mirror-github-bspeice/fractorium
1
0
Fork 0
mirror of https://bitbucket.org/mfeemster/fractorium.git synced 2025-06-30 13:26:02 -04:00
Code Issues Actions Packages Projects Releases Wiki Activity

Change CHOOSE_XFORM_GRAIN to be 16384, so we can & with 16383 instead of using modulo. This results in a 9% speed increase on the CPU.

Browse Source
This commit is contained in:
mfeemster
2014-12-09 20:15:02 -08:00
parent d42cdfafe4
commit 7abecaf7be
2 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
Source/EmberCL/IterOpenCLKernelCreator.cpp
View File

@ -301,7 +301,7 @@ string IterOpenCLKernelCreator<T>::CreateIterKernelString(Ember<T>& ember, strin
os <<
#ifndef STRAIGHT_RAND
" if (THREAD_ID_Y == 0 && THREAD_ID_X < NWARPS)\n"
" xfsel[THREAD_ID_X] = MwcNext(&mwc) % " << CHOOSE_XFORM_GRAIN << ";\n"//It's faster to do the % here ahead of time than every time an xform is looked up to use inside the loop.
" xfsel[THREAD_ID_X] = MwcNext(&mwc) & " << CHOOSE_XFORM_GRAIN_M1 << ";\n"//It's faster to do the & here ahead of time than every time an xform is looked up to use inside the loop.
"\n"
#endif
" barrier(CLK_LOCAL_MEM_FENCE);\n"
@ -322,7 +322,7 @@ string IterOpenCLKernelCreator<T>::CreateIterKernelString(Ember<T>& ember, strin
{
os <<
#ifdef STRAIGHT_RAND
" secondPoint.m_LastXfUsed = xformDistributions[MwcNext(&mwc) % " << CHOOSE_XFORM_GRAIN << " + (" << CHOOSE_XFORM_GRAIN << " * (firstPoint.m_LastXfUsed + 1u))];\n\n";
" secondPoint.m_LastXfUsed = xformDistributions[MwcNext(&mwc) & " << CHOOSE_XFORM_GRAIN_M1 << " + (" << CHOOSE_XFORM_GRAIN << " * (firstPoint.m_LastXfUsed + 1u))];\n\n";
#else
" secondPoint.m_LastXfUsed = xformDistributions[xfsel[THREAD_ID_Y] + (" << CHOOSE_XFORM_GRAIN << " * (firstPoint.m_LastXfUsed + 1u))];\n\n";//Partial cuburn hybrid.
#endif
@ -331,7 +331,7 @@ string IterOpenCLKernelCreator<T>::CreateIterKernelString(Ember<T>& ember, strin
{
os <<
#ifdef STRAIGHT_RAND
" secondPoint.m_LastXfUsed = xformDistributions[MwcNext(&mwc) % " << CHOOSE_XFORM_GRAIN << "];\n\n";//For testing, using straight rand flam4/fractron style instead of cuburn.
" secondPoint.m_LastXfUsed = xformDistributions[MwcNext(&mwc) & " << CHOOSE_XFORM_GRAIN_M1 << "];\n\n";//For testing, using straight rand flam4/fractron style instead of cuburn.
#else
" secondPoint.m_LastXfUsed = xformDistributions[xfsel[THREAD_ID_Y]];\n\n";
#endif
@ -408,7 +408,7 @@ string IterOpenCLKernelCreator<T>::CreateIterKernelString(Ember<T>& ember, strin
"\n"
//Populate randomized xform index buffer with new random values.
" if (THREAD_ID_Y == 0 && THREAD_ID_X < NWARPS)\n"
" xfsel[THREAD_ID_X] = MwcNext(&mwc) % " << CHOOSE_XFORM_GRAIN << ";\n"
" xfsel[THREAD_ID_X] = MwcNext(&mwc) & " << CHOOSE_XFORM_GRAIN_M1 << ";\n"
"\n"
" barrier(CLK_LOCAL_MEM_FENCE);\n"
//Another thread will have written to this thread's location, so read the new value and use it for accumulation below.