Jingning Han [Wed, 28 Oct 2015 18:34:45 +0000 (11:34 -0700)]
Account for variable txfm sizes in coeff token packing
This commit makes the coefficient token packtization process account
for variable transform block sizes supported in a single processing
block. It fixes an enc/dec mismatch issue when var-tx, ext-tx, and
misc-fixes experiments are all turned on.
Jingning Han [Tue, 27 Oct 2015 23:50:27 +0000 (16:50 -0700)]
Add tx_type counts in key frame
Properly update the transform type counts in key frame coding at
decoder. It fixes an enc/dec mismatch issue when both ext-tx and
misc-fixes are turned on.
Peter de Rivaz [Tue, 27 Oct 2015 10:50:00 +0000 (10:50 +0000)]
Accumulate EXT_TX counts for multithread
EXT_TX introduces some new symbols to be decoded.
The encoder counts how many times these are used.
In multithreaded mode, the counts from the worker threads
need to be accumulated into the main thread.
This change means that VP10/VPxEncoderThreadTest now works
with more choices of cpu-used and number of passes.
Jingning Han [Mon, 26 Oct 2015 19:32:30 +0000 (12:32 -0700)]
Fix lossless coding
Use inter_block_yrd as rate-distortion optimization for lossless
coding. This fixes transform coefficient buffer swap use case and
resolves the unit test failure related to lossless coding.
Jingning Han [Mon, 26 Oct 2015 18:09:55 +0000 (11:09 -0700)]
Make transform block partition scheme support use largest txfm setting
This commit properly resets the recursive transform block partition
array in the settings of using largest transform block size at frame
header level. It fixes one of the unit test failure related to the use
of frame level fixed transform block size with 440 color format.
Alex Converse [Wed, 14 Oct 2015 18:03:14 +0000 (11:03 -0700)]
palette: Replace rand() call with custom LCG.
The custom LCG is based on the POSIX recommend constants for a 16-bit
rand(). This implementation uses less computation than typical standard
library procedures which have been extended for 32-bit support, is
guaranteed to be reentrant, and identical everywhere.
base_frame_target is supposed to track the idealized bit
allocation based on error score and not the actual bits
allocated to each frame.
The clamping of this value based on the VBR min and max pct values
was causing a bug where in some cases the loop that adjusts the
active max quantizer for each GF group was running out of bits at
the end of a KF group. This caused a spike in Q and some ugly artifacts.
A second change makes sure that the calculation of the active
Q range for a group DOES, however, take account of clamping.
Jingning Han [Fri, 23 Oct 2015 21:27:21 +0000 (14:27 -0700)]
Properly handle non-420 color format in recursive transform scheme
This commit makes the recursive transform block partitioning properly
handle the non-420 color format. It resolves an enc/dec mismatch
issue in that setting when var-tx experiment is turned on.
Jingning Han [Fri, 23 Oct 2015 00:25:00 +0000 (17:25 -0700)]
Use explicit block position in foreach_transformed_block
Add the row and column index to the argument list of unit functions
called by foreach_transformed_block wrapper. This avoids the
repeated internal parsing according to the block index.
Jingning Han [Thu, 22 Oct 2015 19:06:32 +0000 (12:06 -0700)]
Reset tx_type in recursive transform block partitioning
Temporarily reset the transform type in the inter modes when
recursive transform block partitioning is used. This resolves an
enc/dec mismatch issue in nextgenv2 codebase when both var-tx and
ext-tx experiments are turned on.
Geza Lore [Thu, 15 Oct 2015 17:28:31 +0000 (18:28 +0100)]
Optimize vp9_highbd_block_error_8bit assembly.
A new version of vp9_highbd_error_8bit is now available which is
optimized with AVX assembly. AVX itself does not buy us too much, but
the non-destructive 3 operand format encoding of the 128bit SSEn integer
instructions helps to eliminate move instructions. The Sandy Bridge
micro-architecture cannot eliminate move instructions in the processor
front end, so AVX will help on these machines.
Further 2 optimizations are applied:
1. The common case of computing block error on 4x4 blocks is optimized
as a special case.
2. All arithmetic is speculatively done on 32 bits only. At the end of
the loop, the code detects if overflow might have happened and if so,
the whole computation is re-executed using higher precision arithmetic.
This case however is extremely rare in real use, so we can achieve a
large net gain here.
The optimizations rely on the fact that the coefficients are in the
range [-(2^15-1), 2^15-1], and that the quantized coefficients always
have the same sign as the input coefficients (in the worst case they are
0). These are the same assumptions that the old SSE2 assembly code for
the non high bitdepth configuration relied on. The unit tests have been
updated to take this constraint into consideration when generating test
input data.
Ronald S. Bultje [Tue, 20 Oct 2015 17:31:40 +0000 (13:31 -0400)]
vp10: clip MVs before adding to find_ref_mvs() list.
This causes the output of find_ref_mvs() to always be unique or zero.
A nice side-effect of this is that it also causes the output of
find_ref_mvs_sub8x8() to be unique-or-zero, and it will not ignore
available candidate MVs under certain conditions.
Geza Lore [Mon, 19 Oct 2015 13:05:35 +0000 (14:05 +0100)]
Optimize vpx_quantize_{b,b_32x32} assembler.
Added optimization of the 8 bit assembly quantizer routines. This makes
these functions up to 100% faster, depending on encoding parameters.
This patch maskes the encoder faster in both the high bitdepth and 8bit
configurations. In the high bitdepth configuration, it effects profile 0
only.
Based on my profiling using 1080p input the net gain is between 1-3% for
the 8 bit config, and around 2.5-4.5% for the high bitdepth config,
depending on target bitrate. The difference between the 8 bit and high
bitdepth configurations for the same encoder run is reduced by 1% in all
cases I have profiled.
vp10: allow MV refs to point outside visible image.
In VP9, the ref MV had to point to a block that itself fully resided
within the visible image, i.e. all borders of the image had to be
within the visible borders of the coded frame. This is somewhat
illogical, and had obscure side effects, e.g. clamping of fairly
reasonable motion vectors such as 0,0 were clipped to negative values
if the block was overhanging on frame edges (such as the last rows
on 1080p content), which makes no sense whatsoever.
Instead, relax clamping constraints such that the ref MVs are allowed
to point to blocks exactly outside the visible edges in both Y as well
as UV planes, including the 8tap filter edges (that's why the offset is
8 pixels + block size).
projects / libvpx / log