Re: Problems with Xinerama using a Matrox Millenium / Creator 3D with an Ultra 30.

[Martin <inkubus@interalpha.co.uk>]

On Mon, 2005-03-21 at 01:18, Detlef wrote:
> I dug a bit deeper and xdpyinfo shows a slight difference between the
> visuals regarding the RGB masks and the largest cursor (see the partly cut
> output of xdpyinfo below). I suspect the RGB masks the reason for the
> screens being incompatible and have posted this to the X-Newsgroups but
> haven't received any answer (yet). 
> 
> Matrox: 
<snip>
>   largest cursor:    1280x1024
<snip>
>     red, green, blue masks:    0xff0000, 0xff00, 0xff
<snip>
> Creator 3D
<snip>
>   largest cursor:    64x64
<snip>
>     red, green, blue masks:    0xff, 0xff00, 0xff0000
>     significant bits in color specification:    8 bits
At a guess I'd say the cursor issues are solveable but that the colour
map ones aren't, without a lot of hacking - or perhaps being red/green
colourblind.

> > It's good to know it works, esp. that you can get X running on a Sun box
> > with a non Sun graphics card - very encouraging.
> The only writeoff is that no real acceleration is used, neither on the the
> Creator 3d nor on the Millenium. 
Unless you are planning lots of 3D or video work I wouldn't have thought
it was a major issue.

> In addition the Creator likes only working in 24bit mode although xpdyinfo
> reports all other depths as well. 
> Other than that it works now fine in double head mode (without Xinerama). 
Cool.

> Now I run it deliberately in 24bit on the Creator and 16 Bit on the Matrox,
> which makes the Matrox slightly faster, but still not as fast as the C-3D.
> 
> I wonder if the Matrox driver could come with the next debian Kernel as a
> module. 
Ideally it would be useful to have all of the FB drivers built as (at
least) modules as I suspect this approach may work with other PCI
graphics cards.

Have you attempted the same set up with X.org?

Cheers,
 - Martin

-- 
Martin
inkubus@interalpha.co.uk
"Seasons change, things come to pass"