Add hardware video codec support info to README.md.#652
Conversation
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
xianshijing-lk
left a comment
xianshijing-lk
left a comment
There was a problem hiding this comment.
some nits. please address them
|
|
||
| ## Hardware video codec support | ||
|
|
||
| The underlying Rust SDK ships with platform-specific hardware-accelerated encoders/decoders, which are enabled automatically when the requested codec and OS match. |
There was a problem hiding this comment.
nit,
These are used automatically when available and compatible with the runtime environment (OS, drivers, GPU, and codec).
| | macOS | H264, H265 | ✓ | ✓ | VideoToolbox | | ||
| | Linux (AMD GPU) | H264 | ✓ | | VAAPI | | ||
| | Linux x64 (NVIDIA GPU) | H264, H265 | ✓ | ✓ | NVENC / NVDEC (NVIDIA Video Codec SDK) | | ||
| | Linux ARM (NVIDIA Jetson) | H264, H265 | ? | ? | NVENC / NVDEC (to be confirmed) | |
There was a problem hiding this comment.
what it means by ?
do you mean TBD ? if not sure, should we just remove Linux ARM for now ?
| | Linux x64 (NVIDIA GPU) | H264, H265 | ✓ | ✓ | NVENC / NVDEC (NVIDIA Video Codec SDK) | | ||
| | Linux ARM (NVIDIA Jetson) | H264, H265 | ? | ? | NVENC / NVDEC (to be confirmed) | | ||
|
|
||
| Software encoders (libvpx for VP8/VP9, libaom for AV1, OpenH264 for H264) are used as a fallback on platforms or codecs not listed above. |
There was a problem hiding this comment.
are used as a fallback when hardware acceleration is not available.
|
|
||
| Software encoders (libvpx for VP8/VP9, libaom for AV1, OpenH264 for H264) are used as a fallback on platforms or codecs not listed above. | ||
|
|
||
| > **Note:** NVIDIA data-center / AI-inference GPUs such as the **H100, H200, A100** physically omit the NVENC/NVDEC engines — running on these cards will silently fall back to the software encoder/decoder regardless of the table above. |
There was a problem hiding this comment.
nit, maybe
Note: Availability depends on the specific machine configuration, including GPU model, driver support, and runtime environment.
Note: NVIDIA data-center / AI-inference GPUs such as the H100, H200, A100 do not include NVENC/NVDEC engines. On these GPUs, encoding and decoding will fall back to software implementations.
2 participants
No description provided.