Premultiplied vs non-premultiplied colors in solid and gradient pictures

[Keith Packard]

On Fri, 2005-07-01 at 10:04 +0200, Lars Knoll wrote:

> As I said, I can readd the checks without any problems, but the compositiong 
> operations will be a factor of 3 slower; at least on x86. Together with exa 
> the difference will probably be noticable.

Wrong answers fast aren't that interesting; and these errors are not
minor, rather you get black where you wanted white in some cases. You
don't have to fall back to the really ugly slow code, you can add
saturation to your code; that may well double the size of the code, but
would then generate correct results.

Besides, what we really want is a pure MMX/SSE implementation anyway;
the plain C coded version is not that important in the long run.

-keith

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20050701/d02bad94/attachment.pgp>