist + m_Alpha * (LogMap(dist) - dist);
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
helper.Out.x = helper.In.x + LogMap(m_MulX) * LogScale(random.x) * coeff,
helper.Out.y = helper.In.y + LogMap(m_MulY) * LogScale(random.y) * coeff,
helper.Out.z = helper.In.z + LogMap(m_MulZ) * LogScale(random.z) * coeff,
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
outPoint.m_ColorX = fabs(fmod(outPoint.m_ColorX + LogMap(m_MulC) * LogScale(random.w) * coeff, T(1)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
break;
}
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
More C++11
2014-09-10 01:41:26 -04:00
int i = 0;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string blurType = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string blurShape = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string blurStrength = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string minDist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string invertDist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulX = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulY = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulZ = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string mulC = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string centerX = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string centerY = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string centerZ = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string alpha = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string rMax = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
ss << "\t{\n"
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
<< "\t\tconst real_t randx = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randy = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randz = MwcNext0505(mwc);\n"
<< "\t\tconst real_t randc = MwcNext0505(mwc);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\treal_t radius;\n"
<< "\n"
<< "\t\tswitch ((int)" << blurShape << ")\n"
<< "\t\t{\n"
<< "\t\t case 0:\n"
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
<< "\t\t radius = sqrt(Sqr(vIn.x - " << centerX << ") + Sqr(vIn.y - " << centerY << ") + Sqr(vIn.z - " << centerZ << "));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t break;\n"
<< "\t\t case 1:\n"
0.4.0.8 Beta 07/26/2014 0.4.0.8 Beta 07/26/2014 --Bug Fixes Fix falloff, falloff2, falloff3.
2014-07-26 20:26:15 -04:00
<< "\t\t radius = max(fabs(vIn.x - " << centerX << "), max(fabs(vIn.y - " << centerY << "), (fabs(vIn.z - " << centerZ << "))));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t break;\n"
<< "\t\t}\n"
<< "\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\tconst real_t dist = max(((" << invertDist << " != 0 ? max(1 - radius, (real_t)(0.0)) : max(radius, (real_t)(0.0))) - " << minDist << ") * " << rMax << ", (real_t)(0.0));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\tswitch ((int)" << blurType << ")\n"
<< "\t\t{\n"
<< "\t\tcase 0:\n"
<< "\t\t {\n"
<< "\t\t real_t sigma = dist * randy * M_2PI;\n"
<< "\t\t real_t phi = dist * randz * M_PI;\n"
<< "\t\t real_t rad = dist * randx;\n"
<< "\t\t real_t sigmas = sin(sigma);\n"
<< "\t\t real_t sigmac = cos(sigma);\n"
<< "\t\t real_t phis = sin(phi);\n"
<< "\t\t real_t phic = cos(phi);\n"
<< "\n"
<< "\t\t vOut.x = vIn.x + " << mulX << " * rad * sigmac * phic;\n"
<< "\t\t vOut.y = vIn.y + " << mulY << " * rad * sigmac * phis;\n"
<< "\t\t vOut.z = vIn.z + " << mulZ << " * rad * sigmas;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + " << mulC << " * randc * dist, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t break;\n"
<< "\t\tcase 1:\n"
<< "\t\t if (vIn.x == 0 && vIn.y == 0 && vIn.z == 0)\n"
<< "\t\t {\n"
<< "\t\t vOut.x = vIn.x;\n"
<< "\t\t vOut.y = vIn.y;\n"
<< "\t\t vOut.z = vIn.z;\n"
<< "\t\t }\n"
<< "\t\t else\n"
<< "\t\t {\n"
Explicitly namespace calls to sqrt and isnan osx/linux have alternative top-level implementations of these methods, so the calls are ambiguous given the - so these should be used explicitly
2015-04-13 07:32:58 -04:00
<< "\t\t real_t rIn = sqrt(precalcSumSquares + SQR(vIn.z));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t real_t sigma = asin(vIn.z / rIn) + " << mulZ << " * randz * dist;\n"
<< "\t\t real_t phi = precalcAtanyx + " << mulY << " * randy * dist;\n"
<< "\t\t real_t r = rIn + " << mulX << " * randx * dist;\n"
<< "\t\t real_t sigmas = sin(sigma);\n"
<< "\t\t real_t sigmac = cos(sigma);\n"
<< "\t\t real_t phis = sin(phi);\n"
<< "\t\t real_t phic = cos(phi);\n"
<< "\n"
<< "\t\t vOut.x = r * sigmac * phic;\n"
<< "\t\t vOut.y = r * sigmac * phis;\n"
<< "\t\t vOut.z = r * sigmas;\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + " << mulC << " * randc * dist, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t break;\n"
<< "\t\tcase 2:\n"
<< "\t\t {\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\t real_t coeff = " << rMax << " <= EPS ? dist : dist + " << alpha << " * (LogMap(dist) - dist);\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\n"
<< "\t\t vOut.x = vIn.x + LogMap(" << mulX << ") * LogScale(randx) * coeff,\n"
<< "\t\t vOut.y = vIn.y + LogMap(" << mulY << ") * LogScale(randy) * coeff,\n"
<< "\t\t vOut.z = vIn.z + LogMap(" << mulZ << ") * LogScale(randz) * coeff,\n"
Ditch T(f) define in opencl programs in preference to already-defined real_t. Also removed some erroneously-added double-'T(' statements
2015-02-26 16:18:55 -05:00
<< "\t\t outPoint->m_ColorX = fabs(fmod(outPoint->m_ColorX + LogMap(" << mulC << ") * LogScale(randc) * coeff, (real_t)(1.0)));\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t\t }\n"
<< "\t\t break;\n"
<< "\t\t}\n"
<< "\t}\n";
return ss.str();
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
m_RMax = T(0.04) * m_BlurStrength;
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_BlurType, prefix + "falloff3_blur_type", 0, INTEGER, 0, 3));
m_Params.push_back(ParamWithName<T>(&m_BlurShape, prefix + "falloff3_blur_shape", 0, INTEGER, 0, 1));
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
m_Params.push_back(ParamWithName<T>(&m_BlurStrength, prefix + "falloff3_blur_strength", 1, REAL, EPS, TMAX));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Params.push_back(ParamWithName<T>(&m_MinDistance, prefix + "falloff3_min_distance", T(0.5), REAL, 0, TMAX));
m_Params.push_back(ParamWithName<T>(&m_InvertDistance, prefix + "falloff3_invert_distance", 0, INTEGER, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulX, prefix + "falloff3_mul_x", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulY, prefix + "falloff3_mul_y", 1, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulZ, prefix + "falloff3_mul_z", 0, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_MulC, prefix + "falloff3_mul_c", 0, REAL, 0, 1));
m_Params.push_back(ParamWithName<T>(&m_CenterX, prefix + "falloff3_center_x"));
m_Params.push_back(ParamWithName<T>(&m_CenterY, prefix + "falloff3_center_y"));
m_Params.push_back(ParamWithName<T>(&m_CenterZ, prefix + "falloff3_center_z"));
m_Params.push_back(ParamWithName<T>(&m_Alpha, prefix + "falloff3_alpha"));
m_Params.push_back(ParamWithName<T>(true, &m_RMax, prefix + "falloff3_rmax"));
}
private:
T m_BlurType;
T m_BlurShape;
T m_BlurStrength;
T m_MinDistance;
T m_InvertDistance;
T m_MulX;
T m_MulY;
T m_MulZ;
T m_MulC;
T m_CenterX;
T m_CenterY;
T m_CenterZ;
T m_Alpha;
T m_RMax;
};
/// <summary>
/// xtrb.
/// </summary>
template <typename T>
class EMBER_API XtrbVariation : public ParametricVariation<T>
{
public:
XtrbVariation(T weight = 1.0) : ParametricVariation<T>("xtrb", VAR_XTRB, weight)
{
Init();
}
PARVARCOPY(XtrbVariation)
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Func(IteratorHelper<T>& helper, Point<T>& outPoint, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand) override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
--User changes -Add support for multiple GPU devices. --These options are present in the command line and in Fractorium. -Change scheme of specifying devices from platform,device to just total device index. --Single number on the command line. --Change from combo boxes for device selection to a table of all devices in Fractorium. -Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Bug fixes -EmberAnimate, EmberRender, FractoriumSettings, FinalRenderDialog: Fix wrong order of arguments to Clamp() when assigning thread priority. -VariationsDC.h: Fix NVidia OpenCL compilation error in DCTriangleVariation. -FractoriumXformsColor.cpp: Checking for null pixmap pointer is not enough, must also check if the underlying buffer is null via call to QPixmap::isNull(). --Code changes -Ember.h: Add case for FLAME_MOTION_NONE and default in ApplyFlameMotion(). -EmberMotion.h: Call base constructor. -EmberPch.h: #pragma once only on Windows. -EmberToXml.h: --Handle different types of exceptions. --Add default cases to ToString(). -Isaac.h: Remove unused variable in constructor. -Point.h: Call base constructor in Color(). -Renderer.h/cpp: --Add bool to Alloc() to only allocate memory for the histogram. Needed for multi-GPU. --Make CoordMap() return a const ref, not a pointer. -SheepTools.h: --Use 64-bit types like the rest of the code already does. --Fix some comment misspellings. -Timing.h: Make BeginTime(), EndTime(), ElapsedTime() and Format() be const functions. -Utils.h: --Add new functions Equal() and Split(). --Handle more exception types in ReadFile(). --Get rid of most legacy blending of C and C++ argument parsing. -XmlToEmber.h: --Get rid of most legacy blending of C and C++ code from flam3. --Remove some unused variables. -EmberAnimate: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --If a render fails, exit since there is no point in continuing an animation with a missing frame. --Pass variables to threaded save better, which most likely fixes a very subtle bug that existed before. --Remove some unused variables. -EmberGenome, EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. -EmberRender: --Support multi-GPU processing that alternates full frames between devices. --Use OpenCLInfo instead of OpenCLWrapper for --openclinfo option. --Remove bucketT template parameter, and hard code float in its place. --Only print values when not rendering with OpenCL, since they're always 0 in that case. -EmberCLPch.h: --#pragma once only on Windows. --#include <atomic>. -IterOpenCLKernelCreator.h: Add new kernel for summing two histograms. This is needed for multi-GPU. -OpenCLWrapper.h: --Move all OpenCL info related code into its own class OpenCLInfo. --Add members to cache the values of global memory size and max allocation size. -RendererCL.h/cpp: --Redesign to accomodate multi-GPU. --Constructor now takes a vector of devices. --Remove DumpErrorReport() function, it's handled in the base. --ClearBuffer(), ReadPoints(), WritePoints(), ReadHist() and WriteHist() now optionally take a device index as a parameter. --MakeDmap() override and m_DmapCL member removed because it no longer applies since the histogram is always float since the last commit. --Add new function SumDeviceHist() to sum histograms from two devices by first copying to a temporary on the host, then a temporary on the device, then summing. --m_Calls member removed, as it's now per-device. --OpenCLWrapper removed. --m_Seeds member is now a vector of vector of seeds, to accomodate a separate and different array of seeds for each device. --Added member m_Devices, a vector of unique_ptr of RendererCLDevice. -EmberCommon.h --Added Devices() function to convert from a vector of device indices to a vector of platform,device indices. --Changed CreateRenderer() to accept a vector of devices to create a single RendererCL which will split work across multiple devices. --Added CreateRenderers() function to accept a vector of devices to create multiple RendererCL, each which will render on a single device. --Add more comments to some existing functions. -EmberCommonPch.h: #pragma once only on Windows. -EmberOptions.h: --Remove --platform option, it's just sequential device number now with the --device option. --Make --out be OPT_USE_RENDER instead of OPT_RENDER_ANIM since it's an error condition when animating. It makes no sense to write all frames to a single image. --Add Devices() function to parse comma separated --device option string and return a vector of device indices. --Make int and uint types be 64-bit, so intmax_t and size_t. --Make better use of macros. -JpegUtils.h: Make string parameters to WriteJpeg() and WritePng() be const ref. -All project files: Turn off buffer security check option in Visual Studio (/Gs-) -deployment.pri: Remove the line OTHER_FILES +=, it's pointless and was causing problems. -Ember.pro, EmberCL.pro: Add CONFIG += plugin, otherwise it wouldn't link. -EmberCL.pro: Add new files for multi-GPU support. -build_all.sh: use -j4 and QMAKE=${QMAKE:/usr/bin/qmake} -shared_settings.pri: -Add version string. -Remove old DESTDIR definitions. -Add the following lines or else nothing would build: CONFIG(release, debug|release) { CONFIG += warn_off DESTDIR = ../../../Bin/release } CONFIG(debug, debug|release) { DESTDIR = ../../../Bin/debug } QMAKE_POST_LINK += $$quote(cp --update ../../../Data/flam3-palettes.xml $${DESTDIR}$$escape_expand(\n\t)) LIBS += -L/usr/lib -lpthread -AboutDialog.ui: Another futile attempt to make it look correct on Linux. -FinalRenderDialog.h/cpp: --Add support for multi-GPU. --Change from combo boxes for device selection to a table of all devices. --Ensure device selection makes sense. --Remove "FinalRender" prefix of various function names, it's implied given the context. -FinalRenderEmberController.h/cpp: --Add support for multi-GPU. --Change m_FinishedImageCount to be atomic. --Move CancelRender() from the base to FinalRenderEmberController<T>. --Refactor RenderComplete() to omit any progress related functionality or image saving since it can be potentially ran in a thread. --Consolidate setting various renderer fields into SyncGuiToRenderer(). -Fractorium.cpp: Allow for resizing of the options dialog to show the entire device table. -FractoriumCommon.h: Add various functions to handle a table showing the available OpenCL devices on the system. -FractoriumEmberController.h/cpp: Remove m_FinalImageIndex, it's no longer needed. -FractoriumRender.cpp: Scale the interactive sub batch count and quality by the number of devices used. -FractoriumSettings.h/cpp: --Temporal samples defaults to 100 instead of 1000 which was needless overkill. --Add multi-GPU support, remove old device,platform pair. -FractoriumToolbar.cpp: Disable OpenCL toolbar button if there are no devices present on the system. -FractoriumOptionsDialog.h/cpp: --Add support for multi-GPU. --Consolidate more assignments in DataToGui(). --Enable/disable CPU/OpenCL items in response to OpenCL checkbox event. -Misc: Convert almost everything to size_t for unsigned, intmax_t for signed.
2015-09-12 21:33:45 -04:00
intmax_t m, n;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T alpha, beta, offsetAl, offsetBe, offsetGa, x, y;
//Transfer to trilinear coordinates, normalized to real distances from triangle sides.
DirectTrilinear(helper.In.x, helper.In.y, alpha, beta);
m = Floor<T>(alpha / m_S2a);
offsetAl = alpha - m * m_S2a;
n = Floor<T>(beta / m_S2b);
offsetBe = beta - n * m_S2b;
offsetGa = m_S2c - m_Ac * offsetAl - m_Bc * offsetBe;
if (offsetGa > 0)
{
Hex(offsetAl, offsetBe, offsetGa, alpha, beta, rand);
}
else
{
offsetAl = m_S2a - offsetAl;
offsetBe = m_S2b - offsetBe;
offsetGa = -offsetGa;
Hex(offsetAl, offsetBe, offsetGa, alpha, beta, rand);
alpha = m_S2a - alpha;
beta = m_S2b - beta;
}
alpha += m * m_S2a;
beta += n * m_S2b;
InverseTrilinear(alpha, beta, x, y, rand);
helper.Out.x = m_Weight * x;
helper.Out.y = m_Weight * y;
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
helper.Out.z = (m_VarType == VARTYPE_REG) ? 0 : helper.In.z;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLString() const override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
ostringstream ss, ss2;
0.4.1.3 Beta 10/14/2014 --User Changes Size is no longer fixed to the window size. Size scaling is done differently in the final render dialog. This fixes several bugs. Remove Xml saving size from settings and options dialog, it no longer applies. Final render can be broken into strips. Set default save path to the desktop if none is found in the settings file. Set default output size to 1920x1080 if none is found in the settings file. --Bug Fixes Better memory size reporting in final render dialog. --Code Changes Migrate to C++11, Qt 5.3.1, and Visual Studio 2013. Change most instances of unsigned int to size_t, and int to intmax_t. Add m_OrigPixPerUnit and m_ScaleType to Ember for scaling purposes. Replace some sprintf_s() calls in XmlToEmber with ostringstream. Move more non-templated members into RendererBase. Add CopyVec() overload that takes a per element function pointer. Add vector Memset(). Replace '&' with '+' instead of "&amp;" in XmlToEmber for much faster parsing. Break strips rendering out into EmberCommon and call from EmberRender and Fractorium. Make AddAndWriteBuffer() just call WriteBuffer(). Make AddAndWriteImage() delete the existing image first before replacing it. Add SetOutputTexture() to RendererCL to support making new textures in response to resize events. Remove multiple return statements in RendererCL, and replace with a bool that tracks results. Add ToDouble(), MakeEnd(), ToString() and Exists() wrappers in Fractorium. Add Size() wrapper in EmberFile. Make QString function arguments const QString&, and string with const string&. Make ShowCritical() wrapper for invoking a message box from another thread. Add combo box to TwoButtonWidget and rename.
2014-10-14 11:53:15 -04:00
intmax_t i = 0, varIndex = IndexInXform();
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss2 << "_" << XformIndexInEmber() << "]";
string index = ss2.str();
string power = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string radius = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string width = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string dist = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string a = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string b = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string sinC = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string cosC = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ha = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string hb = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string hc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ab = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ac = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ba = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string bc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string ca = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string cb = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2a = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2b = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2c = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2ab = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2ac = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string s2bc = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string width1 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string width2 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string width3 = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string absN = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
string cn = "parVars[" + ToUpper(m_Params[i++].Name()) + index;
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
ss << "\t{\n"
<< "\t\tint m, n;\n"
<< "\t\treal_t alpha, beta, offsetAl, offsetBe, offsetGa, x, y;\n"
<< "\n"
<< "\t\t{\n"//DirectTrilinear function extracted out here.
<< "\t\t alpha = vIn.y + " << radius << ";\n"
<< "\t\t beta = vIn.x * " << sinC << " - vIn.y * " << cosC << " + " << radius << ";\n"
<< "\t\t}\n"
<< "\n"
<< "\t\tm = floor(alpha / " << s2a << ");\n"
<< "\t\toffsetAl = alpha - m * " << s2a << ";\n"
<< "\t\tn = floor(beta / " << s2b << ");\n"
<< "\t\toffsetBe = beta - n * " << s2b << ";\n"
<< "\t\toffsetGa = " << s2c << " - " << ac << " * offsetAl - " << bc << " * offsetBe;\n"
<< "\n"
<< "\t\tif (offsetGa > 0)\n"
<< "\t\t{\n"
<< "\n"
<< "\t\t Hex(offsetAl, offsetBe, offsetGa,\n"
<< "\t\t " << width << ",\n"
<< "\t\t " << ha << ",\n"
<< "\t\t " << hb << ",\n"
<< "\t\t " << hc << ",\n"
<< "\t\t " << ab << ",\n"
<< "\t\t " << ba << ",\n"
<< "\t\t " << bc << ",\n"
<< "\t\t " << ca << ",\n"
<< "\t\t " << cb << ",\n"
<< "\t\t " << s2a << ",\n"
<< "\t\t " << s2b << ",\n"
<< "\t\t " << s2ab << ",\n"
<< "\t\t " << s2ac << ",\n"
<< "\t\t " << s2bc << ",\n"
<< "\t\t " << width1 << ",\n"
<< "\t\t " << width2 << ",\n"
<< "\t\t " << width3 << ",\n"
<< "\t\t &alpha, &beta, mwc);\n"
<< "\n"
<< "\t\t}\n"
<< "\t\telse\n"
<< "\t\t{\n"
<< "\t\t offsetAl = " << s2a << " - offsetAl;\n"
<< "\t\t offsetBe = " << s2b << " - offsetBe;\n"
<< "\t\t offsetGa = -offsetGa;\n"
<< "\n"
<< "\t\t Hex(offsetAl, offsetBe, offsetGa,\n"
<< "\t\t " << width << ",\n"
<< "\t\t " << ha << ",\n"
<< "\t\t " << hb << ",\n"
<< "\t\t " << hc << ",\n"
<< "\t\t " << ab << ",\n"
<< "\t\t " << ba << ",\n"
<< "\t\t " << bc << ",\n"
<< "\t\t " << ca << ",\n"
<< "\t\t " << cb << ",\n"
<< "\t\t " << s2a << ",\n"
<< "\t\t " << s2b << ",\n"
<< "\t\t " << s2ab << ",\n"
<< "\t\t " << s2ac << ",\n"
<< "\t\t " << s2bc << ",\n"
<< "\t\t " << width1 << ",\n"
<< "\t\t " << width2 << ",\n"
<< "\t\t " << width3 << ",\n"
<< "\t\t &alpha, &beta, mwc);\n"
<< "\n"
<< "\t\t alpha = " << s2a << " - alpha;\n"
<< "\t\t beta = " << s2b << " - beta;\n"
<< "\t\t}\n"
<< "\n"
<< "\t\talpha += m * " << s2a << ";\n"
<< "\t\tbeta += n * " << s2b << ";\n"
<< "\n"
<< "\t\t{\n"//InverseTrilinear function extracted out here.
<< "\t\t real_t inx = (beta - " << radius << " + (alpha - " << radius << ") * " << cosC << ") / " << sinC << ";\n"
<< "\t\t real_t iny = alpha - " << radius << ";\n"
<< "\t\t real_t angle = (atan2(iny, inx) + M_2PI * MwcNextRange(mwc, (int)" << absN << ")) / " << power << ";\n"
<< "\t\t real_t r = xform->m_VariationWeights[" << varIndex << "] * pow(SQR(inx) + SQR(iny), " << cn << ");\n"
<< "\n"
<< "\t\t x = r * cos(angle);\n"
<< "\t\t y = r * sin(angle);\n"
<< "\t\t}\n"
<< "\n"
<< "\t\tvOut.x = xform->m_VariationWeights[" << varIndex << "] * x;\n"
<< "\t\tvOut.y = xform->m_VariationWeights[" << varIndex << "] * y;\n"
Numerous fixes 0.4.0.5 Beta 07/18/2014 --User Changes Allow for vibrancy values > 1. Add flatten and unflatten menu items. Automatically flatten like Apophysis does. Add plugin and new_linear tags to Xml to be compatible with Apophysis. --Bug Fixes Fix blur, blur3d, bubble, cropn, cross, curl, curl3d, epispiral, ho, julia3d, julia3dz, loonie, mirror_x, mirror_y, mirror_z, rotate_x, sinusoidal, spherical, spherical3d, stripes. Unique filename on final render was completely broken. Two severe OpenCL bugs. Random seeds were biased and fusing was being reset too often leading to results that differ from the CPU. Subtle, but sometimes severe bug in the setup of the xaos weights. Use properly defined epsilon by getting the value from std::numeric_limits, rather than hard coding 1e-6 or 1e-10. Omit incorrect usage of epsilon everywhere. It should not be automatically added to denominators. Rather, it should only be used if the denominator is zero. Force final render progress bars to 100 on completion. Sometimes they didn't seem to make it there. Make variation name and params comparisons be case insensitive. --Code Changes Make ForEach and FindIf wrappers around std::for_each and std::find_if.
2014-07-19 02:33:18 -04:00
<< "\t\tvOut.z = " << ((m_VarType == VARTYPE_REG) ? "0" : "vIn.z") << ";\n"
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
<< "\t}\n";
return ss.str();
}
--Code changes -Better use of const functions. -Ember.pro should not define BUILDING_EMBERCL, instead define BUILDING_EMBER.
2015-06-29 23:13:53 -04:00
virtual string OpenCLFuncsString() const
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
return
"\n"
"void Hex(real_t al, real_t be, real_t ga,\n"
" real_t width,\n"
" real_t ha,\n"
" real_t hb,\n"
" real_t hc,\n"
" real_t ab,\n"
" real_t ba,\n"
" real_t bc,\n"
" real_t ca,\n"
" real_t cb,\n"
" real_t s2a,\n"
" real_t s2b,\n"
" real_t s2ab,\n"
" real_t s2ac,\n"
" real_t s2bc,\n"
" real_t width1,\n"
" real_t width2,\n"
" real_t width3,\n"
" real_t* al1, real_t* be1, uint2* mwc)\n"
"{\n"
" real_t ga1, de1, r = MwcNext01(mwc);\n"
"\n"
" if (be < al)\n"
" {\n"
" if (ga < be)\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" de1 = width * be;\n"
" ga1 = width * ga;\n"
" }\n"
" else\n"
" {\n"
" ga1 = width1 * ga + width2 * hc * ga / be;\n"
" de1 = width1 * be + width2 * s2ab * (3 - ga / be);\n"
" }\n"
"\n"
" *al1 = s2a - ba * de1 - ca * ga1;\n"
" *be1 = de1;\n"
" }\n"
" else\n"
" {\n"
" if (ga < al)\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" ga1 = width * ga;\n"
" de1 = width * be;\n"
" }\n"
" else\n"
" {\n"
" de1 = width1 * be + width2 * hb * be / ga;\n"
" ga1 = width1 * ga + width2 * s2ac * (3 - be / ga);\n"
" }\n"
"\n"
" *al1 = s2a - ba * de1 - ca * ga1;\n"
" *be1 = de1;\n"
" }\n"
" else\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" *al1 = width * al;\n"
" *be1 = width * be;\n"
" }\n"
" else\n"
" {\n"
" *be1 = width1 * be + width2 * hb * be / al;\n"
" *al1 = width1 * al + width2 * s2ac * (3 - be / al);\n"
" }\n"
" }\n"
" }\n"
" }\n"
" else\n"
" {\n"
" if (ga < al)\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" de1 = width * al;\n"
" ga1 = width * ga;\n"
" }\n"
" else\n"
" {\n"
" ga1 = width1 * ga + width2 * hc * ga / al;\n"
" de1 = width1 * al + width2 * s2ab * (3 - ga / al);\n"
" }\n"
"\n"
" *be1 = s2b - ab * de1 - cb * ga1;\n"
" *al1 = de1;\n"
" }\n"
" else\n"
" {\n"
" if (ga < be)\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" ga1 = width * ga;\n"
" de1 = width * al;\n"
" }\n"
" else\n"
" {\n"
" de1 = width1 * al + width2 * ha * al / ga;\n"
" ga1 = width1 * ga + width2 * s2bc * (3 - al / ga);\n"
" }\n"
"\n"
" *be1 = s2b - ab * de1 - cb * ga1;\n"
" *al1 = de1;\n"
" }\n"
" else\n"
" {\n"
" if (r >= width3)\n"
" {\n"
" *be1 = width * be;\n"
" *al1 = width * al;\n"
" }\n"
" else\n"
" {\n"
" *al1 = width1 * al + width2 * ha * al / be;\n"
" *be1 = width1 * be + width2 * s2bc * (3 - al / be);\n"
" }\n"
" }\n"
" }\n"
" }\n"
"}\n"
"\n"
;
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
virtual void Precalc() override
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
{
T s2, sinA2, cosA2, sinB2, cosB2, sinC2, cosC2;
T br = T(0.047) + m_A;
T cr = T(0.047) + m_B;
T ar = T(M_PI) - br - cr;
T temp = ar / 2;
sincos(temp, &sinA2, &cosA2);
temp = br / 2;
sincos(temp, &sinB2, &cosB2);
temp = cr / 2;
sincos(temp, &sinC2, &cosC2);
sincos(cr, &m_SinC, &m_CosC);
T a = m_Radius * (sinC2 / cosC2 + sinB2 / cosB2);
T b = m_Radius * (sinC2 / cosC2 + sinA2 / cosA2);
T c = m_Radius * (sinB2 / cosB2 + sinA2 / cosA2);
m_Width1 = 1 - m_Width;
m_Width2 = 2 * m_Width;
m_Width3 = 1 - m_Width * m_Width;
s2 = m_Radius * (a + b + c);
m_Ha = s2 / a / 6;
m_Hb = s2 / b / 6;
m_Hc = s2 / c / 6;
m_Ab = a / b;// a div on b
m_Ac = a / c;
m_Ba = b / a;
m_Bc = b / c;
m_Ca = c / a;
m_Cb = c / b;
m_S2a = 6 * m_Ha;
m_S2b = 6 * m_Hb;
m_S2c = 6 * m_Hc;
m_S2bc = s2 / (b + c) / 6;
m_S2ab = s2 / (a + b) / 6;
m_S2ac = s2 / (a + c) / 6;
if (m_Power == 0)
m_Power = 2;
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
m_AbsN = T(int(fabs(m_Power)));
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
m_Cn = m_Dist / m_Power / 2;
}
protected:
void Init()
{
string prefix = Prefix();
m_Params.clear();
m_Params.push_back(ParamWithName<T>(&m_Power, prefix + "xtrb_power", 2, INTEGER_NONZERO));
m_Params.push_back(ParamWithName<T>(&m_Radius, prefix + "xtrb_radius", 1));
m_Params.push_back(ParamWithName<T>(&m_Width, prefix + "xtrb_width", T(0.5)));
m_Params.push_back(ParamWithName<T>(&m_Dist, prefix + "xtrb_dist", 1));
m_Params.push_back(ParamWithName<T>(&m_A, prefix + "xtrb_a", 1));
m_Params.push_back(ParamWithName<T>(&m_B, prefix + "xtrb_b", 1));
m_Params.push_back(ParamWithName<T>(true, &m_SinC, prefix + "xtrb_sinc"));//Precalcs.
m_Params.push_back(ParamWithName<T>(true, &m_CosC, prefix + "xtrb_cosc"));
m_Params.push_back(ParamWithName<T>(true, &m_Ha, prefix + "xtrb_ha"));
m_Params.push_back(ParamWithName<T>(true, &m_Hb, prefix + "xtrb_hb"));
m_Params.push_back(ParamWithName<T>(true, &m_Hc, prefix + "xtrb_hc"));
m_Params.push_back(ParamWithName<T>(true, &m_Ab, prefix + "xtrb_ab"));
m_Params.push_back(ParamWithName<T>(true, &m_Ac, prefix + "xtrb_ac"));
m_Params.push_back(ParamWithName<T>(true, &m_Ba, prefix + "xtrb_ba"));
m_Params.push_back(ParamWithName<T>(true, &m_Bc, prefix + "xtrb_bc"));
m_Params.push_back(ParamWithName<T>(true, &m_Ca, prefix + "xtrb_ca"));
m_Params.push_back(ParamWithName<T>(true, &m_Cb, prefix + "xtrb_cb"));
m_Params.push_back(ParamWithName<T>(true, &m_S2a, prefix + "xtrb_s2a"));
m_Params.push_back(ParamWithName<T>(true, &m_S2b, prefix + "xtrb_s2b"));
m_Params.push_back(ParamWithName<T>(true, &m_S2c, prefix + "xtrb_s2c"));
m_Params.push_back(ParamWithName<T>(true, &m_S2ab, prefix + "xtrb_s2ab"));
m_Params.push_back(ParamWithName<T>(true, &m_S2ac, prefix + "xtrb_s2ac"));
m_Params.push_back(ParamWithName<T>(true, &m_S2bc, prefix + "xtrb_s2bc"));
m_Params.push_back(ParamWithName<T>(true, &m_Width1, prefix + "xtrb_width1"));
m_Params.push_back(ParamWithName<T>(true, &m_Width2, prefix + "xtrb_width2"));
m_Params.push_back(ParamWithName<T>(true, &m_Width3, prefix + "xtrb_width3"));
m_Params.push_back(ParamWithName<T>(true, &m_AbsN, prefix + "xtrb_absn"));
m_Params.push_back(ParamWithName<T>(true, &m_Cn, prefix + "xtrb_cn"));
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
private:
inline void DirectTrilinear(T x, T y, T& al, T& be)
{
al = y + m_Radius;
be = x * m_SinC - y * m_CosC + m_Radius;
}
inline void InverseTrilinear(T al, T be, T& x, T& y, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand)
{
T inx = (be - m_Radius + (al - m_Radius) * m_CosC) / m_SinC;
T iny = al - m_Radius;
More Linux work. Convert all casts to new style, away from legacy.
2014-12-07 02:51:44 -05:00
T angle = (atan2(iny, inx) + M_2PI * (rand.Rand(int(m_AbsN)))) / m_Power;
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
T r = m_Weight * pow(SQR(inx) + SQR(iny), m_Cn);
x = r * cos(angle);
y = r * sin(angle);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
void Hex(T al, T be, T ga, T& al1, T& be1, QTIsaac<ISAAC_SIZE, ISAAC_INT>& rand)
{
T ga1, de1, r = rand.Frand01<T>();
if (be < al)
{
if (ga < be)
{
if (r >= m_Width3)
{
de1 = m_Width * be;
ga1 = m_Width * ga;
}
else
{
ga1 = m_Width1 * ga + m_Width2 * m_Hc * ga / be;
de1 = m_Width1 * be + m_Width2 * m_S2ab * (3 - ga / be);
}
al1 = m_S2a - m_Ba * de1 - m_Ca * ga1;
be1 = de1;
}
else
{
if (ga < al)
{
if (r >= m_Width3)
{
ga1 = m_Width * ga;
de1 = m_Width * be;
}
else
{
de1 = m_Width1 * be + m_Width2 * m_Hb * be / ga;
ga1 = m_Width1 * ga + m_Width2 * m_S2ac * (3 - be / ga);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
al1 = m_S2a - m_Ba * de1 - m_Ca * ga1;
be1 = de1;
}
else
{
if (r >= m_Width3)
{
al1 = m_Width * al;
be1 = m_Width * be;
}
else
{
be1 = m_Width1 * be + m_Width2 * m_Hb * be / al;
al1 = m_Width1 * al + m_Width2 * m_S2ac * (3 - be / al);
}
}
}
}
else
{
if (ga < al)
{
if (r >= m_Width3)
{
de1 = m_Width * al;
ga1 = m_Width * ga;
}
else
{
ga1 = m_Width1 * ga + m_Width2 * m_Hc * ga / al;
de1 = m_Width1 * al + m_Width2 * m_S2ab * (3 - ga / al);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
be1 = m_S2b - m_Ab * de1 - m_Cb * ga1;
al1 = de1;
}
else
{
if (ga < be)
{
if (r >= m_Width3)
{
ga1 = m_Width * ga;
de1 = m_Width * al;
}
else
{
de1 = m_Width1 * al + m_Width2 * m_Ha * al / ga;
ga1 = m_Width1 * ga + m_Width2 * m_S2bc * (3 - al / ga);
}
Initial VS2013, C++11 and linux commit. This most likely won't build and suffers from some compiler issues. More commits to follow.
2014-09-01 00:25:15 -04:00
Initial source commit Initial source commit
2014-07-08 03:11:14 -04:00
be1 = m_S2b - m_Ab * de1 - m_Cb * ga1;
al1 = de1;
}
else
{
if (r >= m_Width3)
{
be1 = m_Width * be;
al1 = m_Width * al;
}
else
{
al1 = m_Width1 * al + m_Width2 * m_Ha * al / be;
be1 = m_Width1 * be + m_Width2 * m_S2bc * (3 - al / be);
}
}
}
}
}
T m_Power;
T m_Radius;
T m_Width;
T m_Dist;
T m_A;
T m_B;
T m_SinC;//Precalcs.
T m_CosC;
T m_Ha;
T m_Hb;
T m_Hc;
T m_Ab;
T m_Ac;
T m_Ba;
T m_Bc;
T m_Ca;
T m_Cb;
T m_S2a;
T m_S2b;
T m_S2c;
T m_S2ab;
T m_S2ac;
T m_S2bc;
T m_Width1;
T m_Width2;
T m_Width3;
T m_AbsN;
T m_Cn;
};
MAKEPREPOSTPARVAR(Bubble2, bubble2, BUBBLE2)
MAKEPREPOSTPARVAR(CircleLinear, CircleLinear, CIRCLELINEAR)
MAKEPREPOSTPARVARASSIGN(CircleRand, CircleRand, CIRCLERAND, ASSIGNTYPE_SUM)
MAKEPREPOSTPARVAR(CircleTrans1, CircleTrans1, CIRCLETRANS1)
MAKEPREPOSTPARVAR(Cubic3D, cubic3D, CUBIC3D)
MAKEPREPOSTPARVAR(CubicLattice3D, cubicLattice_3D, CUBIC_LATTICE3D)
MAKEPREPOSTVAR(Foci3D, foci_3D, FOCI3D)
MAKEPREPOSTPARVAR(Ho, ho, HO)
MAKEPREPOSTPARVAR(Julia3Dq, julia3Dq, JULIA3DQ)
MAKEPREPOSTPARVARASSIGN(Line, line, LINE, ASSIGNTYPE_SUM)
MAKEPREPOSTPARVAR(Loonie3D, loonie_3D, LOONIE3D)
MAKEPREPOSTPARVAR(Mcarpet, mcarpet, MCARPET)
MAKEPREPOSTPARVAR(Waves23D, waves2_3D, WAVES23D)
MAKEPREPOSTPARVARASSIGN(Pie3D, pie3D, PIE3D, ASSIGNTYPE_SUM)
MAKEPREPOSTPARVAR(Popcorn23D, popcorn2_3D, POPCORN23D)
MAKEPREPOSTVAR(Sinusoidal3D, sinusoidal3D, SINUSOIDAL3D)
MAKEPREPOSTPARVAR(Scry3D, scry_3D, SCRY3D)
MAKEPREPOSTPARVAR(Shredlin, shredlin, SHRED_LIN)
MAKEPREPOSTPARVAR(SplitBrdr, SplitBrdr, SPLIT_BRDR)
MAKEPREPOSTVAR(Wdisc, wdisc, WDISC)
MAKEPREPOSTPARVAR(Falloff, falloff, FALLOFF)
MAKEPREPOSTPARVAR(Falloff2, falloff2, FALLOFF2)
MAKEPREPOSTPARVAR(Falloff3, falloff3, FALLOFF3)
MAKEPREPOSTPARVAR(Xtrb, xtrb, XTRB)
}
Reference in New Issue Copy Permalink