[Ffmpeg-devel] [RFC] VC1 Transform in AltiVec

Michael Niedermayer michaelni
Tue Jul 18 15:00:35 CEST 2006


On Tue, Jul 18, 2006 at 01:49:51PM +0200, Luca Barbato wrote:
> Guillaume Poirier wrote:
> > 
> > sorry, but I don't understand why your code samples weren't
> > appropriate. Could you please elaborate? I'm sure I could learn from
> > your mistake.
> If the michael code is pseudocode and src[n] means take a vector of
> shorts out that this offset the code is ok, if the code is scalar mine
> is completely wrong...

well, its scalars, it was copied from the c idct, it was just to demonstrate
how to do the transform wih 16bit instead of 32bit variables

but ...
the same transform is applied to src+1, src+2, ..., src+7
so maybe your code isnt completely wrong ...


Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

In the past you could go to a library and read, borrow or copy any book
Today you'd get arrested for mere telling someone where the library is

More information about the ffmpeg-devel mailing list