On Sat, 9 Jul 2022, 15:05 Ben Parham, <behfart...@gmail.com> wrote: > Hi again, > Sure, I will do it based on that. Thank you for your help. If I face an > issue I'll contact you again. > I appreciate it. > > On Sat, Jul 9, 2022 at 7:55 PM Dennis Mungai <dmng...@gmail.com> wrote: > > > On Sat, 9 Jul 2022 at 14:40, Ben Parham <behfart...@gmail.com> wrote: > > > > > Thank you, Dennis. It works with 6 in-preset p4. Now I am running one > > > stream, but I will run 5 streams in RTX A2000. Does it support this > > number > > > in all streams? > > > > > > > Hello there, > > That GPU has no limitations on the simultaneous number of NVENC sessions > it > > can handle. > > However, throughput will be significantly impacted by its' low power > > profile limit (70W?) as you add more transcode jobs. > > This is an issue I've observed in production with a similar GPU profile, > > the Tesla T4. > > You may consider tuning these encoder presets to compensate for > throughput > > drops with multiple simultaneous transcode sessions, if real time > encoding > > is a priority to you. > > And on NVDEC: VRAM usage can also be quite steep with multiple sessions. > > Monitor that carefully via nvidia-smi as you add more transcode jobs. >
Sure thing, no worries. > _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
