This patch adds column-based tiling. The idea is to make each tile
independently decodable (after reading the common frame header) and
also independendly encodable (minus within-frame cost adjustments in
the RD loop) to speed-up hardware & software en/decoders if they used
multi-threading. Column-based tiling has the added advantage (over
other tiling methods) that it minimizes realtime use-case latency,
since all threads can start encoding data as soon as the first SB-row
worth of data is available to the encoder.
There is some test code that does random tile ordering in the decoder,
to confirm that each tile is indeed independently decodable from other
tiles in the same frame. At tile edges, all contexts assume default
values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode),
and motion vector search and ordering do not cross tiles in the same
frame.
t log
Tile independence is not maintained between frames ATM, i.e. tile 0 of
frame 1 is free to use motion vectors that point into any tile of frame
0. We support 1 (i.e. no tiling), 2 or 4 column-tiles.
The loopfilter crosses tile boundaries. I discussed this briefly with Aki
and he says that's OK. An in-loop loopfilter would need to do some sync
between tile threads, but that shouldn't be a big issue.
Resuls: with tiling disabled, we go up slightly because of improved edge
use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf,
~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5%
on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is
concentrated in the low-bitrate end of clips, and most of it is because
of the loss of edges at tile boundaries and the resulting loss of intra
predictors.
TODO:
- more tiles (perhaps allow row-based tiling also, and max. 8 tiles)?
- maybe optionally (for EC purposes), motion vectors themselves
should not cross tile edges, or we should emulate such borders as
if they were off-frame, to limit error propagation to within one
tile only. This doesn't have to be the default behaviour but could
be an optional bitstream flag.
Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f
Update the code to call the new convolution functions to do subpixel
prediction rather than the existing functions. Remove the old C and
assembly code, since it is unused. This causes a 50% performance
reduction on the decoder, but that will be resolved when the asm for
the new functions is available.
There is no consensus for whether 6-tap or 2-tap predictors will be
supported in the final codec, so these filters are implemented in
terms of the 8-tap code, so that quality testing of these modes
can continue. Implementing the lower complexity algorithms is a
simple exercise, should it be necessary.
This code produces slightly better results in the EIGHTTAP_SMOOTH
case, since the filter is now applied in only one direction when
the subpel motion is only in one direction. Like the previous code,
the filtering is skipped entirely on full-pel MVs. This combination
seems to give the best quality gains, but this may be indicative of a
bug in the encoder's filter selection, since the encoder could
achieve the result of skipping the filtering on full-pel by selecting
one of the other filters. This should be revisited.
Quality gains on derf positive on almost all clips. The only clip
that seemed to be hurt at all datarates was football
(-0.115% PSNR average, -0.587% min). Overall averages 0.375% PSNR,
0.347% SSIM.
Change-Id: I7d469716091b1d89b4b08adde5863999319d69ff
This commit introduces a new convolution function which will be used to
replace the existing subpixel interpolation functions. It is much the
same as the existing functions, but allows for changing the filter
kernel on a per-pixel basis, and doesn't bake in knowledge of the
filter to be applied or the size of the resulting block into the
function name.
Replacing the existing subpel filters will come in a later commit.
Change-Id: Ic9a5615f2f456cb77f96741856fc650d6d78bb91
This commit changes the 4x4 iDCT to use same algorithm & constants as
other iDCTs. The 4x4 fDCT is also changed to be based on the new iDCT.
Change-Id: Ib1a902693228af903862e1f5a08078c36f2089b0
This commit makes the NearestMV match the chosen
best reference MV. It can be a 0,0 or non zero vector
which means the the compound nearest mv mode can
combine a 0,0 and a non zero vector.
Change-Id: I2213d09996ae2916e53e6458d7d110350dcffd7a
and called this function in vp9_dequant_idct_add_32x32_c when
eob == 1. For the test clip used, the decoder performance improved
by 21+%. Based on Yaowu's 16 point idct work.
Change-Id: Ib579a90fed531d45777980e04bf0c9b23c093c43
Separate out code to set the main encode speed
related rd thresholds. Some values changed from
the initial defaults for various new modes.
Quality test results pending but even the addition
of some further non-zero defaults helps encode speed
somewhat in limited testing on derf clips.
Adjustment of thresholds for quality / speed tradeoff
to follow.
Change-Id: I117ee473157e151a1b93193d5f393449328de20d
fixed a function prototypes to prevent compiler warnings;
removed a function not in use;
un-capitialize "Refstride" to ref_stride
Change-Id: Ib4472b6084f357d96328c6a06e795b6813a9edba
This commit changes the inverse 16 point dct to use the same algorithm
as the one for 32 point idct. In fact, now 16 point dct uses the exact
version of the souce code for even portion of the 32 point idct.
Tests showed current implementation has significant better accuracy
than the previous version. With this implementation and the minor bug
fix on forward 16 point dct, encoding tests showed about 0.2% better
compression of CIF set, test results on std-hd setting pending.
Change-Id: I68224b60c816ba03434e9f08bee147c7e344fb63
The commit fixes a minor error in 16 point fdct where in a rotation can
produce result of -1 instead of 0.
Change-Id: I45aac4a52bcd06225c6d04e643547a13e1c1aade
This commit changes the 32x32 idct to use integer only. The algorithm
was taken directly from "A Fast Computational Algorithm for the
Discrete Cosine Tranform" by W. Chen, et al., which was published in
IEEE Transaction on Communication Vol. Com.-25 No. 9, 1977. The signal
flow graph in the original paper is for a 32 point forward dct, the
current implementation of inverse DCT was done by follow the graph in
reversed direction.
With this implementation, the 32 point inverse dct contains a 16 point
inverse dct in its even portion, similarly the 16 point idct further
contains 8 point and 4 point inverse dcts.
As of patch 4, encoding tests showed there is no compression loss when
compared against the floating point baseline. Numbers even showed very
small postives. (cif: .01%, std-hd: .05%).
Change-Id: I2d2d17a424b0b04b42422ef33ec53f5802b0f378
This is identical to the later decisions made in encode_superblock().
This commit doesn't actually change anything, but makes the mbmi state
more consistent between the RD loop and the final encode result.
Change-Id: I9e735afb7c5a52e5b61728cb88c67ef9b9bf59be
The RD loop would change the pointer after the first mode (DC) was tested,
leading to corrupt block objects being provided for the others. This
would essentially render the i8x8 predictor useless.
Change-Id: I16c5906ca64fb34878ac32ce59af8974e4582bb8
Remove eob_max_offset markers and replace
with the generic skip_block flag to indicate
to the quantizer that all coeffs to be set to 0
and eob position set to 0;
Change-Id: Id477e8f8d4ec1a5562758904071013c24b76bfd7
Adds a special combination mode specific to intra prediciton
mode D45.
Current results with the compound inter/intra experiment:
derf: 0.2%
yt: 0.55%
std-hd: 0.75%
hd: 0.74%
Change-Id: I8976bdf3b9b0b66ab8c5c628bbc62c14fc72ca86
First step in simplifying the segment mode and
segment EOB flags into a simpler segment skip
flag that implies 0,0 mv and EOB at position 0.
Change-Id: Ib750cac31a7a02dc21082580498efd9f7d8d72a5
Simplification to eliminate a number of very large data
data structures. All zero run, zbin boosts for different
transform sizes are now limited to a maximum run length
of 15 before they max out the boost.
Some further work still needs be done to refactor, rationalize
and optimize the multiple quantizer functions.
The simplification coupled with tweaks to the 16 element array
now used for all transform sizes, has minimal effect on quality.
Change-Id: I6f3948b8ca0418b60d4db9030ff19026a34ed423
We'd backup and restore all cols for a 64x64 SB, but the array wouldn't
be big enough to hold all that data.
Change-Id: Ic68ea721bf07e0b2f3937bd16b0b734bcc743ce1
Adds a flag to disable features that would inhibit frame parallel
decoding. This includes backward adaptation and MV sorting based
on search in ref frame buffer.
Also includes some minor clean-ups.
Change-Id: I434846717a47b7bcb244b37ea670c5cdf776f14d
Added a quick eob == 0 check. Once the integer version of the dct32x32 is
complete, we can check for other eob cases.
For the 1080p clip used, the decoder performance improved by 4%.
Change-Id: I9390b6ed3c8be0c0c0a0c44c578d9a031d6e026e
