Releases: rigaya/QSVEnc
Releases · rigaya/QSVEnc
8.04
- Add support for VapourSynth API V4.
- Add option to encode only when input audio codec differs from codec specified by --audio-codec. (--audio-encode-other-codec-only)
8.03
- Fix --qp-offset not working for AV1. ( #273 )
- Fix error when encoding H.264 to RTMP/FLV output.
- Fix mkv not created when encoding with -c av_libsvtav1.
- Work around OpenCL driver issue on Linux Intel iGPU where read_imageui caused "undefined reference to __spirv_ImageSampleExplicitLod_Ruint4".
- Fix build error on Arch Linux. ( #276 )
8.02
- Add option to fallback to 8bit encoding when 10bit encoding is not supported by the hardware.(--fallback-bitdepth)
8.01
- Avoid unintended fps values when front of input file is corrupted.
- Improve handling when input files have negative pts.
- Improve quality of burned in subtitles in --vpp-subburn processing by changing libass initialization method.
- Improve progress indicator when using --parallel.
- Previously was only showing the speed of first thread, now will show speed of all threads.
- Add support for using --parallel with multiple pipes.
- Update libass.dll
- libass 0.9.0 -> 0.17.4 (x64), 0.14.0 (x86)
- harfbuzz 11.4.4 (new)
- libunibreak 6.1 (new)
- Update rpm build environment to fedora41. ( #269 )
8.00
- Update libvpl to 2.15.
- Add options for mfx_ai_superres in --vpp-resize. (superres-mode, superres-algo)
Note: superres-mode does not seem to work at least at 6913 driver.
- Add options for mfx_ai_superres in --vpp-resize. (superres-mode, superres-algo)
- Add QSVEnc.auo2 as new native AviUtl2 plugin.
- Add feature to use filters with avcodec encoders (mainly for my test purpose).
- Available with
-c av_xxx, with option to set encoder options (--avcodec-prms).
Example: -c av_libsvtav1 --avcodec-prms preset=6,crf=30,svtav1-params=enable-variance-boost=1:variance-boost-strength=2"
Other codec options include av_libvvenc, av_libvpx-vp9, etc.
- Available with
- Update ffmpeg libraries. (Windows)
- ffmpeg 7.1+ (20240822) -> 8.0
- libpng 1.6.44 -> 1.6.50
- expat 2.6.2 -> 2.7.1
- fribidi 1.0.11 -> 1.0.16
- libogg 1.3.5 -> 1.3.6
- libxml2 2.12.6 -> 2.14.5
- libvpl 2.13.0 -> 2.15.0
- libvpx 1.14.1 -> 1.15.2
- dav1d 1.4.3 -> 1.5.1
- libxxhash 0.8.2 -> 0.8.3
- glslang 15.0.0 -> 15.4.0
- dovi_tool 2.1.2 -> 2.3.1
- libjpeg-turbo 2.1.0 -> 3.1.1
- lcms2 2.16 -> 2.17
- zimg 3.0.5 -> 3.0.6
- libplacebo 7.349.0 -> 7.351.0
- libsvtav1 3.1.0 (new!) x64 only
- libvvenc 1.13.1 (new!) x64 only
- Remove mmt/tlv patch
7.94
7.93
7.92
7.91
- Fix 10-bit processing in --vpp-afs.
- Improve precision of --vpp-afs for both 8bit and 10bit.
- Add option to explicitly specify field pattern in --vpp-deinterlace.
- Fix incorrect frame rate when outputting in y4m format with raw output.
- Fix processing sometimes stopping during raw output.
- Add support for --option-file on Linux.
- Fix handling when "end" is omitted in --dynamic-rc. ( #259 )
7.90
- Add support for combining --output-format with
-c raw. ( #257 )
Now supports cases like-c raw --output-format nut. - Fix black/white processing in 10-bit depth for --vpp-edgelevel.
- Improve interlace detection when using --avsw.