[Ffmpeg-devel] Re: [RFC] mpeg2 422 encoding.

[Baptiste COUDURIER]

Hi

Michael Niedermayer wrote:
> [...]
> 
> this patch causes a 0.4% speed loss for mpeg4 encoding, this is IMHO not
> acceptable, speed losses add up and if we accept 10 such patches ...
> 
> [...]
>
> things should be implemented as:
> 
> static always_inline some_func_intenal(csx, csy){
>     ...
> }
> 
> some_func(){
>     if(chroma_shift_x==1 && chroma_shift_y==1)
>         some_func_intenal(1, 1);
>     else
>         some_func_intenal(chroma_shift_x, chroma_shift_y);
> }
> 
> that way we could easily switch between a big & fast or small and 0.4%slower
> version by just forcing some_func_intenal to be inlined or not
> 
> [...]
> 

All right, I tried to implement it the way you suggested, I hope I
understood it right.

I still have a bug when I use -mbd bits in 422, and I could not find why
yet, picture has yellow/green/blue chroma blocks on P/B frames.
What does mbd bits imply that is well hiddent ?

Thanks Michael for your time.

-- 
Baptiste COUDURIER                              GnuPG Key Id: 0x5C1ABAAA
SMARTJOG S.A.                                    http://www.smartjog.com
Key fingerprint                 8D77134D20CC9220201FC5DB0AC9325C5C1ABAAA
Phone: +33 1 49966312
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: mpeg2_422.patch
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20060404/231108bd/attachment.txt>