2014-07-08 03:11:14 -04:00
# pragma once
# include "Variation.h"
namespace EmberNs
{
/// <summary>
/// DC Bubble.
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
/// This accesses the summed output point in a rare and different way.
2014-07-08 03:11:14 -04:00
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCBubbleVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCBubbleVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_bubble " , eVariationId : : VAR_DC_BUBBLE , weight , true )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCBubbleVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
T r = helper . m_PrecalcSumSquares ;
2014-08-03 19:16:10 -04:00
T r4_1 = Zeps ( r / 4 + 1 ) ;
2014-07-08 03:11:14 -04:00
r4_1 = m_Weight / r4_1 ;
helper . Out . x = r4_1 * helper . In . x ;
helper . Out . y = r4_1 * helper . In . y ;
helper . Out . z = m_Weight * ( 2 / r4_1 - 1 ) ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
T sumX , sumY ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
sumX = helper . In . x ;
sumY = helper . In . y ;
}
else
{
sumX = outPoint . m_X ;
sumY = outPoint . m_Y ;
}
T tempX = helper . Out . x + sumX ;
T tempY = helper . Out . y + sumY ;
2016-01-12 23:42:12 -05:00
outPoint . m_ColorX = fmod ( std : : abs ( m_Bdcs * ( Sqr < T > ( tempX + m_CenterX ) + Sqr < T > ( tempY + m_CenterY ) ) ) , T ( 1.0 ) ) ;
2014-07-08 03:11:14 -04:00
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
string scale = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
2014-07-08 03:11:14 -04:00
string centerX = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string centerY = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string bdcs = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t real_t r = precalcSumSquares; \n "
2014-08-03 19:16:10 -04:00
< < " \t \t real_t r4_1 = Zeps(r / 4 + 1); \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t r4_1 = " < < weight < < " / r4_1; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t vOut.x = r4_1 * vIn.x; \n "
< < " \t \t vOut.y = r4_1 * vIn.y; \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.z = " < < weight < < " * (2 / r4_1 - 1); \n "
2014-07-08 03:11:14 -04:00
< < " \n "
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
< < " \t \t real_t sumX, sumY; \n \n " ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = vIn.x; \n "
< < " \t \t sumY = vIn.y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
else
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = outPoint->m_X; \n "
< < " \t \t sumY = outPoint->m_Y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
ss
2015-12-31 16:41:59 -05:00
< < " \t \t real_t tempX = vOut.x + sumX; \n "
< < " \t \t real_t tempY = vOut.y + sumY; \n "
< < " \n "
< < " \t \t outPoint->m_ColorX = fmod(fabs( " < < bdcs < < " * (Sqr(tempX + " < < centerX < < " ) + Sqr(tempY + " < < centerY < < " ))), (real_t)(1.0)); \n "
< < " \t } \n " ;
2014-07-08 03:11:14 -04:00
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
virtual vector < string > OpenCLGlobalFuncNames ( ) const override
{
return vector < string > { " Sqr " , " Zeps " } ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_Bdcs = 1 / ( m_Scale = = 0 ? T ( 10E-6 ) : m_Scale ) ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_CenterX , prefix + " dc_bubble_centerx " ) ) ; //Params.
m_Params . push_back ( ParamWithName < T > ( & m_CenterY , prefix + " dc_bubble_centery " ) ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Scale , prefix + " dc_bubble_scale " , 1 ) ) ;
2014-07-08 03:11:14 -04:00
m_Params . push_back ( ParamWithName < T > ( true , & m_Bdcs , prefix + " dc_bubble_bdcs " ) ) ; //Precalc.
}
private :
T m_CenterX ; //Params.
T m_CenterY ;
T m_Scale ;
T m_Bdcs ; //Precalc.
} ;
/// <summary>
/// DC Carpet.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCCarpetVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCCarpetVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_carpet " , eVariationId : : VAR_DC_CARPET , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCCarpetVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
int x0 = rand . RandBit ( ) ? - 1 : 1 ;
int y0 = rand . RandBit ( ) ? - 1 : 1 ;
T x = helper . In . x + x0 ;
T y = helper . In . y + y0 ;
T x0_xor_y0 = T ( x0 ^ y0 ) ;
T h = - m_H + ( 1 - x0_xor_y0 ) * m_H ;
2014-09-01 00:25:15 -04:00
helper . Out . x = m_Weight * ( m_Xform - > m_Affine . A ( ) * x + m_Xform - > m_Affine . B ( ) * y + m_Xform - > m_Affine . E ( ) ) ;
2014-07-08 03:11:14 -04:00
helper . Out . y = m_Weight * ( m_Xform - > m_Affine . C ( ) * x + m_Xform - > m_Affine . D ( ) * y + m_Xform - > m_Affine . F ( ) ) ;
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
helper . Out . z = DefaultZ ( helper ) ;
2016-01-12 23:42:12 -05:00
outPoint . m_ColorX = fmod ( std : : abs ( outPoint . m_ColorX * T ( 0.5 ) * ( 1 + h ) + x0_xor_y0 * ( 1 - h ) * T ( 0.5 ) ) , T ( 1.0 ) ) ;
2014-07-08 03:11:14 -04:00
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2014-07-08 03:11:14 -04:00
string origin = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string h = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t int x0 = (MwcNext(mwc) & 1) ? -1 : 1; \n "
< < " \t \t int y0 = (MwcNext(mwc) & 1) ? -1 : 1; \n "
< < " \t \t real_t x = vIn.x + x0; \n "
< < " \t \t real_t y = vIn.y + y0; \n "
< < " \t \t real_t x0_xor_y0 = (real_t)(x0 ^ y0); \n "
< < " \t \t real_t h = - " < < h < < " + (1 - x0_xor_y0) * " < < h < < " ; \n "
< < " \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (xform->m_A * x + xform->m_B * y + xform->m_E); \n "
< < " \t \t vOut.y = " < < weight < < " * (xform->m_C * x + xform->m_D * y + xform->m_F); \n "
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
< < " \t \t vOut.z = " < < DefaultZCl ( )
2019-05-23 22:44:27 -04:00
< < " \t \t outPoint->m_ColorX = fmod(fabs(outPoint->m_ColorX * (real_t)(0.5) * (real_t)(1 + h) + x0_xor_y0 * (real_t)(1 - h) * (real_t)(0.5)), (real_t)(1.0)); \n "
2014-07-08 03:11:14 -04:00
< < " \t } \n " ;
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_H = T ( 0.1 ) * m_Origin ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Origin , prefix + " dc_carpet_origin " ) ) ; //Params.
m_Params . push_back ( ParamWithName < T > ( true , & m_H , prefix + " dc_carpet_h " ) ) ; //Precalc.
}
private :
T m_Origin ; //Params.
T m_H ; //Precalc.
} ;
/// <summary>
/// DC Cube.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCCubeVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCCubeVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_cube " , eVariationId : : VAR_DC_CUBE , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCCubeVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
T x , y , z ;
T p = 2 * rand . Frand01 < T > ( ) - 1 ;
T q = 2 * rand . Frand01 < T > ( ) - 1 ;
2014-12-06 00:05:09 -05:00
uint i = rand . Rand ( 3 ) ;
uint j = rand . RandBit ( ) ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
switch ( i )
{
case 0 :
x = m_Weight * ( j ? - 1 : 1 ) ;
y = m_Weight * p ;
z = m_Weight * q ;
if ( j )
outPoint . m_ColorX = m_ClampC1 ;
else
outPoint . m_ColorX = m_ClampC2 ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
break ;
2015-12-31 16:41:59 -05:00
2014-07-08 03:11:14 -04:00
case 1 :
x = m_Weight * p ;
y = m_Weight * ( j ? - 1 : 1 ) ;
z = m_Weight * q ;
if ( j )
outPoint . m_ColorX = m_ClampC3 ;
else
outPoint . m_ColorX = m_ClampC4 ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
break ;
2015-12-31 16:41:59 -05:00
2014-07-08 03:11:14 -04:00
case 2 :
2014-09-10 01:41:26 -04:00
default :
2014-07-08 03:11:14 -04:00
x = m_Weight * p ;
y = m_Weight * q ;
z = m_Weight * ( j ? - 1 : 1 ) ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
if ( j )
outPoint . m_ColorX = m_ClampC5 ;
else
outPoint . m_ColorX = m_ClampC6 ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
break ;
}
helper . Out . x = x * m_DcCubeX ;
helper . Out . y = y * m_DcCubeY ;
helper . Out . z = z * m_DcCubeZ ;
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
string cubeC1 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
string cubeC2 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeC3 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeC4 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeC5 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeC6 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeX = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeY = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cubeZ = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2014-07-08 03:11:14 -04:00
string clampC1 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
string clampC2 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string clampC3 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string clampC4 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string clampC5 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string clampC6 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
ss < < " \t { \n "
< < " \t \t real_t x, y, z; \n "
< < " \t \t real_t p = 2 * MwcNext01(mwc) - 1; \n "
< < " \t \t real_t q = 2 * MwcNext01(mwc) - 1; \n "
--Bug fixes
-Fix improper usage of rand() in cpow2, cpow3, hypertile1, hypertile3D1, hypertile3D2, juliac, juliaq.
-Fix program crashing during density filtering on some Nvidia cards.
-hypertile3D1 was wrong.
-Parsing phoenix_julia when coming from Apophysis was wrong.
-Density filtering was freezing on certain Nvidia cards.
--Code changes
-Optimize juliac, npolar.
-Add a new function Crand() which behaves like the legacy C rand() which returns an integer between 0 and 32766, inclusive.
-Use RandBit() in some places.
-Remove Zeps() from vignette, it's not needed.
-Restructure OpenCL code for density filtering such that it does not hang after being compiled on some Nvidia cards, such as the gtx 1660. Remove barriers from conditionals where possible.
2020-12-29 00:46:55 -05:00
< < " \t \t uint i = MwcNextRange(mwc, 3u); \n "
< < " \t \t uint j = MwcNext(mwc) & 1u; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t switch (i) \n "
< < " \t \t { \n "
< < " \t \t case 0: \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t x = " < < weight < < " * (j ? -1 : 1); \n "
< < " \t \t y = " < < weight < < " * p; \n "
< < " \t \t z = " < < weight < < " * q; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t if (j) \n "
< < " \t \t outPoint->m_ColorX = " < < clampC1 < < " ; \n "
< < " \t \t else \n "
< < " \t \t outPoint->m_ColorX = " < < clampC2 < < " ; \n "
< < " \n "
< < " \t \t break; \n "
< < " \t \t case 1: \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t x = " < < weight < < " * p; \n "
< < " \t \t y = " < < weight < < " * (j ? -1 : 1); \n "
< < " \t \t z = " < < weight < < " * q; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t if (j) \n "
< < " \t \t outPoint->m_ColorX = " < < clampC3 < < " ; \n "
< < " \t \t else \n "
< < " \t \t outPoint->m_ColorX = " < < clampC4 < < " ; \n "
< < " \n "
< < " \t \t break; \n "
< < " \t \t case 2: \n "
2020-04-11 23:07:12 -04:00
< < " \t \t default: \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t x = " < < weight < < " * p; \n "
< < " \t \t y = " < < weight < < " * q; \n "
< < " \t \t z = " < < weight < < " * (j ? -1 : 1); \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t if (j) \n "
< < " \t \t outPoint->m_ColorX = " < < clampC5 < < " ; \n "
< < " \t \t else \n "
< < " \t \t outPoint->m_ColorX = " < < clampC6 < < " ; \n "
< < " \n "
< < " \t \t break; \n "
< < " \t \t } \n "
< < " \n "
< < " \t \t vOut.x = x * " < < cubeX < < " ; \n "
< < " \t \t vOut.y = y * " < < cubeY < < " ; \n "
< < " \t \t vOut.z = z * " < < cubeZ < < " ; \n "
< < " \t } \n " ;
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_ClampC1 = Clamp < T > ( m_DcCubeC1 , 0 , 1 ) ;
m_ClampC2 = Clamp < T > ( m_DcCubeC2 , 0 , 1 ) ;
m_ClampC3 = Clamp < T > ( m_DcCubeC3 , 0 , 1 ) ;
m_ClampC4 = Clamp < T > ( m_DcCubeC4 , 0 , 1 ) ;
m_ClampC5 = Clamp < T > ( m_DcCubeC5 , 0 , 1 ) ;
m_ClampC6 = Clamp < T > ( m_DcCubeC6 , 0 , 1 ) ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC1 , prefix + " dc_cube_c1 " ) ) ; //Params.
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC2 , prefix + " dc_cube_c2 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC3 , prefix + " dc_cube_c3 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC4 , prefix + " dc_cube_c4 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC5 , prefix + " dc_cube_c5 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeC6 , prefix + " dc_cube_c6 " ) ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeX , prefix + " dc_cube_x " , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeY , prefix + " dc_cube_y " , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_DcCubeZ , prefix + " dc_cube_z " , 1 ) ) ;
2014-07-08 03:11:14 -04:00
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC1 , prefix + " dc_cube_clamp_c1 " ) ) ; //Precalc.
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC2 , prefix + " dc_cube_clamp_c2 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC3 , prefix + " dc_cube_clamp_c3 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC4 , prefix + " dc_cube_clamp_c4 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC5 , prefix + " dc_cube_clamp_c5 " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_ClampC6 , prefix + " dc_cube_clamp_c6 " ) ) ;
}
private :
T m_DcCubeC1 ; //Params.
T m_DcCubeC2 ;
T m_DcCubeC3 ;
T m_DcCubeC4 ;
T m_DcCubeC5 ;
T m_DcCubeC6 ;
T m_DcCubeX ;
T m_DcCubeY ;
T m_DcCubeZ ;
T m_ClampC1 ; //Precalc.
T m_ClampC2 ;
T m_ClampC3 ;
T m_ClampC4 ;
T m_ClampC5 ;
T m_ClampC6 ;
} ;
/// <summary>
/// DC Cylinder.
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
/// This accesses the summed output point in a rare and different way.
2014-07-08 03:11:14 -04:00
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCCylinderVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCCylinderVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_cylinder " , eVariationId : : VAR_DC_CYLINDER , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCCylinderVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
T temp = rand . Frand01 < T > ( ) * M_2PI ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
T sr = std : : sin ( temp ) ;
T cr = std : : cos ( temp ) ;
2014-07-08 03:11:14 -04:00
T r = m_Blur * ( rand . Frand01 < T > ( ) + rand . Frand01 < T > ( ) + rand . Frand01 < T > ( ) + rand . Frand01 < T > ( ) - 2 ) ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
helper . Out . x = m_Weight * std : : sin ( helper . In . x + r * sr ) * m_X ;
2014-07-08 03:11:14 -04:00
helper . Out . y = r + helper . In . y * m_Y ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
helper . Out . z = m_Weight * std : : cos ( helper . In . x + r * cr ) ;
T sumX , sumY ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
sumX = helper . In . x ;
sumY = helper . In . y ;
}
else
{
sumX = outPoint . m_X ;
sumY = outPoint . m_Y ;
}
2014-07-08 03:11:14 -04:00
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
T tempX = helper . Out . x + sumX ;
T tempY = helper . Out . y + sumY ;
2016-01-12 23:42:12 -05:00
outPoint . m_ColorX = fmod ( std : : abs ( T ( 0.5 ) * ( m_Ldcs * ( ( m_Cosa * tempX + m_Sina * tempY + m_Offset ) ) + 1 ) ) , T ( 1.0 ) ) ;
2014-07-08 03:11:14 -04:00
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2014-07-08 03:11:14 -04:00
string offset = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string angle = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string scale = " 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 blur = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string sina = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
string cosa = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string ldcs = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string ldca = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t real_t temp = MwcNext(mwc) * M_2PI; \n "
< < " \t \t real_t sr = sin(temp); \n "
< < " \t \t real_t cr = cos(temp); \n "
< < " \t \t real_t r = " < < blur < < " * (MwcNext01(mwc) + MwcNext01(mwc) + MwcNext01(mwc) + MwcNext01(mwc) - 2); \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * sin(vIn.x + r * sr)* " < < x < < " ; \n "
2014-07-08 03:11:14 -04:00
< < " \t \t vOut.y = r + vIn.y * " < < y < < " ; \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.z = " < < weight < < " * cos(vIn.x + r * cr); \n "
2014-07-08 03:11:14 -04:00
< < " \n "
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
< < " \t \t real_t sumX, sumY; \n \n " ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = vIn.x; \n "
< < " \t \t sumY = vIn.y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
else
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = outPoint->m_X; \n "
< < " \t \t sumY = outPoint->m_Y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
ss
2015-12-31 16:41:59 -05:00
< < " \t \t real_t tempX = vOut.x + sumX; \n "
< < " \t \t real_t tempY = vOut.y + sumY; \n "
< < " \n "
< < " \t \t outPoint->m_ColorX = fmod(fabs((real_t)(0.5) * ( " < < ldcs < < " * (( " < < cosa < < " * tempX + " < < sina < < " * tempY + " < < offset < < " )) + (real_t)(1.0))), (real_t)(1.0)); \n "
< < " \t } \n " ;
2014-07-08 03:11:14 -04:00
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
sincos ( m_Angle , & m_Sina , & m_Cosa ) ;
m_Ldcs = 1 / ( m_Scale = = 0.0 ? T ( 10E-6 ) : m_Scale ) ;
m_Ldca = m_Offset * T ( M_PI ) ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Offset , prefix + " dc_cylinder_offset " ) ) ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Angle , prefix + " dc_cylinder_angle " ) ) ; //Original used a prefix of dc_cyl_, which is incompatible with Ember's design.
m_Params . push_back ( ParamWithName < T > ( & m_Scale , prefix + " dc_cylinder_scale " , T ( 0.5 ) ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_X , prefix + " dc_cylinder_x " , T ( 0.125 ) ) ) ; //Original used a prefix of cyl_, which is incompatible with Ember's design.
m_Params . push_back ( ParamWithName < T > ( & m_Y , prefix + " dc_cylinder_y " , T ( 0.125 ) ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Blur , prefix + " dc_cylinder_blur " , 1 ) ) ;
2014-07-08 03:11:14 -04:00
m_Params . push_back ( ParamWithName < T > ( true , & m_Sina , prefix + " dc_cylinder_sina " ) ) ; //Precalc.
m_Params . push_back ( ParamWithName < T > ( true , & m_Cosa , prefix + " dc_cylinder_cosa " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_Ldcs , prefix + " dc_cylinder_ldcs " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_Ldca , prefix + " dc_cylinder_ldca " ) ) ;
}
private :
T m_Offset ; //Params.
T m_Angle ;
T m_Scale ;
T m_X ;
T m_Y ;
T m_Blur ;
T m_Sina ; //Precalc.
T m_Cosa ;
T m_Ldcs ;
T m_Ldca ;
} ;
/// <summary>
/// DC GridOut.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCGridOutVariation : public Variation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCGridOutVariation ( T weight = 1.0 ) : Variation < T > ( " dc_gridout " , eVariationId : : VAR_DC_GRIDOUT , weight ) { }
2014-07-08 03:11:14 -04:00
VARCOPY ( DCGridOutVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
2016-03-04 21:54:06 -05:00
T x = VarFuncs < T > : : LRint ( helper . In . x ) ;
T y = VarFuncs < T > : : LRint ( helper . In . y ) ;
2014-07-08 03:11:14 -04:00
T c = outPoint . m_ColorX ;
if ( y < = 0 )
{
if ( x > 0 )
{
if ( - y > = x )
{
helper . Out . x = m_Weight * ( helper . In . x + 1 ) ;
helper . Out . y = m_Weight * helper . In . y ;
c + = T ( 0.25 ) ;
}
else
{
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * ( helper . In . y + 1 ) ;
c + = T ( 0.75 ) ;
}
}
else
{
if ( y < = x )
{
helper . Out . x = m_Weight * ( helper . In . x + 1 ) ;
helper . Out . y = m_Weight * helper . In . y ;
c + = T ( 0.25 ) ;
}
else
{
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * ( helper . In . y - 1 ) ;
c + = T ( 0.75 ) ;
}
}
}
else
{
if ( x > 0 )
{
if ( y > = x )
{
helper . Out . x = m_Weight * ( helper . In . x - 1 ) ;
helper . Out . y = m_Weight * helper . In . y ;
c + = T ( 0.25 ) ;
}
else
{
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * ( helper . In . y + 1 ) ;
c + = T ( 0.75 ) ;
}
}
else
{
if ( y > - x )
{
helper . Out . x = m_Weight * ( helper . In . x - 1 ) ;
helper . Out . y = m_Weight * helper . In . y ;
c + = T ( 0.25 ) ;
}
else
{
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * ( helper . In . y - 1 ) ;
c + = T ( 0.75 ) ;
}
}
}
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
helper . Out . z = DefaultZ ( helper ) ;
2014-07-08 03:11:14 -04:00
outPoint . m_ColorX = fmod ( c , T ( 1.0 ) ) ;
}
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss ;
2014-10-14 11:53:15 -04:00
intmax_t varIndex = IndexInXform ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t real_t x = LRint(vIn.x); \n "
< < " \t \t real_t y = LRint(vIn.y); \n "
< < " \t \t real_t c = outPoint->m_ColorX; \n "
< < " \n "
< < " \t \t if (y <= 0) \n "
< < " \t \t { \n "
< < " \t \t if (x > 0) \n "
< < " \t \t { \n "
< < " \t \t if (-y >= x) \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (vIn.x + 1); \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.25); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * (vIn.y + 1); \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.75); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t if (y <= x) \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (vIn.x + 1); \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.25); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * (vIn.y - 1); \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.75); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t if (x > 0) \n "
< < " \t \t { \n "
< < " \t \t if (y >= x) \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (vIn.x - 1); \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.25); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * (vIn.y + 1); \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.75); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t if (y > -x) \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (vIn.x - 1); \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.25); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * (vIn.y - 1); \n "
2015-02-26 16:18:55 -05:00
< < " \t \t c += (real_t)(0.75); \n "
2014-07-08 03:11:14 -04:00
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t } \n "
< < " \n "
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
< < " \t \t vOut.z = " < < DefaultZCl ( )
2015-02-26 16:18:55 -05:00
< < " \t \t outPoint->m_ColorX = fmod(c, (real_t)(1.0)); \n "
2014-07-08 03:11:14 -04:00
< < " \t } \n " ;
return ss . str ( ) ;
}
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
virtual vector < string > OpenCLGlobalFuncNames ( ) const override
{
return vector < string > { " LRint " } ;
}
2014-07-08 03:11:14 -04:00
} ;
/// <summary>
/// DC Linear.
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
/// This accesses the summed output point in a rare and different way.
2014-07-08 03:11:14 -04:00
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCLinearVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCLinearVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_linear " , eVariationId : : VAR_DC_LINEAR , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCLinearVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * helper . In . y ;
helper . Out . z = m_Weight * helper . In . z ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
T sumX , sumY ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
sumX = helper . In . x ;
sumY = helper . In . y ;
}
else
{
sumX = outPoint . m_X ;
sumY = outPoint . m_Y ;
}
T tempX = helper . Out . x + sumX ;
T tempY = helper . Out . y + sumY ;
2016-01-12 23:42:12 -05:00
outPoint . m_ColorX = fmod ( std : : abs ( T ( 0.5 ) * ( m_Ldcs * ( ( m_Cosa * tempX + m_Sina * tempY + m_Offset ) ) + T ( 1.0 ) ) ) , T ( 1.0 ) ) ;
2014-07-08 03:11:14 -04:00
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2014-07-08 03:11:14 -04:00
string offset = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string angle = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string scale = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string ldcs = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
string ldca = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string sina = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string cosa = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
< < " \t \t vOut.z = " < < weight < < " * vIn.z; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
< < " \t \t real_t sumX, sumY; \n \n " ;
2016-01-04 19:50:15 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_PRE )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = vIn.x; \n "
< < " \t \t sumY = vIn.y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
else
{
ss
2015-12-31 16:41:59 -05:00
< < " \t \t sumX = outPoint->m_X; \n "
< < " \t \t sumY = outPoint->m_Y; \n " ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
}
ss
2015-12-31 16:41:59 -05:00
< < " \t \t real_t tempX = vOut.x + sumX; \n "
< < " \t \t real_t tempY = vOut.y + sumY; \n "
< < " \n "
< < " \t \t outPoint->m_ColorX = fmod(fabs((real_t)(0.5) * ( " < < ldcs < < " * (( " < < cosa < < " * tempX + " < < sina < < " * tempY + " < < offset < < " )) + (real_t)(1.0))), (real_t)(1.0)); \n "
< < " \t } \n " ;
2014-07-08 03:11:14 -04:00
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_Ldcs = 1 / ( m_Scale = = 0 ? T ( 10E-6 ) : m_Scale ) ;
m_Ldca = m_Offset * T ( M_PI ) ;
sincos ( m_Angle , & m_Sina , & m_Cosa ) ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Offset , prefix + " dc_linear_offset " ) ) ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Angle , prefix + " dc_linear_angle " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Scale , prefix + " dc_linear_scale " , 1 ) ) ;
2014-07-08 03:11:14 -04:00
m_Params . push_back ( ParamWithName < T > ( true , & m_Ldcs , prefix + " dc_linear_ldcs " ) ) ; //Precalc.
m_Params . push_back ( ParamWithName < T > ( true , & m_Ldca , prefix + " dc_linear_ldca " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_Sina , prefix + " dc_linear_sina " ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_Cosa , prefix + " dc_linear_cosa " ) ) ;
}
private :
T m_Offset ; //Params.
T m_Angle ;
T m_Scale ;
T m_Ldcs ; //Precalc.
T m_Ldca ;
T m_Sina ;
T m_Cosa ;
} ;
/// <summary>
/// DC Triangle.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCTriangleVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCTriangleVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_triangle " , eVariationId : : VAR_DC_TRIANGLE , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCTriangleVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
// set up triangle
2014-09-01 00:25:15 -04:00
const T
2014-07-08 03:11:14 -04:00
xx = m_Xform - > m_Affine . A ( ) , xy = m_Xform - > m_Affine . B ( ) , // X
yx = m_Xform - > m_Affine . C ( ) * - 1 , yy = m_Xform - > m_Affine . D ( ) * - 1 , // Y
ox = m_Xform - > m_Affine . E ( ) , oy = m_Xform - > m_Affine . F ( ) , // O
px = helper . In . x - ox , py = helper . In . y - oy ; // P
// calculate dot products
const T dot00 = xx * xx + xy * xy ; // X * X
const T dot01 = xx * yx + xy * yy ; // X * Y
const T dot02 = xx * px + xy * py ; // X * P
const T dot11 = yx * yx + yy * yy ; // Y * Y
const T dot12 = yx * px + yy * py ; // Y * P
// calculate barycentric coordinates
const T denom = ( dot00 * dot11 - dot01 * dot01 ) ;
const T num_u = ( dot11 * dot02 - dot01 * dot12 ) ;
const T num_v = ( dot00 * dot12 - dot01 * dot02 ) ;
// u, v must not be constant
T u = num_u / denom ;
T v = num_v / denom ;
int inside = 0 , f = 1 ;
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
// case A - point escapes edge XY
if ( u + v > 1 )
{
f = - 1 ;
if ( u > v )
{
ClampLteRef < T > ( u , 1 ) ;
v = 1 - u ;
}
else
{
ClampLteRef < T > ( v , 1 ) ;
u = 1 - v ;
}
}
else if ( ( u < 0 ) | | ( v < 0 ) ) // case B - point escapes either edge OX or OY
{
ClampRef < T > ( u , 0 , 1 ) ;
ClampRef < T > ( v , 0 , 1 ) ;
}
else
{
inside = 1 ; // case C - point is in triangle
}
2014-09-01 00:25:15 -04:00
2014-07-08 03:11:14 -04:00
// handle outside points
if ( m_ZeroEdges & & ! inside )
{
u = v = 0 ;
}
else if ( ! inside )
{
u = ( u + rand . Frand01 < T > ( ) * m_A * f ) ;
v = ( v + rand . Frand01 < T > ( ) * m_A * f ) ;
ClampRef < T > ( u , - 1 , 1 ) ;
ClampRef < T > ( v , - 1 , 1 ) ;
if ( ( u + v > 1 ) & & ( m_A > 0 ) )
{
if ( u > v )
{
ClampLteRef < T > ( u , 1 ) ;
v = 1 - u ;
}
else
{
ClampLteRef < T > ( v , 1 ) ;
u = 1 - v ;
}
}
}
// set output
helper . Out . x = m_Weight * ( ox + u * xx + v * yx ) ;
helper . Out . y = m_Weight * ( oy + u * xy + v * yy ) ;
helper . Out . z = m_Weight * helper . In . z ;
2016-01-12 23:42:12 -05:00
outPoint . m_ColorX = fmod ( std : : abs ( u + v ) , T ( 1.0 ) ) ;
2014-07-08 03:11:14 -04:00
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2014-07-08 03:11:14 -04:00
string scatterArea = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string zeroEdges = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string a = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t const real_t \n "
< < " \t \t xx = xform->m_A, xy = xform->m_B, \n "
< < " \t \t yx = xform->m_C * -1, yy = xform->m_D * -1, \n "
< < " \t \t ox = xform->m_E, oy = xform->m_F, \n "
< < " \t \t px = vIn.x - ox, py = vIn.y - oy; \n "
< < " \n "
< < " \t \t const real_t dot00 = xx * xx + xy * xy; \n "
< < " \t \t const real_t dot01 = xx * yx + xy * yy; \n "
< < " \t \t const real_t dot02 = xx * px + xy * py; \n "
< < " \t \t const real_t dot11 = yx * yx + yy * yy; \n "
< < " \t \t const real_t dot12 = yx * px + yy * py; \n "
< < " \n "
< < " \t \t const real_t denom = (dot00 * dot11 - dot01 * dot01); \n "
< < " \t \t const real_t num_u = (dot11 * dot02 - dot01 * dot12); \n "
< < " \t \t const real_t num_v = (dot00 * dot12 - dot01 * dot02); \n "
< < " \n "
< < " \t \t real_t u = num_u / denom; \n "
< < " \t \t real_t v = num_v / denom; \n "
< < " \t \t int inside = 0, f = 1; \n "
< < " \n "
< < " \t \t if (u + v > 1) \n "
< < " \t \t { \n "
< < " \t \t f = -1; \n "
< < " \n "
< < " \t \t if (u > v) \n "
< < " \t \t { \n "
< < " \t \t u = u > 1 ? 1 : u; \n "
< < " \t \t v = 1 - u; \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t v = v > 1 ? 1 : v; \n "
< < " \t \t u = 1 - v; \n "
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t else if ((u < 0) || (v < 0)) \n "
< < " \t \t { \n "
< < " \t \t u = u < 0 ? 0 : u > 1 ? 1 : u; \n "
< < " \t \t v = v < 0 ? 0 : v > 1 ? 1 : v; \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t inside = 1; \n "
< < " \t \t } \n "
< < " \n "
--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
< < " \t \t if ( " < < zeroEdges < < " != 0.0 && !inside) \n "
2014-07-08 03:11:14 -04:00
< < " \t \t { \n "
< < " \t \t u = v = 0; \n "
< < " \t \t } \n "
< < " \t \t else if (!inside) \n "
< < " \t \t { \n "
< < " \t \t u = (u + MwcNext01(mwc) * " < < a < < " * f); \n "
< < " \t \t v = (v + MwcNext01(mwc) * " < < a < < " * f); \n "
< < " \t \t u = u < -1 ? -1 : u > 1 ? 1 : u; \n "
< < " \t \t v = v < -1 ? -1 : v > 1 ? 1 : v; \n "
< < " \n "
< < " \t \t if ((u + v > 1) && ( " < < a < < " > 0)) \n "
< < " \t \t { \n "
< < " \t \t if (u > v) \n "
< < " \t \t { \n "
< < " \t \t u = u > 1 ? 1 : u; \n "
< < " \t \t v = 1 - u; \n "
< < " \t \t } \n "
< < " \t \t else \n "
< < " \t \t { \n "
< < " \t \t v = v > 1 ? 1 : v; \n "
< < " \t \t u = 1 - v; \n "
< < " \t \t } \n "
< < " \t \t } \n "
< < " \t \t } \n "
< < " \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * (ox + u * xx + v * yx); \n "
< < " \t \t vOut.y = " < < weight < < " * (oy + u * xy + v * yy); \n "
< < " \t \t vOut.z = " < < weight < < " * vIn.z; \n "
2015-02-26 16:18:55 -05:00
< < " \t \t outPoint->m_ColorX = fmod(fabs(u + v), (real_t)(1.0)); \n "
2014-07-08 03:11:14 -04:00
< < " \t } \n " ;
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_A = Clamp < T > ( m_ScatterArea , - 1 , 1 ) ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
2016-01-04 19:50:15 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_ScatterArea , prefix + " dc_triangle_scatter_area " , 0 , eParamType : : REAL , - 1 , 1 ) ) ; //Params.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_ZeroEdges , prefix + " dc_triangle_zero_edges " , 0 , eParamType : : INTEGER , 0 , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_A , prefix + " dc_triangle_a " ) ) ; //Precalc.
2014-07-08 03:11:14 -04:00
}
private :
T m_ScatterArea ; //Params.
T m_ZeroEdges ;
T m_A ; //Precalc.
} ;
/// <summary>
/// DC Transl.
/// The original used dc_ztransl and post_dcztransl incompatible with Ember's design.
/// These will follow the same naming convention as all other variations.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCZTranslVariation : public ParametricVariation < T >
2014-07-08 03:11:14 -04:00
{
public :
2016-01-04 19:50:15 -05:00
DCZTranslVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_ztransl " , eVariationId : : VAR_DC_ZTRANSL , weight )
2014-07-08 03:11:14 -04:00
{
Init ( ) ;
}
PARVARCOPY ( DCZTranslVariation )
2014-09-01 00:25:15 -04:00
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
2014-07-08 03:11:14 -04:00
{
T zf = m_Factor * ( outPoint . m_ColorX - m_X0_ ) / m_X1_m_x0 ;
if ( m_Clamp ! = 0 )
ClampRef < T > ( zf , 0 , 1 ) ;
helper . Out . x = m_Weight * helper . In . x ;
helper . Out . y = m_Weight * helper . In . y ;
if ( m_Overwrite = = 0 )
helper . Out . z = m_Weight * helper . In . z * zf ;
else
helper . Out . z = m_Weight * zf ;
}
2014-09-01 00:25:15 -04:00
2015-06-29 23:13:53 -04:00
virtual string OpenCLString ( ) const override
2014-07-08 03:11:14 -04:00
{
ostringstream ss , ss2 ;
2014-10-14 11:53:15 -04:00
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
2014-07-08 03:11:14 -04:00
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
string x0 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Params.
string x1 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string factor = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2014-07-08 03:11:14 -04:00
string overwrite = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ; //Precalc.
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string clamp = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string x0_ = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string x1_ = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string x1_m_x0 = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2014-07-08 03:11:14 -04:00
ss < < " \t { \n "
< < " \t \t real_t zf = " < < factor < < " * (outPoint->m_ColorX - " < < x0_ < < " ) / " < < x1_m_x0 < < " ; \n "
< < " \n "
< < " \t \t if ( " < < clamp < < " != 0) \n "
< < " \t \t zf = zf < 0 ? 0 : zf > 1 ? 1 : zf; \n "
< < " \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vIn.x; \n "
< < " \t \t vOut.y = " < < weight < < " * vIn.y; \n "
2014-07-08 03:11:14 -04:00
< < " \n "
< < " \t \t if ( " < < overwrite < < " == 0) \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.z = " < < weight < < " * vIn.z * zf; \n "
2014-07-08 03:11:14 -04:00
< < " \t \t else \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.z = " < < weight < < " * zf; \n "
2014-07-08 03:11:14 -04:00
< < " \t } \n " ;
return ss . str ( ) ;
}
2014-09-01 00:25:15 -04:00
virtual void Precalc ( ) override
2014-07-08 03:11:14 -04:00
{
m_X0_ = m_X0 < m_X1 ? m_X0 : m_X1 ;
m_X1_ = m_X0 > m_X1 ? m_X0 : m_X1 ;
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_X1_m_x0 = Zeps ( m_X1_ - m_X0_ ) ;
2014-07-08 03:11:14 -04:00
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_X0 , prefix + " dc_ztransl_x0 " , 0 , eParamType : : REAL , 0 , 1 ) ) ; //Params.
m_Params . push_back ( ParamWithName < T > ( & m_X1 , prefix + " dc_ztransl_x1 " , 1 , eParamType : : REAL , 0 , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Factor , prefix + " dc_ztransl_factor " , 1 ) ) ;
2016-01-04 19:50:15 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Overwrite , prefix + " dc_ztransl_overwrite " , 1 , eParamType : : INTEGER , 0 , 1 ) ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Clamp , prefix + " dc_ztransl_clamp " , 0 , eParamType : : INTEGER , 0 , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_X0_ , prefix + " dc_ztransl_x0_ " ) ) ; //Precalc.
m_Params . push_back ( ParamWithName < T > ( true , & m_X1_ , prefix + " dc_ztransl_x1_ " ) ) ;
2014-07-08 03:11:14 -04:00
m_Params . push_back ( ParamWithName < T > ( true , & m_X1_m_x0 , prefix + " dc_ztransl_x1_m_x0 " ) ) ;
}
private :
T m_X0 ; //Params.
T m_X1 ;
T m_Factor ;
T m_Overwrite ;
T m_Clamp ;
T m_X0_ ; //Precalc.
T m_X1_ ;
T m_X1_m_x0 ;
} ;
2015-12-31 16:41:59 -05:00
# define SHAPE_SQUARE 0
# define SHAPE_DISC 1
# define SHAPE_BLUR 2
# define MAP_FLAT 0
# define MAP_SPHERICAL 1
# define MAP_HSPHERE 2
# define MAP_QSPHERE 3
# define MAP_BUBBLE 4
# define MAP_BUBBLE2 5
/// <summary>
/// dc_perlin.
/// </summary>
template < typename T >
2016-02-18 21:58:24 -05:00
class DCPerlinVariation : public ParametricVariation < T >
2015-12-31 16:41:59 -05:00
{
public :
2016-01-04 19:50:15 -05:00
DCPerlinVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " dc_perlin " , eVariationId : : VAR_DC_PERLIN , weight )
2015-12-31 16:41:59 -05:00
{
Init ( ) ;
}
PARVARCOPY ( DCPerlinVariation )
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
{
v3T v ;
T vx , vy , col , r , theta , s , c , p , e ;
int t = 0 , iShape = int ( m_Shape ) , iMap = int ( m_Map ) , iOctaves = int ( m_Octaves ) , iBailout = int ( m_SelectBailout ) ;
do
{
// Default edge value
e = 0 ;
// Assign vx, vy according to shape
switch ( iShape )
{
case SHAPE_SQUARE :
vx = ( 1 + m_Edge ) * ( rand . Frand01 < T > ( ) - T ( 0.5 ) ) ;
vy = ( 1 + m_Edge ) * ( rand . Frand01 < T > ( ) - T ( 0.5 ) ) ;
r = SQR ( vx ) > SQR ( vy ) ? std : : sqrt ( SQR ( vx ) ) : std : : sqrt ( SQR ( vy ) ) ;
if ( r > 1 - m_Edge )
e = T ( 0.5 ) * ( r - 1 + m_Edge ) / m_Edge ;
break ;
case SHAPE_DISC :
r = rand . Frand01 < T > ( ) + rand . Frand01 < T > ( ) ;
r = ( r > 1 ) ? 2 - r : r ;
r * = ( 1 + m_Edge ) ;
if ( r > 1 - m_Edge )
e = T ( 0.5 ) * ( r - 1 + m_Edge ) / m_Edge ;
theta = rand . Frand01 < T > ( ) * M_2PI ;
sincos ( theta , & s , & c ) ;
vx = T ( 0.5 ) * r * s ;
vy = T ( 0.5 ) * r * c ;
break ;
case SHAPE_BLUR :
2016-01-04 19:50:15 -05:00
default :
2015-12-31 16:41:59 -05:00
r = ( 1 + m_Edge ) * rand . Frand01 < T > ( ) ;
if ( r > 1 - m_Edge )
e = T ( 0.5 ) * ( r - 1 + m_Edge ) / m_Edge ;
theta = rand . Frand01 < T > ( ) * M_2PI ;
sincos ( theta , & s , & c ) ;
vx = T ( 0.5 ) * r * s ;
vy = T ( 0.5 ) * r * c ;
break ;
}
// Assign V for noise vector position according to map
switch ( iMap )
{
case MAP_FLAT :
v . x = m_Scale * vx ;
v . y = m_Scale * vy ;
v . z = m_Scale * m_Z ;
break ;
case MAP_SPHERICAL :
r = 1 / Zeps < T > ( SQR ( vx ) + SQR ( vy ) ) ;
v . x = m_Scale * vx * r ;
v . y = m_Scale * vy * r ;
v . z = m_Scale * m_Z ;
break ;
case MAP_HSPHERE :
r = 1 / ( SQR ( vx ) + SQR ( vy ) + T ( 0.5 ) ) ;
v . x = m_Scale * vx * r ;
v . y = m_Scale * vy * r ;
v . z = m_Scale * m_Z ;
break ;
case MAP_QSPHERE :
r = 1 / ( SQR ( vx ) + SQR ( vy ) + T ( 0.25 ) ) ;
v . x = m_Scale * vx * r ;
v . y = m_Scale * vy * r ;
v . z = m_Scale * m_Z ;
break ;
case MAP_BUBBLE :
r = T ( 0.25 ) - ( SQR ( vx ) + SQR ( vy ) ) ;
if ( r < 0 )
r = std : : sqrt ( - r ) ;
else
r = std : : sqrt ( r ) ;
v . x = m_Scale * vx ;
v . y = m_Scale * vy ;
v . z = m_Scale * ( r + m_Z ) ;
break ;
case MAP_BUBBLE2 :
2016-01-04 19:50:15 -05:00
default :
2015-12-31 16:41:59 -05:00
r = T ( 0.25 ) - ( SQR ( vx ) + SQR ( vy ) ) ;
if ( r < 0 )
r = std : : sqrt ( - r ) ;
else
r = std : : sqrt ( r ) ;
v . x = m_Scale * vx ;
v . y = m_Scale * vy ;
v . z = m_Scale * ( 2 * r + m_Z ) ;
break ;
}
p = m_VarFuncs - > PerlinNoise3D ( v , m_Amps , m_Freqs , iOctaves ) ;
// Add edge effects
if ( p > 0 )
e = p * ( 1 + e * e * 20 ) + 2 * e ;
else
e = p * ( 1 + e * e * 20 ) - 2 * e ;
}
while ( ( e < m_NotchBottom | | e > m_NotchTop ) & & t + + < iBailout ) ;
// Add blur effect to transform
helper . Out . x = m_Weight * vx ;
helper . Out . y = m_Weight * vy ;
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
helper . Out . z = DefaultZ ( helper ) ;
2015-12-31 16:41:59 -05:00
col = m_Centre + m_Range * p ;
outPoint . m_ColorX = col - Floor < T > ( col ) ;
}
virtual vector < string > OpenCLGlobalFuncNames ( ) const override
{
2016-01-17 19:10:06 -05:00
return vector < string > { " Zeps " , " Sqr " , " SimplexNoise3D " , " PerlinNoise3D " } ;
2015-12-31 16:41:59 -05:00
}
virtual vector < string > OpenCLGlobalDataNames ( ) const override
{
return vector < string > { " NOISE_INDEX " , " NOISE_POINTS " } ;
}
virtual string OpenCLString ( ) const override
{
ostringstream ss , ss2 ;
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
ss2 < < " _ " < < XformIndexInEmber ( ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string weight = WeightDefineString ( ) ;
2015-12-31 16:41:59 -05:00
string index = ss2 . str ( ) + " ] " ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string shape = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string map = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string selectCentre = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string selectRange = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string centre = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string range = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string edge = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string scale = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string octaves = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string amps = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string freqs = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string z = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2015-12-31 16:41:59 -05:00
string selectBailout = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
string notchBottom = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string notchTop = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2015-12-31 16:41:59 -05:00
ss < < " \t { \n "
2017-08-16 20:33:11 -04:00
< < " \t \t real4 v; \n "
2015-12-31 16:41:59 -05:00
< < " \t \t real_t vx, vy, col, r, theta, s, c, p, e; \n "
< < " \t \t int t = 0, iShape = (int) " < < shape < < " , iMap = (int) " < < map < < " , iOctaves = (int) " < < octaves < < " , iBailout = (int) " < < selectBailout < < " ; \n "
< < " \n "
< < " \t \t do \n "
< < " \t \t { \n "
< < " \t \t e = 0; \n "
< < " \n "
< < " \t \t switch (iShape) \n "
< < " \t \t { \n "
< < " \t \t case " < < SHAPE_SQUARE < < " : \n "
< < " \t \t vx = (1 + " < < edge < < " ) * (MwcNext01(mwc) - 0.5); \n "
< < " \t \t vy = (1 + " < < edge < < " ) * (MwcNext01(mwc) - 0.5); \n "
< < " \t \t r = SQR(vx) > SQR(vy) ? sqrt(SQR(vx)) : sqrt(SQR(vy)); \n "
< < " \n "
< < " \t \t if (r > 1 - " < < edge < < " ) \n "
< < " \t \t e = 0.5 * (r - 1 + " < < edge < < " ) / " < < edge < < " ; \n "
< < " \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < SHAPE_DISC < < " : \n "
< < " \t \t r = MwcNext01(mwc) + MwcNext01(mwc); \n "
< < " \t \t r = (r > 1) ? 2 - r : r; \n "
< < " \t \t r *= (1 + " < < edge < < " ); \n "
< < " \n "
< < " \t \t if (r > 1 - " < < edge < < " ) \n "
< < " \t \t e = 0.5 * (r - 1 + " < < edge < < " ) / " < < edge < < " ; \n "
< < " \n "
< < " \t \t theta = MwcNext01(mwc) * M_2PI; \n "
< < " \t \t s = sincos(theta, &c); \n "
< < " \t \t vx = 0.5 * r * s; \n "
< < " \t \t vy = 0.5 * r * c; \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < SHAPE_BLUR < < " : \n "
2016-04-13 23:59:57 -04:00
< < " \t \t default: \n "
2015-12-31 16:41:59 -05:00
< < " \t \t r = (1 + " < < edge < < " ) * MwcNext01(mwc); \n "
< < " \n "
< < " \t \t if (r > 1 - " < < edge < < " ) \n "
< < " \t \t e = 0.5 * (r - 1 + " < < edge < < " ) / " < < edge < < " ; \n "
< < " \n "
< < " \t \t theta = MwcNext01(mwc) * M_2PI; \n "
< < " \t \t s = sincos(theta, &c); \n "
< < " \t \t vx = 0.5 * r * s; \n "
< < " \t \t vy = 0.5 * r * c; \n "
< < " \t \t break; \n "
< < " \t \t } \n "
< < " \n "
< < " \t \t switch (iMap) \n "
< < " \t \t { \n "
< < " \t \t case " < < MAP_FLAT < < " : \n "
< < " \t \t v.x = " < < scale < < " * vx; \n "
< < " \t \t v.y = " < < scale < < " * vy; \n "
< < " \t \t v.z = " < < scale < < " * " < < z < < " ; \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < MAP_SPHERICAL < < " : \n "
< < " \t \t r = 1 / Zeps(SQR(vx) + SQR(vy)); \n "
< < " \t \t v.x = " < < scale < < " * vx * r; \n "
< < " \t \t v.y = " < < scale < < " * vy * r; \n "
< < " \t \t v.z = " < < scale < < " * " < < z < < " ; \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < MAP_HSPHERE < < " : \n "
< < " \t \t r = 1 / (SQR(vx) + SQR(vy) + 0.5); \n "
< < " \t \t v.x = " < < scale < < " * vx * r; \n "
< < " \t \t v.y = " < < scale < < " * vy * r; \n "
< < " \t \t v.z = " < < scale < < " * " < < z < < " ; \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < MAP_QSPHERE < < " : \n "
< < " \t \t r = 1 / (SQR(vx) + SQR(vy) + 0.25); \n "
< < " \t \t v.x = " < < scale < < " * vx * r; \n "
< < " \t \t v.y = " < < scale < < " * vy * r; \n "
< < " \t \t v.z = " < < scale < < " * " < < z < < " ; \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < MAP_BUBBLE < < " : \n "
< < " \t \t r = 0.25 - (SQR(vx) + SQR(vy)); \n "
< < " \n "
< < " \t \t if (r < 0) \n "
< < " \t \t r = sqrt(-r); \n "
< < " \t \t else \n "
< < " \t \t r = sqrt(r); \n "
< < " \n "
< < " \t \t v.x = " < < scale < < " * vx; \n "
< < " \t \t v.y = " < < scale < < " * vy; \n "
< < " \t \t v.z = " < < scale < < " * (r + " < < z < < " ); \n "
< < " \t \t break; \n "
< < " \n "
< < " \t \t case " < < MAP_BUBBLE2 < < " : \n "
2016-04-13 23:59:57 -04:00
< < " \t \t default: \n "
2015-12-31 16:41:59 -05:00
< < " \t \t r = 0.25 - (SQR(vx) + SQR(vy)); \n "
< < " \n "
< < " \t \t if (r < 0) \n "
< < " \t \t r = sqrt(-r); \n "
< < " \t \t else \n "
< < " \t \t r = sqrt(r); \n "
< < " \n "
< < " \t \t v.x = " < < scale < < " * vx; \n "
< < " \t \t v.y = " < < scale < < " * vy; \n "
< < " \t \t v.z = " < < scale < < " * (2 * r + " < < z < < " ); \n "
< < " \t \t break; \n "
< < " \t \t } \n "
< < " \n "
2016-04-13 23:59:57 -04:00
< < " \t \t p = PerlinNoise3D(&v, globalShared + NOISE_INDEX, globalShared + NOISE_POINTS, " < < amps < < " , " < < freqs < < " , iOctaves); \n "
2015-12-31 16:41:59 -05:00
< < " \n "
< < " \t \t if (p > 0) \n "
< < " \t \t e = p * (1 + e * e * 20) + 2 * e; \n "
< < " \t \t else \n "
< < " \t \t e = p * (1 + e * e * 20) - 2 * e; \n "
2016-04-13 23:59:57 -04:00
< < " \t \t } \n "
2015-12-31 16:41:59 -05:00
< < " \t \t while ((e < " < < notchBottom < < " || e > " < < notchTop < < " ) && t++ < iBailout); \n "
< < " \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t vOut.x = " < < weight < < " * vx; \n "
< < " \t \t vOut.y = " < < weight < < " * vy; \n "
--User changes
-Allow for pausing the renderer in the main window. This makes is more efficient when entering many parameters, such as when following a tutorial.
-Add support for new variations: erf, gamma, jac_cn, jac_dn, jac_sn, logDB, pressure_wave, pRose3D, splits3D, w, waves2b, x, xerf, y, z.
-Inform user of the start and stop of file parsing in EmberAnimate because the files could potentially be very large.
-Move the follwing fields to a new table called Animation: Interpolation, Affine Interpolation, Temporal Samples, Temporal Filter Width, Temporal Filter Type.
-These currently have no effect on the interactive renderer and instead are used when running flames through EmberGenome to generate sequences, and then animating them in Fractorium or EmberAnimate.
-Add new parameter overrides for EmberRender and EmberAnimate which directly assign values to all flames being rendered, rather than scale:
--quality
--demin
--demax
--Bug fixes
-Left pad instead of right pad names of sequence outputs from EmberGenome.
-Unique file naming was broken for files which already had an underscore in them.
-Properly report that png is the default format of EmberRender and EmberAnimate output instead of erroneously claiming it was jpg.
-Make command line programs search these folders in this order for the palette file:
./
~/.fractorium
~/.config/fractorium
/usr/share/fractorium
/usr/local/share/fractorium
-Fix possible bad values in hexes.
-Better assignment of Z variables.
-Fix boarders due to use of poorly implemented rint() function from flam3. Use std::rint() now.
-wedge_sph was completely wrong due to having accidentally swapped the mapping of two parameters.
-Make juliascope juliascope_power parameter be of type REAL_NONZERO since it's used as a denominator.
-Make Z assignment compatible with the originals in:
-arch, bcircle, bCollide, bent, bent2, bisplit, blob, blur_linear, blur_square, bMod, boarders, boarders2, bSwirl, bTransform, butterfly, cardioid, cell, circleblur, circlize, circlize2, circus, collideoscope, cos, cosine, cosh, coth, cpow, cpow2, crescents, cropn, csc, csch, curl, curve, dc_gridout, deltaa, diamond, disc2, eclipse, eCollide, edisc, eJulia, elliptic, eMod, eMotion, ennepers, epispiral, ePush, eRotate, eScale, eSwirl, ex, exp, expo, exponential, fan, fdisc, fibonacci, fibonacci2, fisheye, flipcircle, flipy, flower, flux, funnel, glynnia, GlynnSim1, GlynnSim2, GlynnSim3, gridout, handkerchief, heart, hole, idisc, julia, julian2, juliaNab, kaleidoscope, lazyTravis, Lissajous, mask, MobiusN, mobius_strip, modulus, murl, murl2, npolar, ortho, oscilloscope, parabola, perspective, petal, phoenix_julia, pie (was also inconsistent between cpu and gpu), poincare, popcorn, popcorn2, power, pow_block, rational3, rays, rblur, rings, rippled, roundspher, sec, secant2, sigmoid, sin, sineblur, sinh, sinusgrid, sphericaln, spiralwing, spirograph, split, squarize, squirrel, squish, sschecks, starblur, stripes, stwin, super_shape, tan, tancos, tangent, tanh, TwinTrian, twoface, unpolar, waves, wavesn, wedge_julia, whorl, xheart, zblur, zscale.
--Code changes
-Generalize Variation::PrecalcHelper() and rename to PrePostPrecalcHelper().
--Do the same for the OpenCL version and rename it PrePostPrecalcOpenCLString().
-Rename Variation::m_AssignType to m_PrePostAssignType since it's only relevant to pre/post variations.
2016-01-29 20:02:15 -05:00
< < " \t \t vOut.z = " < < DefaultZCl ( )
2015-12-31 16:41:59 -05:00
< < " \t \t col = " < < centre < < " + " < < range < < " * p; \n "
< < " \t \t outPoint->m_ColorX = col - floor(col); \n "
< < " \t } \n " ;
return ss . str ( ) ;
}
virtual void Precalc ( ) override
{
m_NotchBottom = m_SelectCentre - m_SelectRange ;
m_NotchBottom = ( m_NotchBottom > T ( 0.75 ) ) ? T ( 0.75 ) : m_NotchBottom ;
m_NotchBottom = ( m_NotchBottom < - 2 ) ? - 3 : m_NotchBottom ;
m_NotchTop = m_SelectCentre + m_SelectRange ;
m_NotchTop = ( m_NotchTop < T ( - 0.75 ) ) ? T ( - 0.75 ) : m_NotchTop ;
m_NotchTop = ( m_NotchTop > 3 ) ? 3 : m_NotchTop ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . reserve ( 15 ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_Shape , prefix + " dc_perlin_shape " , 0 , eParamType : : INTEGER , 0 , 2 ) ) ; //Params.
m_Params . push_back ( ParamWithName < T > ( & m_Map , prefix + " dc_perlin_map " , 0 , eParamType : : INTEGER , 0 , 5 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_SelectCentre , prefix + " dc_perlin_select_centre " , 0 , eParamType : : REAL , - 1 , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_SelectRange , prefix + " dc_perlin_select_range " , 1 , eParamType : : REAL , T ( 0.1 ) , 2 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Centre , prefix + " dc_perlin_centre " , T ( 0.25 ) ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Range , prefix + " dc_perlin_range " , T ( 0.25 ) ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Edge , prefix + " dc_perlin_edge " ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Scale , prefix + " dc_perlin_scale " , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Octaves , prefix + " dc_perlin_octaves " , 2 , eParamType : : INTEGER , 1 , 5 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Amps , prefix + " dc_perlin_amps " , 2 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Freqs , prefix + " dc_perlin_freqs " , 2 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Z , prefix + " dc_perlin_z " ) ) ;
2016-01-04 19:50:15 -05:00
m_Params . push_back ( ParamWithName < T > ( & m_SelectBailout , prefix + " dc_perlin_select_bailout " , 10 , eParamType : : INTEGER , 2 , 1000 ) ) ;
2015-12-31 16:41:59 -05:00
m_Params . push_back ( ParamWithName < T > ( true , & m_NotchBottom , prefix + " dc_perlin_notch_bottom " ) ) ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
m_Params . push_back ( ParamWithName < T > ( true , & m_NotchTop , prefix + " dc_perlin_notch_top " ) ) ;
2015-12-31 16:41:59 -05:00
}
private :
T m_Shape ; //Params.
T m_Map ;
T m_SelectCentre ;
T m_SelectRange ;
T m_Centre ;
T m_Range ;
T m_Edge ;
T m_Scale ;
T m_Octaves ;
T m_Amps ;
T m_Freqs ;
T m_Z ;
T m_SelectBailout ;
T m_NotchBottom ; //Precalc.
T m_NotchTop ;
2016-04-13 23:59:57 -04:00
shared_ptr < VarFuncs < T > > m_VarFuncs = VarFuncs < T > : : Instance ( ) ;
2015-12-31 16:41:59 -05:00
} ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
/// <summary>
/// randCubes.
/// </summary>
template < typename T >
class RandCubesVariation : public ParametricVariation < T >
{
public :
RandCubesVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " randCubes " , eVariationId : : VAR_RAND_CUBES , weight )
{
Init ( ) ;
}
PARVARCOPY ( RandCubesVariation )
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
{
auto blockx = Floor < T > ( std : : log ( rand . Frand01 < T > ( ) ) * ( rand . RandBit ( ) ? m_Spread : - m_Spread ) ) ;
auto blocky = Floor < T > ( std : : log ( rand . Frand01 < T > ( ) ) * ( rand . RandBit ( ) ? m_Spread : - m_Spread ) ) ;
T z = VarFuncs < T > : : Hash ( int ( blockx * m_Seed ) ) + VarFuncs < T > : : Hash ( int ( blockx + blocky * m_Seed ) ) ; //random height for each block
if ( m_VarType = = eVariationType : : VARTYPE_REG )
{
outPoint . m_X = 0 ; //This variation intentionally assigns instead of summing.
outPoint . m_Y = 0 ;
outPoint . m_Z = 0 ;
}
helper . Out . x = ( blockx * m_Density + rand . Frand01 < T > ( ) ) * m_BlockSize ;
helper . Out . y = ( blocky * m_Density + rand . Frand01 < T > ( ) ) * m_BlockSize ;
helper . Out . z = m_BlockHeight * z * std : : pow ( rand . Frand01 < T > ( ) , T ( 0.125 ) ) ; //fade out down
outPoint . m_ColorX = z / 2 ; //block height -> palette location
}
virtual string OpenCLString ( ) const override
{
ostringstream ss , ss2 ;
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
string weight = WeightDefineString ( ) ;
string blocksize = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string blockheight = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string spread = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string seed = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string density = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
ss < < " \t { \n "
< < " \t \t real_t blockx = floor(log(MwcNext01(mwc)) * ((MwcNext(mwc) & 1) ? " < < spread < < " : - " < < spread < < " )); \n "
< < " \t \t real_t blocky = floor(log(MwcNext01(mwc)) * ((MwcNext(mwc) & 1) ? " < < spread < < " : - " < < spread < < " )); \n "
2018-09-15 06:11:12 -04:00
< < " \t \t real_t z = Hash(blockx * " < < seed < < " ) + Hash(fma(blocky, " < < seed < < " , blockx)); \n " ;
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
if ( m_VarType = = eVariationType : : VARTYPE_REG )
{
ss < < " \t \t outPoint->m_X = 0; \n "
< < " \t \t outPoint->m_Y = 0; \n "
< < " \t \t outPoint->m_Z = 0; \n " ;
}
2018-09-15 06:11:12 -04:00
ss < < " \t \t vOut.x = fma(blockx, " < < density < < " , MwcNext01(mwc)) * " < < blocksize < < " ; \n "
< < " \t \t vOut.y = fma(blocky, " < < density < < " , MwcNext01(mwc)) * " < < blocksize < < " ; \n "
2019-05-23 22:44:27 -04:00
< < " \t \t vOut.z = " < < blockheight < < " * z * pow(MwcNext01(mwc), (real_t)(0.125)); \n "
--User changes
-Support 4k monitors, and in general, properly scale any monitor that is not HD.
-Allow for a spatial filter of radius zero, which means do not use a spatial filter.
-Add new variations: concentric, cpow3, helicoid, helix, rand_cubes, sphereblur.
-Use a new method for computing elliptic which is more precise. Developed by Discord user Claude.
-Remove the 8 variation per xform limitation on the GPU.
-Allow for loading the last flame file on startup, rather than randoms.
-Use two different default quality values in the interactive renderer, one each for CPU and GPU.
-Creating linked xforms was using non-standard behavior. Make it match Apo and also support creating multiple linked xforms at once.
--Bug fixes
-No variations in an xform used to have the same behavior as a single linear variation with weight 1. While sensible, this breaks backward compatibility. No variations now sets the output point to zeroes.
-Prevent crashing the program when adjusting a value on the main window while a final render is in progress.
-The xaos table was inverted.
--Code changes
-Convert projects to Visual Studio 2017.
-Change bad vals from +- 1e10 to +-1e20.
-Reintroduce the symmetry tag in xforms for legacy support in programs that do not use color_speed.
-Compiler will not let us use default values in templated member functions anymore.
2017-11-26 20:27:00 -05:00
< < " \t \t outPoint->m_ColorX = z / 2; \n "
< < " \t } \n " ;
return ss . str ( ) ;
}
virtual vector < string > OpenCLGlobalFuncNames ( ) const override
{
return vector < string > { " Hash " } ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_BlockSize , prefix + " randCubes_blockSize " , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_BlockHeight , prefix + " randCubes_blockHeight " , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Spread , prefix + " randCubes_spread " , 3 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Seed , prefix + " randCubes_seed " , 1 , eParamType : : INTEGER ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Density , prefix + " randCubes_density " , 1 ) ) ;
}
private :
T m_BlockSize ;
T m_BlockHeight ;
T m_Spread ;
T m_Seed ;
T m_Density ;
} ;
2018-02-04 14:44:22 -05:00
/// <summary>
/// pixel_flow.
/// </summary>
template < typename T >
class PixelFlowVariation : public ParametricVariation < T >
{
public :
PixelFlowVariation ( T weight = 1.0 ) : ParametricVariation < T > ( " pixel_flow " , eVariationId : : VAR_PIXEL_FLOW , weight )
{
Init ( ) ;
}
PARVARCOPY ( PixelFlowVariation )
virtual void Func ( IteratorHelper < T > & helper , Point < T > & outPoint , QTIsaac < ISAAC_SIZE , ISAAC_INT > & rand ) override
{
T xl , yl ;
2021-01-03 08:32:04 -05:00
sincos ( m_Rad , & yl , & xl ) ;
2018-02-04 14:44:22 -05:00
auto blockx = ( int ) Floor ( helper . In . x * m_Width ) ; //Calculate which block we're in.
blockx + = int ( 2 - 4 * VarFuncs < T > : : Hash ( int ( blockx * m_Seed + 1 ) ) ) ; //Varying width and length.
auto blocky = ( int ) Floor ( helper . In . y * m_Width ) ;
blocky + = int ( 2 - 4 * VarFuncs < T > : : Hash ( int ( blocky * m_Seed + 1 ) ) ) ;
T fLen = ( VarFuncs < T > : : Hash ( int ( blocky + blockx * - m_Seed ) ) + VarFuncs < T > : : Hash ( int ( blockx + blocky * m_Seed * T ( 0.5 ) ) ) ) * T ( 0.5 ) ; //doesnt matter just needs to be random enough
T r01 = rand . Frand01 < T > ( ) ;
T fade = fLen * r01 * r01 * r01 * r01 ; //Fading effect.
helper . Out . x = m_Weight * m_Len * xl * fade ;
helper . Out . y = m_Weight * m_Len * yl * fade ;
helper . Out . z = DefaultZ ( helper ) ;
if ( m_EnableDC )
outPoint . m_ColorX = r01 ; //Direct color.
}
virtual string OpenCLString ( ) const override
{
ostringstream ss , ss2 ;
intmax_t i = 0 , varIndex = IndexInXform ( ) ;
ss2 < < " _ " < < XformIndexInEmber ( ) < < " ] " ;
string index = ss2 . str ( ) ;
string weight = WeightDefineString ( ) ;
string angle = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2018-09-15 06:11:12 -04:00
string len = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2018-02-04 14:44:22 -05:00
string width = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2018-09-15 06:11:12 -04:00
string seed = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string dc = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
string rad = " parVars[ " + ToUpper ( m_Params [ i + + ] . Name ( ) ) + index ;
2018-02-04 14:44:22 -05:00
ss < < " \t { \n "
< < " \t \t real_t xl, yl; \n "
2021-01-03 08:32:04 -05:00
< < " \t \t yl = sincos( " < < rad < < " , &xl); \n "
2018-02-04 14:44:22 -05:00
< < " \t \t int blockx = (int)floor(vIn.x * " < < width < < " ); \n "
2018-09-15 06:11:12 -04:00
< < " \t \t blockx += (int)(2 - 4 * Hash((int)(blockx * (int) " < < seed < < " + 1))); \n "
2018-02-04 14:44:22 -05:00
< < " \t \t int blocky = (int)floor(vIn.y * " < < width < < " ); \n "
2018-09-15 06:11:12 -04:00
< < " \t \t blocky += (int)(2 - 4 * Hash((int)(blocky * (int) " < < seed < < " + 1))); \n "
< < " \t \t real_t fLen = (Hash((int)(blocky + blockx * -(int) " < < seed < < " )) + Hash((int)(blockx + blocky * (int) " < < seed < < " * (real_t)0.5))) * (real_t)0.5; \n "
2018-02-04 14:44:22 -05:00
< < " \t \t real_t r01 = MwcNext01(mwc); \n "
< < " \t \t real_t fade = fLen * r01 * r01 * r01 * r01; \n "
< < " \t \t vOut.x = " < < weight < < " * " < < len < < " * xl * fade; \n "
< < " \t \t vOut.y = " < < weight < < " * " < < len < < " * yl * fade; \n "
< < " \t \t vOut.z = " < < DefaultZCl ( )
< < " \t \t if ( " < < dc < < " ) \n "
< < " \t \t \t outPoint->m_ColorX = r01; \n "
< < " \t } \n " ;
return ss . str ( ) ;
}
virtual void Precalc ( ) override
{
m_Rad = m_Angle * DEG_2_RAD_T ;
}
virtual vector < string > OpenCLGlobalFuncNames ( ) const override
{
return vector < string > { " Hash " } ;
}
protected :
void Init ( )
{
string prefix = Prefix ( ) ;
m_Params . clear ( ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Angle , prefix + " pixel_flow_angle " , 90 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Len , prefix + " pixel_flow_len " , T ( 0.1 ) ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Width , prefix + " pixel_flow_width " , 200 ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_Seed , prefix + " pixel_flow_seed " , 42 , eParamType : : INTEGER ) ) ;
m_Params . push_back ( ParamWithName < T > ( & m_EnableDC , prefix + " pixel_flow_enable_dc " , 0 , eParamType : : INTEGER , 0 , 1 ) ) ;
m_Params . push_back ( ParamWithName < T > ( true , & m_Rad , prefix + " pixel_flow_rad " ) ) ;
}
private :
T m_Angle ;
T m_Len ;
T m_Width ;
T m_Seed ;
T m_EnableDC ;
T m_Rad ; //Precalc.
} ;
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
MAKEPREPOSTPARVAR ( DCBubble , dc_bubble , DC_BUBBLE )
2014-07-08 03:11:14 -04:00
MAKEPREPOSTPARVAR ( DCCarpet , dc_carpet , DC_CARPET )
2016-01-04 19:50:15 -05:00
MAKEPREPOSTPARVARASSIGN ( DCCube , dc_cube , DC_CUBE , eVariationAssignType : : ASSIGNTYPE_SUM )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
MAKEPREPOSTPARVAR ( DCCylinder , dc_cylinder , DC_CYLINDER )
2014-07-08 03:11:14 -04:00
MAKEPREPOSTVAR ( DCGridOut , dc_gridout , DC_GRIDOUT )
--User changes
-Add new variations: bubbleT3D, crob, hexaplay3D, hexcrop, hexes, hexnix3D, loonie2, loonie3, nBlur, octapol and synth.
-Allow for pre/post versions of dc_bubble, dc_cylinder and dc_linear whereas before they were omitted.
-When saving a file with multiple embers in it, detect if time values are all the same and if so, start them at zero and increment by 1 for each ember.
-Allow for numerous quality increases to be coalesced into one. It will pick up at the end of the current render.
-Show selection highlight on variations tree in response to mouse hover. This makes it easier to see for which variation or param the current mouse wheel action will apply.
-Make default temporal samples be 100, whereas before it was 1000 which was overkill.
-Require the shift key to be held with delete for deleting an ember to prevent it from triggering when the user enters delete in the edit box.
-This wasn't otherwise fixable without writing a lot more code.
--Bug fixes
-EmberGenome was crashing when generating a sequence from a source file with more than 2 embers in it.
-EmberGenome was improperly handling the first frame of a merge after the last frame of the loop.
-These bugs were due to a previous commit. Revert parts of that commit.
-Prevent a zoom value of less than 0 when reading from xml.
-Slight optimization of the crescents, and mask variations, if the compiler wasn't doing it already.
-Unique file naming was broken because it was looking for _# and the default names ended with -#.
-Disallow renaming of an ember in the library tree to an empty string.
-Severe bug that prevented some variations from being read correctly from params generated outside this program.
-Severe OpenCL randomization bug. The first x coordinates of the first points in the first kernel call of the first ember of a render since the OpenCL renderer object was created were not random and were mostly -1.
-Severe bug when populating xform selection distributions that could sometimes cause a crash due to roundoff error. Fix by using double.
-Limit the max number of variations in a random ember to MAX_CL_VARS, which is 8. This ensures they'll look the same on CPU and GPU.
-Prevent user from saving stylesheet to default.qss, it's a special reserved filename.
--Code changes
-Generalize using the running sum output point inside of a variation for all cases: pre, reg and post.
-Allow for array variables in variations where the address of each element is stored in m_Params.
-Qualify all math functions with std::
-No longer use our own Clamp() in OpenCL, instead use the standard clamp().
-Redesign how functions are used in the variations OpenCL code.
-Add tests to EmberTester to verify some of the new functionality.
-Place more const and override qualifiers on functions where appropriate.
-Add a global rand with a lock to be used very sparingly.
-Use a map instead of a vector for bad param names in Xml parsing.
-Prefix affine interpolation mode defines with "AFFINE_" to make their purpose more clear.
-Allow for variations that change state during iteration by sending a separate copy of the ember to each rendering thread.
-Implement this same functionality with a local struct in OpenCL. It's members are the total of all variables that need to change state within an ember.
-Add Contains() function to Utils.h.
-EmberRender: print names of kernels being printed with --dump_kernel option.
-Clean up EmberTester to handle some of the recent changes.
-Fix various casts.
-Replace % 2 with & 1, even though the compiler was likely doing this already.
-Add new file Variations06.h to accommodate new variations.
-General cleanup.
2015-11-22 17:15:07 -05:00
MAKEPREPOSTPARVAR ( DCLinear , dc_linear , DC_LINEAR )
2014-07-08 03:11:14 -04:00
MAKEPREPOSTPARVAR ( DCTriangle , dc_triangle , DC_TRIANGLE )
MAKEPREPOSTPARVAR ( DCZTransl , dc_ztransl , DC_ZTRANSL )
2015-12-31 16:41:59 -05:00
MAKEPREPOSTPARVAR ( DCPerlin , dc_perlin , DC_PERLIN )
2018-09-18 23:49:38 -04:00
MAKEPREPOSTPARVARASSIGN ( RandCubes , randCubes , RAND_CUBES , eVariationAssignType : : ASSIGNTYPE_SUM )
MAKEPREPOSTPARVAR ( PixelFlow , pixel_flow , PIXEL_FLOW )
2014-07-08 03:11:14 -04:00
}
