Olaf Hering (olh)
accepted
request 5398
over 3 years ago
(revision 41)
Build 10 and 12bit libraries unconditionally. (forwarded request 5396 from Aloysius)
Olaf Hering (olh)
accepted
request 5395
over 3 years ago
(revision 40)
- Update to version 3.5
New features:
* Real-time VBV for ABR (Average BitRate) encodes in –pass 2
using --vbv-live-multi-pass. Improves VBV compliance with no
significant impact on coding efficiency.
Enhancements to existing features:
* Improved hist-based scene cut algorithm: Reduces false
positives by leveraging motion and scene transition info.
* Support for RADL pictures at IDR scene cuts: Improves coding
efficiency with no significant impact on performance.
* Bidirectional scene cut aware Frame Quantizer Selection:
Saves bits than forward masking with no noticeable perceptual
quality difference.
API changes:
* Additions to x265_param structure to support the newly added
features and encoder enhancements.
* New x265_param options --min-vbv-fullness and
--max-vbv-fullness to control min and max VBV fullness.
Bug fixes:
* Incorrect VBV lookahead in --analysis-load + --scale-factor.
* Encoder hang when VBV is used with slices.
* QP spikes in the row-level VBV rate-control when WPP enabled.
* Encoder crash in --abr-ladder.
- Use new homepage and download URLs.
- Add subpackage for HDR10+ library
- Add update.sh
Olaf Hering (olh)
accepted
request 5114
over 4 years ago
(revision 39)
- Update to version 3.4
New features:
* Edge-aware quadtree partitioning to terminate CU depth
recursion based on edge information. --rskip level 2 enables
the feature and --rskip-edge-threshold denotes the minimum
expected edge-density percentage within the CU, below which
the recursion is skipped. Experimental feature.
* Application-level feature --abr-ladder for automating
efficient ABR ladder generation. Shows ~65% savings in the
over-all turn-around time required for the generation of a
typical Apple HLS ladder in Intel(R) Xeon(R) Platinum 8280
CPU @ 2.70GHz over a sequential ABR-ladder generation
approach that leverages save-load architecture.
Enhancements to existing features:
* Improved efficiency in 2-pass rate-control algorithm. The
savings in the bitrate is ~1.72% with visual improvement in
quality in the initial 1-2 secs.
Encoder enhancements:
* Faster ARM64 encodes enabled by ASM contributions from
Huawei. The speed-up over no-asm version for 1080p encodes @
medium preset is ~15% in a 16 core H/W.
* Strict VBV conformance in zone encoding.
Bug fixes:
* Multi-pass encode failures with --frame-dup.
* Corrupted bitstreams with --hist-scenecut when input depth
and internal bit-depth differ.
* Incorrect analysis propagation in multi-level save-load
architecture.
* Failure in detecting NUMA packages installed in non-standard
directories.
Olaf Hering (olh)
accepted
request 5053
over 4 years ago
(revision 38)
- Update to version 3.3
New features:
* Adaptive frame duplication** to identify and skip encoding
of near-identical frames and signal the duplication info to the
decoder via pic_struct SEI. :option:`frame-dup` to enable frame
duplication and :option:`--dup-threshold` to set the threshold
for frame similarity (optional).
* Boundary aware quantization** to cut off bits from frames
following scene-cut. This leverages the inability of HVS to
perceive fine details during scene changes and saves bits.
:option:`--scenecut-aware-qp` , :option:`--scenecut-window` and
:option:`--max-qp-delta` to enable boundary aware frame
quantization, to set window size (optional) and to set QP offset
(optional).
* Improved scene-cut detection** using edge and chroma
histograms. :option:`--hist-scenecut` to enable the feature and
:option:`--hist-threshold` (optional) to provide threshold for
determining scene cuts.
Enhancements to existing features:
* :option:`--hme-range` to modify search range for HME levels
L0, L1, and L2.
* Improved performance of AQ mode 4 by reducing memory foot
print.
* Introduced :option:`--analysis-save-reuse-level` and
:option:`--analysis-load-reuse-level` to de-couple reuse levels
of :option:`--analysis-save` and :option:`--analysis-load`.
Turnaround time of ABR encoding can be reduced by properly
leveraging these options.
Encoder enhancements:
* Improved VBV lookahead to eliminate blocky artifacts in
Olaf Hering (olh)
committed
almost 5 years ago
(revision 36)
remove weird Requires: buildrequires, rejected by rpm 4.15
devel pkg requires a specific lib version
Olaf Hering (olh)
accepted
request 4640
about 5 years ago
(revision 35)
- Update to version 3.2.1
* Fix output change in analysis load
* Fix encoder crash with zones and add test for zones
* Fix: Performance drop in aq-mode 4 This patch moves the
memory handling part of the edge information required for
aq-mode 4 to the Frame class-level in that way it can be
reused by the threads.
* Fix: Performance drop in aq-mode 4 This patch moves the
memory handling part of the edge information required for
aq-mode 4 to the Frame class-level in that way it can be
reused by the threads.
* Fix output change observed during analysis-load for
inter-refine levels 2 and 3.
* Adaptive Frame duplication This patch does the following. 1.
Replaces 2-3 near-identical frames with one frame and sets
pic_struct based on frame doubling / tripling. 2. Add option
"--frame-dup" and "--dup-threshold' to enable frame
duplication and to set threshold for frame similarity
(optional).
* Fix: AQ mode 4 commit (21db162) introduces slowdown even is
not used AQ mode 4.
enzokiel
accepted
request 4371
almost 6 years ago
(revision 31)
- Support 10 and 12 bit color depths
- Update to version 3.0
New features
* option:: '--dolby-vision-profile <integer|float>' generates
bitstreams confirming to the specified Dolby Vision profile.
Currently profile 5, profile 8.1 and profile 8.2 enabled,
Default 0 (disabled)
* option:: '--dolby-vision-rpu' File containing Dolby Vision RPU
metadata. If given, x265's Dolby Vision metadata parser will
fill the RPU field of input pictures with the metadata read
from the file. The library will interleave access units with
RPUs in the bitstream. Default NULL (disabled).
* option:: '--zonefile <filename>' specifies a text file which
contains the boundaries of the zones where each of zones are
configurable.
* option:: '--qp-adaptation-range' Delta-QP range by QP
adaptation based on a psycho-visual model. Default 1.0.
* option:: '--refine-ctu-distortion <0/1>' store/normalize ctu
distortion in analysis-save/load. Default 0.
* Experimental feature option:: '--hevc-aq' enables adaptive
quantization. It scales the quantization step size according
to the spatial activity of one coding unit relative to frame
average spatial activity. This AQ method utilizes the minimum
variance of sub-unit in each coding unit to represent the
coding unit's spatial complexity.
Encoder enhancements
* Preset: change param defaults for veryslow and slower preset.
Replace slower preset with defaults used in veryslow preset
and change param defaults in veryslow preset as per
experimental results.
Olaf Hering (olh)
committed
almost 7 years ago
(revision 27)
v2.7 fails to build in armv7l
Malcolm Lewis (malcolmlewis)
accepted
request 4239
over 6 years ago
(revision 29)