summaryrefslogtreecommitdiff
path: root/vp9/encoder
AgeCommit message (Collapse)Author
2017-02-24Merge "Make vp9_scale_and_extend_frame_ssse3 work for hbd when bitdepth = 8."Jerome Jiang
2017-02-24consolidate block_error functionsJohann
vp9_highbd_block_error_8bit_c was a very simple wrapper around vp9_block_error_c. The SSE2 implemention was practically identical to the non-HBD one. It was missing some minor improvements which only went into the original version. In quick speed tests, the AVX implementation showed minimal improvement over SSE2 when it does not detect overflow. However, when overflow is detected the function is run a second time. The OperationCheck test seems to trigger this case and reverses any speed benefits by running ~60% slower. AVX2 on the other hand is always 30-40% faster. Change-Id: I9fcb9afbcb560f234c7ae1b13ddb69eca3988ba1
2017-02-24Merge "block error sse2: use tran_low_t"Johann Koenig
2017-02-23Make vp9_scale_and_extend_frame_ssse3 work for hbd when bitdepth = 8.Jerome Jiang
Only works for bitdepth = 8 when compiled with high bitdepth flag. 4x speed ups for handling 1:2 down/upsampling. Validated manually for: 1) Dynamic resize for a single layer encoding 2) SVC encoding with 3 spatial layers Results are bitexact with the patch and the speed gain (~4x) in the scaling was verified. BUG=webm:1371 Change-Id: I1bdb5f4d4bd0df67763fc271b6aa355e60f34712
2017-02-24block error sse2: use tran_low_tJohann
Change-Id: Ib04990e4a7bda9fbf501f294da2057a2b2595deb
2017-02-23Merge "vp9: 1pass CBR: modify condition for reducing loop filter."Marco Paniconi
2017-02-22Merge "vp9: Non-rd pickmode: use simple block_yrd under some conditons."Jerome Jiang
2017-02-22vp9: 1pass CBR: modify condition for reducing loop filter.Marco
The reduction showed improvement on RTC when aq-mode=3 is on. Add that (cyclic refresh enabled) to the condition. Only affects 1 pass CBR. Change-Id: I5d0843002d8e31d7c165098a62e7a71146b08664
2017-02-22vp9: Non-rd pickmode: use simple block_yrd under some conditons.Marco
For speed 8 only. 3% speed up for QVGA and 6.3% for VGA on Nexus 6. ~3% avgPSNR decrease on rtc_derf and 2.9% on rtc. Disabled for now. Change-Id: I70133f1f6c804d663d594df437bfe7fdb0030d6a
2017-02-22Merge "vp9: aq-mode=3: On key frame reset cr->reduce_refresh to 0."Marco Paniconi
2017-02-22vp9: aq-mode=3: On key frame reset cr->reduce_refresh to 0.Marco
This prevent possible reduction of cyclic refresh after key frame. Change-Id: Idd4e49b69cd95476e7eccfa31b2bd8669569e9e8
2017-02-22vp9: Only compute y_sad for golden in variance partition for speed < 8.Jerome Jiang
Only affects speed 8. No obvious quality regression. Systematic speed ups by ~1% on Nexus 6. Change-Id: Ia904ca28ea041c3281c532911ec38fb7d7f46a17
2017-02-22Merge "Refactored the row based multi-threading code"Yunqing Wang
2017-02-22Merge "Fix segmentation fault caused by denoiser working with spatial SVC."Jerome Jiang
2017-02-21vp9: Incorporate source sum_diff into non-rd partition thresholds.Marco
Increase the variance partition thresholds for superblocks that have low sum-diff (from source analysis prior to encoding frame). Use it for now only for speed >= 7 or for denoising on. Small change on metrics for rtc set: less than ~0.1 avgPNSR decrease on RTC set, for both speed 7 and 8. Change-Id: I38325046ebd5f371f51d6e91233d68ff73561af1
2017-02-21Merge "Drop zbin_ptr and quant_shift_ptr"Johann Koenig
2017-02-21Fix segmentation fault caused by denoiser working with spatial SVC.Jerome Jiang
Re-enable the affected test. BUG=webm:1374 Change-Id: I98cd49403927123546d1d0056660b98c9cb8babb
2017-02-21Merge "Change to prediction decay calculation."Paul Wilkins
2017-02-20vp9: Fix for non-rd pickmode for high-bitdepth build.Marco
Use the simple block_yrd under certain conditions. The optimization code is completed but the speed is still slower (~6% on 720p) than the low-bitdepth build. For now, use the more complex block_yrd under certain conditions (always use it for speed <= 5, otherwise use it on key frames and for bsize >= 32x32). This gives about ~2-3% gain in quality for speed 7 on RTC set (over high bitdepth build), with about the same encoder fps as the low bitdepth build. Change-Id: Ibe92a1945d0bd635f880befb4c815727df62d754
2017-02-20Refactored the row based multi-threading codeRanjit Kumar Tulabandu
Modified the code to facilitate bit-match tests in first pass Added unit-tests to test the row based multi-threading behavior for bit-exactness Change-Id: Ieaf6a8f935bb1075597e0a3b52d9989c8546d7df
2017-02-17Change to prediction decay calculation.paulwilkins
This change subtracts out low complexity intra regions that are also low error in the inter domain, in the calculation of the frame prediction decay. The rationale here his that low complexity regions (such as sky) do not imply high prediction decay in the same way as high error intra or neutral blocks. The effect of this is small in most clips but in a few clips it can be > 10%. (E.g. In to tree) Change-Id: If67ac23d17fca14285cad2defa464c61c9ea861c
2017-02-17Merge "cosmetics: Fix spelling mistake in compile flag name."James Zern
2017-02-16Merge "block error avx2: use tran_low_t"Johann Koenig
2017-02-16cosmetics: Fix spelling mistake in compile flag name.paulwilkins
agressive -> aggressive after: ce7b38459 Aggressive VBR method. Change-Id: Ie0f30b1bbc77ed9f32bec047b4a9b3d0cf4853f5
2017-02-16Drop zbin_ptr and quant_shift_ptrJohann
vp9[_highbd]_quantize]_fp[_32x32] and vp9_fdct8x8_quant do not make use of these parameters. scan is used for C code and iscan is used for SIMD implementations. Change-Id: I908a0ff7d3febac33da97e0596e040ec7bc18ca5
2017-02-16block error avx2: use tran_low_tJohann
Change-Id: Ic5f3a1f569d6f82afeaf4fcd7235374bb460db3c
2017-02-16Merge changes I267050a5,Iebade0ef,Id96a8df3Johann Koenig
* changes: quantize_fp_32x32 highbd ssse3: enable existing function quantize_fp highbd ssse3: use tran_low_t for coeff quantize_fp highbd sse2: use tran_low_t for coeff
2017-02-16Merge "Structured the mode ordering code to avoid redundant memcpy"Yunqing Wang
2017-02-16quantize_fp highbd ssse3: use tran_low_t for coeffJohann
Change-Id: Iebade0efc0efbb0a80a0f3adbef4962e3a2f25e8
2017-02-16quantize_fp highbd sse2: use tran_low_t for coeffJohann
Change-Id: Id96a8df33354a7987ce890a3d6798c7375ffa4aa
2017-02-16bitdepth conversion: really use num elementsJohann
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
2017-02-16Structured the mode ordering code to avoid redundant memcpyRanjit Kumar Tulabandu
Change-Id: I4f5d6b54018bd1928cd9e5e42619e6f55b334803
2017-02-16Merge "Disconnect ARF breakout from frame boost."Paul Wilkins
2017-02-16Merge "Remove unnecessary factor."Paul Wilkins
2017-02-16Merge "Bug in scale_sse_threshold()"Paul Wilkins
2017-02-16Merge "Additional first pass stats."Paul Wilkins
2017-02-15vp9: Some code cleanup for aq-mode = 3.Marco
The weight segment needs to only be computed once per frame, so remove it from the funciton vp9_cyclic_refresh_rc_bits_per_mb(), which is called within a loop inside vp9_rc_regulate_q. Change-Id: Ia0e18b89abb97e42c466d4dbc47700d7f76555db
2017-02-15vp9. Use same source_sad threshold for all speeds.Marco
Only affects real-time mode. Change-Id: Iba836f110c4da936f5173cc0f54424d5b6121bff
2017-02-15Vp9: Speed 8 aq-mode=3: Reduce computation in estimating bits per mb.Marco
vp9_compute_qdelta_by_rate has almost 2% overhead in profiling on Nexus 6. Reduce the calling of that function in speed 8 by estimating the delta-q. Both rtc and rtc_derf show little/no change in avg psnr/ssim. Encoding speed is 2~3% faster on Nexus 6. Change-Id: If25933715783f31104a18a5092ea347b1221b5f5
2017-02-15Disconnect ARF breakout from frame boost.paulwilkins
This small change replaces the frame boost check in the arf group length break out clause with a test against a prediction decay value. The boost value is in fact partly dependent on the decay value but this change means that the per frame boost calculation can be adjusted without influencing the group length calculation. The value chosen gives a close match on all the test sets with the previous code (on average) but it was noted that a lower threshold was slightly better for 1080P and up and a slightly higher value for small image sizes. Change-Id: I4d5b9f67d5b17b0d99ea3f796d3d6202fd61ee0c
2017-02-15Remove unnecessary factor.paulwilkins
Removed unnecessary scaling factor to simplify. Change-Id: I3fc9c5975a2597e72f1324e09dd586dea1facfa7
2017-02-15Bug in scale_sse_threshold()paulwilkins
The function scale_sse_threshold() returns a threshold scaled if necessary for use with 10 and 12 bit from an 8 bit baseline. SSE error values would be expected to rise for the 10 and 12 bit cases where there are more bits of precision. Hence the threshold used for the test should also be scaled up. Change-Id: I4009c98b6eecd1bf64c3c38aaa56598e0136b03d
2017-02-15Additional first pass stats.paulwilkins
Added counts that split the intra coded blocks into low and high variance. Change-Id: Ic540144b34d5141659081bb22f7ee16fd6861f14
2017-02-15Merge "Aggressive VBR method."Paul Wilkins
2017-02-15Merge "vp9 fdct higbd neon: connect existing highbd calls"Johann Koenig
2017-02-15Merge "Row based multi-threading of encoding stage"Yunqing Wang
2017-02-15Row based multi-threading of encoding stageRanjit Kumar Tulabandu
(Yunqing Wang) This patch implements the row-based multi-threading within tiles in the encoding pass, and substantially speeds up the multi-threaded encoder in VP9. Speed tests at speed 1 on STDHD(using 4 tiles) set show that the average speedups of the encoding pass(second pass in the 2-pass encoding) is 7% while using 2 threads, 16% while using 4 threads, 85% while using 8 threads, and 116% while using 16 threads. Change-Id: I12e41dbc171951958af9e6d098efd6e2c82827de
2017-02-14vp9 fdct higbd neon: connect existing highbd callsJohann
Change-Id: Ia8f822bd6e70b3911bc433a5a750bfb6f9a3a75c
2017-02-14Merge "quantize_fp highbd neon: use tran_low_t for coeff"Johann Koenig
2017-02-14apply clang-formatclang-format
Change-Id: I75e4a9e0b37bd4586f26c8d6c1fa27f3f6ff1bce
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-10 09:31:59 +0000