Make the quantizer exact.
authorTimothy B. Terriberry <tterribe@xiph.org>
Tue, 29 Jun 2010 00:15:09 +0000 (17:15 -0700)
committerTimothy B. Terriberry <tterribe@xiph.org>
Fri, 23 Jul 2010 15:48:01 +0000 (08:48 -0700)
commite04e293522a3cf3761eae3690b8efbc2aa69848b
tree5cb27f30b97c8296bcd2caa06bde2c344f895aac
parent08eed049d4f08943079483cdd5d5d9f865457a67
Make the quantizer exact.

This replaces the approximate division-by-multiplication in the
 quantizer with an exact one that costs just one add and one
 shift extra.
The asm versions have not been updated in this patch, and thus
 have been disabled, since the new method requires different
 multipliers which are not compatible with the old method.

Change-Id: I53ac887af0f969d906e464c88b1f4be69c6b1206
vp8/encoder/arm/csystemdependent.c
vp8/encoder/block.h
vp8/encoder/encodeframe.c
vp8/encoder/ethreading.c
vp8/encoder/onyx_int.h
vp8/encoder/quantize.c
vp8/encoder/x86/x86_csystemdependent.c