Ok, maybe sws_scale running on 24 CPU cores is faster than this
https://github.com/lemenkov/libyuv on one core.
Seriously, what do you mean by "initialize the context manually".  How to
force it to use SSSE3 instruction and so on?

ср, 14 сент. 2022 г. в 13:25, wolverin via Libav-user <[email protected]
>:

>
>
> I recommend not to use sws_sclale at all. It is a real bottleneck of all
> encoding process.
>
>
>
> If you use the multithreaded version of AVFilter or initialize the context
> manually to use multithreading, then everything works fast enough.
>
> _______________________________________________
> Libav-user mailing list
> [email protected]
> https://ffmpeg.org/mailman/listinfo/libav-user
>
> To unsubscribe, visit link above, or email
> [email protected] with subject "unsubscribe".
>


-- 
С уважением Алексей Комаров.
_______________________________________________
Libav-user mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/libav-user

To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".

Reply via email to