[FFmpeg-devel] [PATCH] mmx implementation of vc-1 inverse transformations

Michael Niedermayer michaelni
Tue Aug 26 03:46:24 CEST 2008

On Tue, Aug 26, 2008 at 02:05:49AM +0100, M?ns Rullg?rd wrote:
> Michael Niedermayer <michaelni at gmx.at> writes:
> > Also the 8x8 transform either has to be optimal or removed from the patch.
> How do you know when something is optimal?

I do not, but when i find something suboptimal then i know it is not.

So taken litterally i surely have not said it well at all ...
somewhat better would have been saying:
It must not contain any suboptimality that will be found in a review,
and of course it should not contain any known suboptimalities even if
we miss such suboptimalities.
And even better would be to limit that to suboptimalities that can be
fixed in reality (that is not just ones we could proof to exist but ones
that could be fixed by a finite number of known/calculateable steps).

Iam not sure if i have missed any further holes, but i hope not :)

Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Its not that you shouldnt use gotos but rather that you should write
readable code and code with gotos often but not always is less readable
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080826/dff3940a/attachment.pgp>

More information about the ffmpeg-devel mailing list