-
I have found the problem.
The guts of it:
The main section of executeReductionStep (in gpu/gpukernel.cpp) line 836 is currently:
_context->getStreamReduceInterpolant( inputBuffer, resultExtents[0], resultExtents[1],
i, remainingExtent+i, 0, otherExtent, dim, _globalInterpolants[i] );
But when remainingExtent is 47, we do not want to go from i...47+i, We want i..46+i...
2006-12-25 10:33:04 UTC in Brook
-
I just tried it on the favored hardware: dx9 for ATI 1950 xt.
The same problem ocurrs.
2006-11-23 09:24:31 UTC in Brook
-
I found that on my platform (Nvidia 7800GS Windows 2000 CVS version of Brook) the reduce function works incorrectly if the size of the stream is 47. It basically double counts the item at position 46, and skips the item at position 44.
This happens when BRT_RUNTIME is set to dx9. ogl and cpu works fine.
I have not tested this on ATI hardware yet.
This problem ocurrs for many stream...
2006-11-22 21:35:56 UTC in Brook
-
hib committed patchset 130 of module lwneuralnet to the Lightweight Neural Network CVS repository, changing 2 files.
2006-11-12 21:41:40 UTC in Lightweight Neural Network
-
hib committed patchset 125 of module lwneuralnet to the Lightweight Neural Network CVS repository, changing 6 files.
2006-11-06 23:32:38 UTC in Lightweight Neural Network
-
hib committed patchset 124 of module lwneuralnet to the Lightweight Neural Network CVS repository, changing 2 files.
2006-11-06 22:59:32 UTC in Lightweight Neural Network