Q1, This is compatible with LGPL.
Q2, You need to discuss with FFmpeg group. This is not about your work,
it's about ownership of intellectual property and distribution.
The real problem is not about work itself, not on invocation of function
these details, but distribution of your final product.
Have you benchmark the ffmpeg during the runtime to check where is the
bottleneck ?
On Thu, Oct 10, 2019 at 10:11 AM Karan Singh
wrote:
> Hi Team
>
> Any update on this?
>
>
> i am struggling with live streaming of 4k videos with or without GPU.
>
> My question before describing my issue is what
You can get more information here.
https://developer.nvidia.com/nvidia-video-codec-sdk
NVIDIA GPU works with the GPU newer than Kepler, MX150 is too old to
support NVENC, the only way is DXVA.
NVIDIA VideoCodec only supports MPEG1/2 H.264, HEVC and VP9, also the
ffmpeg must be built with the NVE
ata.3!)
> .retain()
> .takeRetainedValue()
>
> In C, you would do something like:
>
> CVPixelBufferRef pixelBuffer =
> CVPixelBufferRetain((CVPixelBufferRef)frame->data[3]);
>
> Cheers!
> fumoboy007
>
> On Sat Aug 18 05:50:45 EEST 2018, Bo Zhou
> wrote:
>
Hi !
I'm using the FFmpeg 4.0.2 and successfully executed the example hw_decode
on Mac. And I could see the trace log
[h264 @ 0x7f97d980] Format videotoolbox_vld chosen by get_format().
[h264 @ 0x7f97d980] Format videotoolbox_vld requires hwaccel
initialisation.
[h264 @ 0x7f97d980] R
