On 06. 06. 2018 17:25, Dennis Mungai wrote: > And with CUVID, in particular: It's decoders are very fragile. > > You did mention that disabling hwaccel (cuvid?) raised your processor load > significantly, correct? > > Can you confirm whether you're applying any filters down the encoder > pipeline? > Also, what processor are you using?
Yes, the load was raised by 100%, not that is really huge (from 6% - 9% to 12% - 15%). But on 29 running ffmpeg instances it matters. Thou I am still trying to understand where the bottleneck is when I try to pass more than 32+ instances, but that is a matter for a new subject. Dependent on a stream I am "burning" subtitles into a video and mapping out only those streams that I am interested in, so one audio + one video (optionally burned subtitles). Everything else (deinterlace) is done in decoding stage. CPU: [ 0.074709] smpboot: CPU0: Intel(R) Core(TM) i7-4770S CPU @ 3.10GHz (family: 0x6, model: 0x3c, stepping: 0x3) Br, Mitja _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".
