4.8 KiB
Versatile Video Codec binaries
Versatile Video Codec binaries has available two tools for you to choose.
Fraunhofer HHI (vvenc/vvdec)
Fraunhofer HHI VVC binaries is just standard easy tool, which allows to encode & decode with arguments and also encodes good quality for slow preset, but it may be slow...
- C++14 programming language
- Supports Windows, macOS, Linux, BSD, Android, iOS (library only with new app or non-signed binary executable on jailbroken device) & WebAssembly.
- SIMDE: SCALAR, SSE4.1, SSE4.2, AVX, AVX2, AVX512 (unsupported yet), NEON (arm only) and WASM (Web browsers only).
- Faster depends speeding up for newer computer machine of CPUs.
- Presets can be used from faster to slower.
- Includes advanced tool of encode (vvencFFapp)
- Two-pass control of bitrate is only allowed.
- Default encoding pixel format is 10-bit.
- For decoding, it can output to YUV, Y4M or pipe.
- Lossless encoding (only for vvencFFapp with CostMode argument)
Old screenshots in 2021 of three desktop operating systems:
Windows:
Mac:
Linux:
uvg266 (VVC Scalable encoder tool)
uvg266 is only scalable encoder which helps to make it faster for older & newer computers, but still a bit good quality than vvenc, seems okay.
- C99 programming language
- Supports Windows, macOS, Linux, BSD, Android & iOS (library only with new app or non-signed binary executable on jailbroken device).
- SIMDE: MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2, NEON & AltiVec (PowerPC only).
- Can only encode 8-bit VVC output for rest operating system supported, but for 10-bit needs compiled with definition, this only works for Android users.
- Faster encoding scalable of old & new computers than vvenc.
- Supports presets from ultrafast to placebo like x264 presets, however it is recommended to use from faster to slower to avoid VVC decoding fail issue.
- Lossless encoding is not recommended due decoding fail.
Minimum requirements
vvenc/vvdec:
- Windows XP and later (requires Visual C++ Redistributable for Visual Studio 2019 installed)
- macOS 10.13 and later (arm64 since macOS 11.0)
- Linux kernel 3.2.0 and later (eg. Ubuntu 12.04 LTS and later)
- BSD (FreeBSD & x64 only)
- Android 4.1 (API 16, Jelly Bean) and later (arm64 & x86_64 since Android 5.0)
uvg266:
- Windows Vista and later (requires Visual C++ Redistributable)
- macOS 10.13 and later (arm64 since macOS 11.0)
- Linux kernel 3.2.0 and later (eg. Ubuntu 12.04 LTS and later)
- BSD (FreeBSD & x64 only)
- Android 4.3 (API 18, Jelly Bean) and later (arm64 & x86_64 since Android 5.0)
Downloads
NOTE: uvg266 compiled by same architectures supported of vvenc & vvdec.
All compiled builds are compressed files on 7-Zip.
| OS | vvenc & vvdec | uvg266 |
|---|---|---|
| Windows | x64/x86 | Download |
| macOS | Arm64/x64 | Download |
| Linux | x86_64/x86/arm64/armv7a | Download |
| BSD | x86_64 | Download |
| Android | x86_64/x86/arm64/armv7a | Download or 10-bit build |
Comparisions between vvenc/vvdec and uvg266
If you're using newer computer or/and you wanted it to encode 10-bit with fast encoding to get good quality, use vvenc recommended.
If you're using older computer or/and more faster encoding of 8-bit than vvenc, use uvg266 recommended.
Also for decoding, use vvdec and also it is faster decoding for older (possible) & newer computers.