[PATCH 2/2] glamor: Use float matrix for render transformations

[Uli Schlachter]

Hi,

On 19.08.2014 00:16, Keith Packard wrote:
> walter harms <wharms at bfs.de> writes:
>> So it would be better to use the 16.16 format for transfer and use double
>> for internal calculations ?
> 
> No, the 16.16 format is fundamentally inappropriate in the
> representation of transformation matrices.

Then how about a 16.32 or 16.48 fixed point format? Such a format wouldn't lose
precision with higher numbers. Perhaps a 32.32 fixed point format might make
some (marginal) sense, too.

Why would a floating point format be better than extending the fixed point
format on the wire?

Uli
-- 
"In the beginning the Universe was created. This has made a lot of
 people very angry and has been widely regarded as a bad move."