On Sun, 16 Oct 2016, Luca Barbato wrote:
On 16/10/2016 22:18, Martin Storsjö wrote:
Now the comparison to libvpx is much more close; we're rarely slower
at all, and even much faster in some cases.
Probably you could update the statement in the commit and push it then =)
I've already updated it actually, although it's all pretty hard to notice
the nuances.
Originally it was:
---8<---
The libvpx functions are marginally faster on A7, A8 and A9, while
our is marginally faster on A53 (which is where it has been tuned
during development). Overall they are pretty similar.
---8<---
Then after some tuning, my verdict was:
---8<---
Our version is consistently marginally faster on A53, consistently
marginally slower on A9, and faster in some tests but slower in some,
on A7 and A8.
---8<---
And now it is:
---8<---
Our version is consistently faster on on A7 and A53, consistently
marginally slower on A9, and faster in some tests but slower in some,
on A7 and A8.
---8<---
// Martin
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
