Commit Graph

9 Commits

Author SHA1 Message Date
Kyle Siefring
8394990b27 block error sse2: sum in 32 bits when possible
Add 31bit pairs before unpacking in x86 block error code

BUG=webm:1210

Change-Id: I5ca8c7f7775585a17fe09d6bbfc25e1f2955eb0a
2017-05-01 09:59:18 -07:00
Johann
3c16bbb73b block error sse2: use tran_low_t
Change-Id: Ib04990e4a7bda9fbf501f294da2057a2b2595deb
2017-02-24 01:33:35 +00:00
Johann
44600442dc bitdepth conversion: really use num elements
The previous implementation confused bit/bytes/elements. It was using
'32' as the multiplier but that was mistakenly adopted because a 32x32
transform embedded the stride.

Change-Id: Ieeb867a332416b9a40580b5e7c9b20088e9e691a
2017-02-16 15:02:48 +00:00
Johann
537949a9df block_error_fp highbd sse2: use tran_low_t for coeff
BUG=webm:1365

Change-Id: Id2ed3ebaaaa6a4b68628c23e08b64ea5f1341761
2017-02-07 15:03:28 +00:00
Johann
41a0a0cb35 Use newer x86inc.asm
Rename updated version of x86inc.asm

Use "private_prefix" instead of "program_name" and make vpx the default
prefix.

Change-Id: I4883a99b2aee8e5dc9f2c16a2e6f4b5d6e4de458
2015-08-07 16:44:44 -07:00
Jingning Han
f2cf3c06a0 Reduce required xmm number by one in block_error_fp
Use 6 xmms instead of 8.

Change-Id: If976ad85d09191d2fb0565399d690f2869dbbcc7
2015-04-01 12:07:35 -07:00
Jingning Han
1470529f62 Refactor block_yrd function for RTC coding mode
This commit separates Hadamard transform/quantization operations
from rate and distortion computation in block_yrd. This allows one
to skip SATD computation when all transform blocks are quantized
to zero. It also uses a new block error function that skips
repeated computation of sum of squared residuals. It reduces the
CPU cycles spent on block error calculation in block_yrd by 40%.

Change-Id: I726acb2454b44af1c3bd95385abecac209959b10
2015-04-01 12:00:43 -07:00
Ronald S. Bultje
af660715c0 Make coefficient skip condition an explicit RD choice.
This commit replaces zrun_zbin_boost, a method of biasing non-zero
coefficients following runs of zero-coefficients to be rounded towards
zero, with an explicit skip-block choice in the RD loop.

The logic is basically that if individual coefficients should be rounded
towards zero (from a RD point of view), the trellis/optimize loop should
take care of it. If whole blocks should be zero (from a RD point of
view), a single RD check is much more efficient than a complete
serialization of the quantization loop.

Quality change: derf +0.5% psnr, +1.6% ssim; yt +0.6% psnr, +1.1% ssim.
SIMD for quantize will follow in a separate patch. Results for other
test sets pending.

Change-Id: Ife5fa641163ac5150ac428011e87188f1937c1f4
2013-06-28 10:28:49 -07:00
Ronald S. Bultje
54b2a59623 Implement SSE2 block_error.
Change vp9_block_error() to return a 64bit error variable, change all
callers to expect a 64bit return value (this will prevent overflows,
which we basically don't check for at all right now). Remove duplicate
block_error() function, which fixed that through truncation. Remove
old (incompatible) mmx/sse2 block_error SIMD versions and replace with
a new one that returns a 64bit value.

Encoding time of first 50 frames of bus @ 1500kbps goes from 3min29 to
3min23, i.e. a 3% overall speedup.

Change-Id: Ib71ac5508b5ee8a80f1753cd85d72df1629abe68
2013-06-21 12:54:52 -07:00