Multithreaded encoding was breaking at low bitrates
Please review/comment. Not sure if this is the best fix.
Change-Id: I87468c765372593fd865bc82e25121ebb8ca6af2
Make bilinearfilter_arm.c compiled only when HAVE_ARMV6, as its definitions
are v6 only. This is normally not a problem for static builds as the file
is elided at link time, but this was not being done properly for the
--enable-shared --enable-pic build.
Change-Id: Ic800a7cde751f74f22555c5b247f99f9df5e550d
Merged multi-resolution motion estimation with regular motion
estimation function in order to remove duplicated part. This
caused slight changes in multi-resulotion encoder quality &
performance.
Change-Id: Ib4ecc7acfebfe5eea959b5b91febae6db7b95fd1
The total_stats, this_frame_stats, and total_left_stats structures
were previously create by a heap allocation, despite being of fixed
size. These structures were allocated and deallocated during
{de,}allocate_compressor_data, which is reinvoked whenever the frame
size changes. Unfortunately, this clobbers the total_stats and
total_left_stats data.
Historically, these were variable size at one time, due to the first
pass motion map, which necessitated their being created by a unique
heap allocation. However, this bug with the total_stats being
clobbered has probably been present since that initial implementation.
These structures are instead moved to be stored within the struct
twopass_rc directly, rather than being heap allocated separately.
Change-Id: I7f9e519e25c58b92969071f0e99fa80307e0682b
When mb_skip_coeff is set, the idct is not necessary. Prior
to this patch, the code would call idcts based on leftover
eob information. This patch will now skip the idct for
SPLIT_MV and clear out the eobs for B_PRED, forcing the idct
to be skipped.
Change-Id: If5b0d2ed3ebd07789d30ec5160df927485fcaa17
These functions are now used by the encoder.
This is WIP with the goal of creating a common idct/add for
the encoder and decoder. A boost of 1.8% was seen for
the HD rt test clip used.
[Tero] Added needed changes to ARM side.
Change-Id: Ibbb8000be09034203d7adffc457d3c3f8b06a5bf
While doing motion search on a macroblock, we usually call
vp8_find_near_mvs once per reference frame. Actually, for
different reference frames, the only difference in calculating
these near_mvs is they may have different sign_bias, which
causes a sign change in resulting near_mvs. In this change, we
only do find_near_mvs for the first reference frame. For other
reference frames, only need to adjust the near_mvs according to
that reference frame's sign_bias value.
Change-Id: I661394b49c6ad79fed7d0f2eb2be239b9c56f149
Add the notion of deferred validation of parameters. We don't want to
validate the cq_level at initialization time, because it won't have
been set via set_param() yet.
Change-Id: Ia1308395e8c10e0b1dc4e9af3a09b2bd6744cc30
The fast-path for skipped MBs was not correctly respecting the
block type during update of the coefficient counts. Extracted
this from part of change I365cfb6ac636f19c545f682e3aeac185253abaef
Change-Id: I53d8cf0a00a98034b97b0ed3414b703bae74a739
Do the test filtering in the existing backup frame buffer instead of
the original. Copy the original data into extra buffer before doing
the filtering. This way there is no need to restore the original
unfiltered frame at the end of level picking process.
This came up in some discussions with Johann. Thanks!
Change-Id: I495f4301d983854673276c34ec0ddf9a9d622122
The example encoder down-samples the input video frames a number of
times with a down-sampling factor, and then encodes and outputs
bitstreams with different resolutions.
Support arbitrary down-sampling factor, and down-sampling factor
can be different for each encoding level.
For example, the encoder can be tested as follows.
1. Configure with multi-resolution encoding enabled:
../libvpx/configure --target=x86-linux-gcc --disable-codecs
--enable-vp8 --enable-runtime_cpu_detect --enable-debug
--disable-install-docs --enable-error-concealment
--enable-multi-res-encoding
2. Run make
3. Encode:
If input video is 1280x720, run:
./vp8_multi_resolution_encoder 1280 720 input.yuv 1.ivf 2.ivf 3.ivf 1
(output: 1.ivf(1280x720); 2.ivf(640x360); 3.ivf(320x180).
The last parameter is set to 1/0 to show/not show PSNR.)
4. Decode:
./simple_decoder 1.ivf 1.yuv
./simple_decoder 2.ivf 2.yuv
./simple_decoder 3.ivf 3.yuv
5. View video:
mplayer 1.yuv -demuxer rawvideo -rawvideo w=1280:h=720 -loop 0 -fps 30
mplayer 2.yuv -demuxer rawvideo -rawvideo w=640:h=360 -loop 0 -fps 30
mplayer 3.yuv -demuxer rawvideo -rawvideo w=320:h=180 -loop 0 -fps 30
The encoding parameters can be modified in vp8_multi_resolution_encoder.c,
for example, target bitrate, frame rate...
Modified API. John helped a lot with that. Thanks!
Change-Id: I03be9a51167eddf94399f92d269599fb3f3d54f5
This value needs to be copied to each thread's data structure.
This fixed artifact problem in multi-thread encoder.
Change-Id: Iab6d9745a1d44846aa503184705376f63a505597
to the dqcoeff or qcoeff buffer. The encoder would
populate the dc coeffs of the y blocks as a separate
stage (recon_dcblock) and the decoder would use a special
version of the idct. This change eliminates the extra copy
and reduces the code footprint.
[Tero] Added needed changes to armv6 and NEON assembly.
Change-Id: I83202ffdbaf83f6e5dd69f4ba2519fcf0b13b3ba
API was not returning correct partition sizes on arm targets.
The armv5 token packing functions were not storing the information to the
partition size table.
As a fix, have one boolcoder instance allocated for each partition so
that partition sizes are internally available after all partitions
were encoded. This will also allow more flexibility in producing
several partitions in parallel.
Use buffer validation (overflow check) in all ARM bitpacking
functions.
Change-Id: I31c8a11d8a7613676f0ff50928cb2a2ab14fd169
Storing vp8_bilinear_filters_mmx in an mmx file and using it in an sse2
file is bad
Moving towards allowing --disable-mmx
Change-Id: I20493b35bdedcdcfc0915e6f05fdbe6c81a4a742
There was an implicit reference frame test order (typically LAST,
GOLD, ARF) in the mode selection logic, but this doesn't provide the
expected results when some reference frames are disabled. For
instance, in real-time mode, the speed selection logic often disables
the ARF modes. So if the user disables the LAST and GOLD frames, the
encoder was always choosing INTRA, when in reality searching the ARF
in this case has the same speed penalty as searching LAST would have
had.
Instead, introduce the notion of a reference frame search order. This
patch preserves the former priorities, so if a frame is disabled, the
other frames bump up a slot to take its place. This patch lays the
groundwork for doing something smarter in the frame test order, for
example considering temporal distance or looking at the frames used by
nearby blocks.
Change-Id: I1199149f8662a408537c653d2c021c7f1d29a700
Extend buffer write validation (overflow check) to single token
partition packing, both mb and row based functions.
Change-Id: I36e19b7d37fc43712d05c70e3ad223d3eb5b973d
Patch set 2: 64 bit build fix
Patch set 3: 64 bit crash fix
[Tero]
Patch set 4: Updated ARMv6 and NEON assembly.
Added also minor NEON optimizations to subtract
functions.
Patch set 5: x86 stride bug fix
Change-Id: I1fcca93e90c89b89ddc204e1c18f208682675c15
The calculated frame_rate is a state variable in the codec, and
shouldn't be maintained in the configuration struct. Move it to the
main part of cpi so that it isn't clobbered when the configuration
struct is updated. The initial framerate estimate is moved from the
vp8_cx_iface.c wrapper into the body of init_config() in onyx_if.c, so
that it is only called once and not reset on every call to
vp8_change_config().
Change-Id: I8d9a3d1283330d1ee297d07e9d78d1f2875f2465
Ronald recently sent me this patch that he did in April.
> From: Ronald S. Bultje <rbultje@google.com>
> Date: Thu, 28 Apr 2011 17:30:15 -0700
> Subject: [PATCH] SSE2 optimizations for
> vp8_build_intra_predictors_mby{,_s}().
HD decode tests have shown a performance boost up to 1.5%,
depending on material.
Patch set 3: Fixed encoder crash.
Change-Id: Ie1fd1fa3dc750eec1a7a20bfa2decc079dcf48c8
Call the idct/add after the tokenize. This is WIP with
the goal of creating a common idct/add for the encoder and
decoder. This move is necessary because the decoder's version
of the idct clobbers qcoeff, which is used by the tokenize.
Change-Id: I6b08d8e8397cd873647fa4fb9469884e3c876756
Added ARM optimized intra 4x4 prediction
- 2x faster on Profiler compared to C-code compiled with -O3
- Function interface changed a little to improve BLOCKD structure
access
Change-Id: I9bc2b723155943fe0cf03dd9ca5f1760f7a81f54
The referenced function (SignalObjectAndWait) isn't used. Reduces the
warnings with mingw32-w64 which defines this.
Change-Id: I4ce592879ec9372bf196dac640204c4d370bd210
vp8cx_mb_init_quantizer() needs to be called at least once to get
all values calculated. This change added one check to decide if
we could skip initialization or not.
Change-Id: I3f65eb548be57580a61444328336bc18c25c085b
Added code to clip the buffer level to the maximum buffer
size. Without this the buffer level would increase
unchecked.
This bug was found when encoding an essentially static
scene at 2Mb/s. The encoder is unable to generate frames
consistent with the high data-rate because Q bottoms out
at Qmin.
As frames generated are consistently undersized the buffer
level increases and does not get checked against the
maximum size specified by the user (or default).
Change-Id: Id8a3c6323d3246da50f7cb53ddbf78b5528032c6
Fix compiler warning for passing a non const array
to a function expecting a const array by using an
intermediary pointer and casting.
Change-Id: I9bdd358ebdc926223993fb8fb2098ffedd2f3fc7
Changed 'int eob' to 'char *eob' in BLOCKD so that both encoder and
decoder will use eobs[25] array from MACROBLOCKD structure. In future,
this will enable use of the decoder side IDCT in the encoder.
Change-Id: I6e1c011628cb8864fd4a0b80f0279ce16a5ca978
Adding support for several partitions within one input fragment.
This is necessary to fully support all possible packetization
combinations in the VP8 RTP profile. Several partitions can
be transmitted in the same packet, and they can only be split
by reading the partition lengths from the bitstream.
Change-Id: If7d7ea331cc78cb7efd74c4a976b720c9a655463
In some situations (f.g. error-resilient is turned on), vp8cx_mb
_init_quantizer() was called once per macroblock. Added checks
to avoid calculations when there is no change.
Change-Id: Ie4f0a5ade2202041254990a4e9d5b03bd1ac5aea
It is discovered that in rare situations the 2nd order block may
produce a few small magnitude coefficients that has no effect on
reconstruction. The situations are a combination of low quantizer
values (high quality) and low energy in residual signals (content
dependent). This commit added code to detect such cases and reset
the 2nd order block to all 0.
Patch 1 to 4 used code to do all-zero-check on idct result buffer,
and tests on derf set showed a consistent gain of .12%-.14% on all
metrics.But due to a recent change Ie31d90b, the idct result buffer
is not longer populated. So patch 5&6 use an alternative method to
detect the situations. Tests on derf set now shows a consistent
quality gain of .16%-.20%.
As suggested by Jim, Patch 7&8 removed the condition of all first
order block not having any coefficient, instead we reset 2nd order
coefficients to all 0 if sum of absolute value of the coefficients
is small. So it does slightly more than just detecting the oddity
as discussed above, but tests on derf set now show a consistent
gain of .20%-.23% on all metrics.
It is worth noting here that this change does not have any effect
on mid/high quantizer range, it only affects the quantizer value
18 or blow. Within this range, the change helps compression by up
to 2.5% on clips in the derf set.
Change-Id: I718e19cf59a4fc2462cb7070832759beb9f7e7dd
Tests showed ~1.2% performance boost on the HD clip used.
Performance will vary based on material.
Change-Id: Icbcf1a828750d5b4ae5252bf596b3ef594042e8a
Interleaved vp8_find_near_mvs and vp8_mv_ref_probs.
2.5% to 4% performance improvement for the HD clips used.
Change-Id: Id888b667cf5ae2f0e19da18743140f055ff7de8d
The partial frame copy function used to copy an extra 8 lines above
and below. The partial frame filtering can only modify 3 pixel rows
above the partial frame. Reduce copy to bare minimum needed, which is
4 lines, so that partial filtering on copied frame is possible.
Define the "magic" fraction number for partial filtering in
loopfilter.h .
Change-Id: I4791ffc541b6884b12759a0d0714a8faf16147ec
Restructure if statement to clarify the error condition. Trigger the
error before clobbering pc-> variables.
Change-Id: Id01cab798a341ce9899078fdcec265a0e942a0b7
Decode the mv mode with if-then-elses instead of traversing
the vp8_mv_ref_tree data structure. This will make it
easier to interleave vp8_find_near_mvs and vp8_mv_ref_probs.
Change-Id: I1e798d6ec40fcaeeff06ccc82f81201978d12f74
check to make sure that cx_data buffer has enough room before
writting to it, prior behavior did not which could result in a crash.
Change-Id: I3fab6f2bc4a96d7c675ea81acd39ece121738b28
During the _pick only the Y plane is examined. In addition, data beyond
the borders of the frame is not read.
Change-Id: Ic549adfca70fc6e0b55f8aab0efe81f0afac89f9
Instead of using the predict buffer, the decoder now writes
the predictor into the recon buffer. For blocks with eob=0,
unnecessary idcts can be eliminated. This gave a performance
boost of ~1.8% for the HD clips used.
Tero: Added needed changes to ARM side and scheduled some
assembly code to prevent interlocks.
Patch Set 6: Merged (I1bcdca7a95aacc3a181b9faa6b10e3a71ee24df3)
into this commit because of similarities in the idct
functions.
Patch Set 7: EC bug fix.
Change-Id: Ie31d90b5d3522e1108163f2ac491e455e3f955e6
It was crashing when number of partitions was bigger than the number
of MB rows (ex. 128x96 with 8 partitions).
Start point was not checked against mb_rows, plus extra
"empty" partitions were not written out.
Change-Id: I9c2f013b9ec022354b658fab4ef799ff8b1de93d
Added the ability to create rate-targeted, temporally
scalable, VP8 compatible bitstreams.
The application vp8_scalable_patterns.c demonstrates how
to use this capability. Users can create output bitstreams
containing upto 5 temporally separable streams encoded
as a single VP8 bitstream.
(previously abandoned as:
I92d1483e887adb274d07ce9e567e4d0314881b0a)
Change-Id: I156250a3fe930be57c069d508c41b6a7a4ea8d6a
buffer_level in VP8_COMP and starting_buffer_level, optimal_buffer_level
and maximum_buffer_size in VP8_CONFIG changed from int to int64_t
to avoid potential crash issues for larger target bit rates.
Change-Id: I0d5ab6c8a44c2fef51f30cd8df4bb4b739c5df26
Previous entropy probs need to be saved (and restored) only when
current updates are not propagated.
Change-Id: Ie6ee0543066e30874e56258be0a6b7d2dd2fdb2b
Uninitialized data could be written to the first pass file when no
motion vectors are present in the frame.
Also fix a number of compiler warnings.
Change-Id: Icc9f53b6d33da9de4563d86d9fd591910473ea90
The data processed by the loopfilter overlaps. At the block level, this
results in some redundant transforms. Grouping the filtering allows for
a single 16x16 transpose (and inversion) instead of three 16x8 transposes
(and three more inversions).
This implementation is x86_64 only. We retain the previous
implementation for x86.
Improvements are obviously material dependant, but it seems to be ~%1 in
tests here.
Change-Id: I467b7ec3655be98fb5f1a94b5d145e5e5a660007
vp8_find_near_mvs() is being called on all possible reference frames
but the data computed may be used if the loop exits early, which can
be due to x->skip beign set to 1.
Optimize this by call vp8_find_near_mvs() laziy only if it is going
to be used and not computed yet.
Change-Id: Iccdbd4c962a670c9f2c99b8aca8096042ca5dc98
Changes to the selection of Q limits for two pass
and two pass CQ mode.
Allowance made for Mode and motion vector costs.
Some refactoring of common code.
For Derf and YT sets CQ mode average improvement
circa 1% (SSIM and Global PSNR).
Some increased tendency to undershoot even when
user CQ not reached.
Patch2: Removed some test code accidentally merged.
Change-Id: Icf74d13af77437c08602571dc7a97e747cce5066
Sync with loopfilter thread just at the beginning of next frame encoding.
This returns control to application faster and allows a better multicore scaling.
When PSNR packets are generated the final filtered frame is needed imediatly
so we cannot delay the sync.
Change-Id: I288d97b5e331d41d6f5bb49d97986fa12ac6f066
Make sure that this header is listed as one of the sources, so that it
will be installed if necessary.
Change-Id: I2427e494488126b179151dc21043c1e2c8ba5991
- Removed fast_fdct4x4_neon and fast_fdct8x4_neon
- Uses now short_fdct4x4 and short_fdct8x4
- Gives ~1-2% speed-up on Cortex-A8/A9
Change-Id: Ib62f2cb2080ae719f8fa1d518a3a5e71278a41ec
Rd and Rm registers should be different in 'mul'. This register
combination results in unpredictable behaviour. GCC will give
a warning and RVCT an error in this case.
Restriction applies only to armv5 targets and not for armv6 and above.
Change-Id: I378d17c51e1f16a6820814fbed43e115aaabb03e
These changes fixes a glitch between the RTP profile and the input
partitions interface. Since there's no way for the user to know the
actual number of partitions, the decoder have to read the
multi_token_paritition bits also when input partitions mode is
enabled.
Included are also a couple of fixes for issues with independent
partitions and uninitialized memory reads.
Change-Id: I6f93b15287d291169ed681898ed3fbcc5dc81837
- Updated walsh transform to match C
(based on Change Id24f3392)
- Changed fast_fdct4x4 and 8x4 to short_fdct4x4 and 8x4
correspondingly
Change-Id: I704e862f40e315b0a79997633c7bd9c347166a8e
Modified original patch If2f07220885c4c3a0cae0dace34ea0e36124f001
according to comments. Scheduled code a little bit to prevent some
interlocks.
Change-Id: I338f02b881098782f82af63d97f042b85e63e902
In the "Removed bmi copy to/from BLOCKD" commit, the copy
to the bmi in BLOCKD was eliminated. The clamp_mvs() used
the bmi in BLOCKD, which now contains incorrect values. This
patch fixes this problem.
Change-Id: I8eca1eaf4015052b0b63e90876f7ad321aba7cff
vp8_update_zbin_extra() is called all the time even though the fast
quantizer doesn't use it. Skip this call if fast quantizer is used.
Change-Id: Ia711c38431930cc2486cf59b8466060ef0e9d9db
This change makes sure that no key frame recoding in real-time mode
even if CONFIG_REALTIME_ONLY is not configured.
Change-Id: Ifc34141f3217a6bb63cc087d78b111fadb35eec2
for SPLITMV and B_PRED modes. Modified code to use the bmi
found in mode_info_context instead of BLOCKD. On the decode
side, the uvmvs are calculated only when required, instead of
every macroblock. This is WIP. (bmi should eventually be
removed from BLOCKD)
Small performance gains noticed for RT encodes and decodes.(VGA)
Change-Id: I2ed7f0fd5ca733655df684aa82da575c77a973e7
Prepend idct function names with vp8_
so that under profiling they show up
associated with libvpx.
Change-Id: I4fe357b50236cb7730a4cc00164c0a3487a1d8b4
The data that the simple horizontal loopfilter reads is aligned, treat
it accordingly.
For the vertical, we only use the bottom 4 bytes, so don't read in 16
(and incur the penalty for unaligned access).
This shows a small improvement on older processors which have a
significant penalty for unaligned reads.
postproc_mmx.c is unused
Change-Id: I87b29bbc0c3b19ee1ca1de3c4f47332a53087b3d
Prepend . to local labels in assembly code. This
allows non unique labels within a file. Also
makes profiling information more informative
by keeping the function name with the loop name.
Change-Id: I7a983cb3a5ba2413d5dafd0a37936b268fb9e37f
Calculations were incorrectly classified as either
SSE3 or SSSE3. Only using SSE2 instructions.
Cleanup function names and make non-RTCD code work
as well.
Change-Id: I48ad0218af0cc51c5078070a08511dee43ecfe09
Calculations were incorrectly classified as either
SSE3 or SSSE3. Only using SSE2 instructions.
Cleanup function names and make non-RTCD code work
as well.
Change-Id: I29f5c2ead342b2086a468029c15e2c1d948b5d97
When active map is specified and the current frame is not a key frame,
golden frame nor a altref frame then copy only those active regions.
This significantly reduces encoding time by as much as 19% on the test
system where realtime encoding is used. This is particularly useful
when the frame size is large (e.g. 2560x1600) and there's only a few
action macroblocks.
Change-Id: If394a813ec2df5a0201745d1348dbde4278f7ad4
Instead of a single mid GF boost apply a few extra bits to
every other frame. This gives a very small average metrics
improvement on both derf and YT sets.
Also use min GF interval as min KF interval.
Change-Id: Iee238b8cae0ffaed850a5a944ac825cee18da485
Since the block will be interpreted as an inter block, the mode will
be interpreted as a motion vector, resulting in bad concealment.
Change-Id: Ifcc685ae1cc883492bce6dbd61e418d91a89b053
Since the block will be interpreted as an inter block, the mode will
be interpreted as a motion vector, resulting in bad concealment.
Change-Id: Ifcc685ae1cc883492bce6dbd61e418d91a89b053
This reverts commit b5ea2fbc2c. Further
testing showed noticable keyframe popping in some cases, reverting this
for now to give time for a proper fix.
Conflicts:
vp8/encoder/onyx_if.c
vp8/encoder/ratectrl.c
Change-Id: I159f53d1bf0e24c035754ab3ded8ccfd58fd04af
EC expects the subblock MVs to be populated, but
f1d6cc79e4 removed this code. This
commit restores it, protected by CONFIG_ERROR_CONCEALMENT. May move this
to the EC code more directly in the future.
Change-Id: I44f8f985720cb9a1bf222e59143f9e69abf56ad2
When error concealment is enabled the first key frame must
be successfully received before error concealment is activated.
Error concealment will be activated when the delta following
delta frame is received.
Also fixed a couple of bugs related to error tracking in
multi-threading. And avoiding decoding corrupt residual
when we have multiple non-resilient partitions.
Change-Id: I45c4bb296e2f05f57624aef500a874faf431a60d
This patch fixes an OOB read when error concealment is enabled and the
partition sizes are corrupt. The partition size read from the bitstream
was not being validated in EC mode.
Change-Id: Ia81dfd4bce1ab29ee78e42320abe52cee8318974
EC expects the subblock MVs to be populated, but
f1d6cc79e4 removed this code. This
commit restores it, protected by CONFIG_ERROR_CONCEALMENT. May move this
to the EC code more directly in the future.
Change-Id: I44f8f985720cb9a1bf222e59143f9e69abf56ad2
When error concealment is enabled the first key frame must
be successfully received before error concealment is activated.
Error concealment will be activated when the delta following
delta frame is received.
Also fixed a couple of bugs related to error tracking in
multi-threading. And avoiding decoding corrupt residual
when we have multiple non-resilient partitions.
Change-Id: I45c4bb296e2f05f57624aef500a874faf431a60d
This patch fixes an OOB read when error concealment is enabled and the
partition sizes are corrupt. The partition size read from the bitstream
was not being validated in EC mode.
Change-Id: Ia81dfd4bce1ab29ee78e42320abe52cee8318974
This patch fixes a bug in the interaction between the recode loop and
spatial resampling. If the codec was in a spatial resampling state,
and a subsequent iteration of the recode loop disables resampling,
then the source buffer must be reset to the unscaled source.
Change-Id: I4e4cd47b943f6cd26a47449dc7f4255b38e27c77
Changed motion search in vp8_find_best_half_pixel_step() to be the
same as in vp8_find_best_sub_pixel_step(), which checks 5 points
instead of 8 points. This only affects real-time mode with
cpu-used >=9. Tests showed it gives 2% encoding speedup with
a quality loss(psnr) of up to 0.5%.
Change-Id: I16049cad1535002346d46cfdfad345bfc3dc5146
the neon code made several assumptions which were broken by a recent
change: https://review.webmproject.org/2676
update the code with new assumptions and guard them with a compile time
assert
Change-Id: I32a8378030759966068f34618d7b4b1b02e101a0
Since this is the only ABI incompatible change since the last release,
convert it to use the control interface instead. The member of the
configuration struct is replaced with the VP8E_SET_MAX_INTRA_BITRATE_PCT
control.
More significant API changes were expected to be forthcoming when this
control was first introduced, and while they continue to be expected,
it's not worth breaking compatibility for only this change.
Change-Id: I799d8dbe24c8bc9c241e0b7743b2b64f81327d59
This change implemented same idea in change "Preload reference area
to an intermediate buffer in sub-pixel motion search." The changes
were made to vp8_find_best_sub_pixel_step() and vp8_find_best_half
_pixel_step() functions which are called when speed >= 5. Test
result (using tulip clip):
1. On Core2 Quad machine(Linux)
rt mode, speed (-5 ~ -8), encoding speed gain: 2% ~ 3%
rt mode, speed (-9 ~ -11), encoding speed gain: 1% ~ 2%
rt mode, speed (-12 ~ -14), no noticeable encoding speed gain
2. On Xeon machine(Linux)
Test on speed (-5 ~ -14) didn't show noticeable speed change.
Change-Id: I21bec2d6e7fbe541fcc0f4c0366bbdf3e2076aa2
There were some situations that the start motion vectors were
out of range. This fix adjusted range checks to make sure they
are checked and clamped.
Change-Id: Ife83b7fed0882bba6d1fa559b6e63c054fd5065d
sharpness was not recalculated in vp8cx_pick_filter_level_fast
remove last_filter_type. all values are calculated, don't need to update
the lfi data when it changes.
always use cm->sharpness_level. the extra indirection was annoying.
don't track last frame_type or sharpness_level manually. frame type
only matters for motion search and sharpness_level is taken care of in
frame_init
move function declarations to their proper header
Change-Id: I7ef037bd4bf8cf5e37d2d36bd03b5e22a2ad91db
In sub-pixel motion search, the search range is small(+/- 3 pixels).
Preload whole search area from reference buffer into a 32-byte
aligned buffer. Then in search, load reference data from this buffer
instead. This keeps data in cache, and reduces the crossing cache-
line penalty. For tulip clip, tests on Intel Core2 Quad machine(linux)
showed encoder speed improvement:
3.4% at --rt --cpu-used =-4
2.8% at --rt --cpu-used =-3
2.3% at --rt --cpu-used =-2
2.2% at --rt --cpu-used =-1
Test on Atom notebook showed only 1.1% speed improvement(speed=-4).
Test on Xeon machine also showed less improvement, since unaligned
data access latency is greatly reduced in newer cores.
Next, I will apply similar idea to other 2 sub-pixel search functions
for encoding speed > 4.
Make this change exclusively for x86 platforms.
Change-Id: Ia7bb9f56169eac0f01009fe2b2f2ab5b61d2eb2f
This is done by expanding luma row to 32-byte alignment, since
there is currently a bunch of code that assumes that
uv_stride == y_stride/2 (see, for example, vp8/common/postproc.c,
common/reconinter.c, common/arm/neon/recon16x16mb_neon.asm,
encoder/temporal_filter.c, and possibly others; I haven't done a
full audit).
It also uses replaces the hardcoded border of 16 in a number of
encoder buffers with VP8BORDERINPIXELS (currently 32), as the
chroma rows start at an offset of border/2.
Together, these two changes have the nice advantage that simply
dumping the frame memory as a contiguous blob produces a valid,
if padded, image.
Change-Id: Iaf5ea722ae5c82d5daa50f6e2dade9de753f1003
allowing the compiler to inline this function. For real-time
encodes, this gave a boost of 1% to 2.5%, depending on the
speed setting.
Change-Id: I3929d176cca086b4261267b848419d5bcff21c02
This patch attempts to improve the handling of CBR streams with
respect to the short term buffering requirements. The "buffer level"
is changed to be an average over the rc buffer, rather than a long
running average. Overshoot is also tracked over the same interval
and the golden frame targets suppressed accordingly to correct for
overly aggressive boosting.
Testing shows that this is fairly consistently positive in one
metric or another -- some clips that show significant decreases
in quality have better buffering characteristics, others show
improvenents in both.
Change-Id: I924c89aa9bdb210271f2e03311e63de3f1f8f920
Optimized C-code of the following functions:
- vp8_tokenize_mb
- tokenize1st_order_b
- tokenize2nd_order_b
Gives ~1-5% speed-up for RT encoding on Cortex-A8/A9
depending on encoding parameters.
Change-Id: I6be86104a589a06dcbc9ed3318e8bf264ef4176c
Do mvp clamping in full-pixel precision instead of 1/8-pixel
precision to avoid error caused by right shifting operation.
Also, further fixed the motion vector limit calculation in change:
b748045470
Change-Id: Ied88a4f7ddfb0476eb9f7afc6ceeddbf209fffd7
Separate simple filter with reduced no. of parameters.
MB filter level picking based on precalculated table. Level table updated for
each frame. Inside and edge limits precalculated and updated just when
sharpness changes. HEV threshhold is constant.
ARM targets use scalars and others vectors.
Change works only with --target=generic-gnu
All other targets have to be updated!
Change-Id: I6b73aca6b525075b20129a371699b2561bd4d51c
Allow the encoder to inform the application that the encoded frame will not
be used as a reference.
Change-Id: I90e41962325ef73d44da03327deb340d6f7f4860
Motion vector limits are calculated using right shifts, which
could give wrong results for negative numbers. James Berry's
test on one clip showed encoder produced some artifacts. This
change fixed that.
Change-Id: I035fc02280b10455b7f6eb388f7c2e33b796b018
In this commit I have added an experimental function
that tests prediction quality either side of a central position
to calculate a suggested boost number for an ARF frame.
The function is passed an offset from the current position and
a number of frames to search forwards and backwards.
It returns a forward, backward and compound boost number.
The new code can be deactivated using #define NEW_BOOST 0
In its current default state the code searches forwards and backwards
from the proposed position of the next alt ref.
The the old code used a boost number calculated by scanning forward
from the previous GF up to the proposed alt ref frame position.
I have also added some code to try and prevent placement of a gf/arf
where there is a brief flash.
Change-Id: I98af789a5181148659f10dd5dd2ff2d4250cd51c
