search for: _vp_quantize_couple_sort

...1) - fully optimized libvorbis for x86 CPUs - http://homepage3.nifty.com/blacksword/ Changes: 2006/08/24 Lancer 20060824 * Lancer is based on aotuv-r1_20051117 now. * add SSE optiomizations to _vp_couple. * add the dividing code of multi channel processing to xmmlib.h. * _vp_quantize_couple_memo, _vp_quantize_couple_sort is multithreaded under the OpenMP. 2006/08/18 Lancer 20060816 * mapping0_forward is more parallelized. * Buffered delay writer for Ogg stream is implemented. It makes possible that floor1_encode in the code book.* is parallelized. * floor1_encode is able to execute in the parallel processing now....

...because the exponent is stored biased. it's an old trick (and part of the reason for why floats are stored this way in the first place). anyhow, i figured i'd post this so i could feel justified in asking about the double indirection -- the double indirection appears to be required because _vp_quantize_couple_sort needs to return a permutation which yields the sorted data, do i understand that right? thanks -dean p.s. 1-3% oggenc perf improvement measured on all of p-m, p4, k8, efficeon with this patch and compilation options "-march=pentium4 -mfpmath=sse". Index: psy.c =========================...