2012/2/28 Mashiat Sarker Shakkhar <[email protected]>:
> The code is not really as optimized as it could be, because optimizing
> it wasn't really the first priority and I am not terribly experienced
> at that.

This was in no way a criticism. On the contrary, I only reacted
because it was already planned to look at this, and I did.

> Intermediate operations require 32-bit integer according to the spec,
> (that is, Pred is int32_t). Ref-code uses 64-bit for some reason that
> I don't understand.

I haven't looked a lot at that. If intermediates really are 64bits in
some cases, then there's no avoiding it. In particular with 8 more
bits in samples and coefficients not specifically adapted, it's quite
possible some intermediate results need more than 32bits.

> I don't have experience with SIMD, but will try to write some if that's
> necessary. Priority for now is to implement 24-bit.

I completely agree. The goal was not to overwhelm you.

Christophe
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to