[I sent this message yesterday but it hasn't appeared on the mailing list and
I haven't received any notice about it being held or anything. Sorry if the
other eventually shows up and this is a duplicate]

Marc Aurele La France <[EMAIL PROTECTED]> writes:

> The log shows you are only using the Matrox adapter.  So, a few questions
> arise:
> 
> - Do you see the same problems using the Mach64 GT?

Hm, so now I can't reproduce the mouse hotpoint out of sync problem on the
Matrox either. I think I was being fooled by a window manager that was
compressing mouse movements and sometimes didn't end up in the correct place.
Using the gimp to locate the hot point the hot point always seems to be in
sync.

The panning is still jerky at 4 pixels per pan, that's still annoying. But
it's not a correctness issue I guess. Any chance of going back to the X 3
method of panning that allowed arbitrary offsets?

The displaced pixels showing up at the right edge of the screen is still there
though. It only happens on the Matrox card, not the ATI.

Some more data: it seems to happen when the screen is panned to any offset
congruent to 4 mod 64. 

For example, when panned to column 4 (with the first column being 0) the
pixels starting at column 128-131 appear in the rightmost four columns of the
screen. When panned to 68 columns 192-195 appear in the rightmost column. This
is with the virtual frame buffer at 1600x1200 and the actual screen resolution
set to 640x480.

The mouse pointer does show up on the duplicated pixels at the right edge when
it passes through those four pixels in the middle.

-- 
greg

_______________________________________________
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert

Reply via email to