I didn't have time to fool with the test day for video, but I have finally gotten around to checking out my personal collection of ATI cards with latest fedora 13 updates, with these results:
RV410 [Radeon X700 Pro (PCIE)] ------------------------------
My main system uses an RV410 PCIE card. Lots of stuff works fine on fedora 13 (in fact everything I actually use works fine). If I stress the card using the mesa-demos gltestperf test, the whole system locks up and has to be power cycled.
https://bugzilla.redhat.com/show_bug.cgi?id=562607
This doesn't appear to have changed in f13, same thing happens in f12.
Radeon R100 QD [Radeon 7200] ----------------------------
This is an insanely ancient card (an original ATI All-In_Wonder). Fedora 12 was the peak of performance for this card. It actually worked a little. With Fedora 13, all I get is a "No Signal" message from my monitor. (Which is all I got for several releases prior to f12 as well :-).
https://bugzilla.redhat.com/show_bug.cgi?id=522970
RV280 [Radeon 9200] (rev 01) ----------------------------
I think I remember trying this with the live CD in f12 with very little success. It works better in fedora 13. It can handle KMS, it can handle turning on compiz, but it starts going funny in the head when I push things a bit:
On Sat, 2010-04-17 at 16:57 -0400, Tom Horsley wrote:
I didn't have time to fool with the test day for video, but I have finally gotten around to checking out my personal collection of ATI cards with latest fedora 13 updates, with these results:
Thanks for the testing, and for filing bugs. Someone from Bugzappers - possibly me :) - should follow up on the reports soon.
Tom Horsley wrote:
Radeon R100 QD [Radeon 7200]
This is an insanely ancient card (an original ATI All-In_Wonder). Fedora 12 was the peak of performance for this card. It actually worked a little. With Fedora 13, all I get is a "No Signal" message from my monitor. (Which is all I got for several releases prior to f12 as well :-).
Tom, does this card run in vesa mode?
On Mon, 03 May 2010 14:50:00 -0400 Bill Davidsen wrote:
Tom, does this card run in vesa mode?
Yea, I just tried that the other day, and if I cobble up an xorg.conf file and specify the vesa driver, I can get video (but of course, vesa can't do the native 1920x1080 resolution, so I get stretched out 1024x768).
On Mon, 2010-05-03 at 15:11 -0400, Tom Horsley wrote:
On Mon, 03 May 2010 14:50:00 -0400 Bill Davidsen wrote:
Tom, does this card run in vesa mode?
Yea, I just tried that the other day, and if I cobble up an xorg.conf file and specify the vesa driver, I can get video (but of course, vesa can't do the native 1920x1080 resolution, so I get stretched out 1024x768).
There's no 'of course' about it, there's no intrinsic reason vesa shouldn't be capable of the native resolution, and if you file a bug on this, ajax will look at it. When I did that for the Vaio P, he fixed the vesa driver to automatically run the P at its native 1600x768 resolution.
Of course, that won't make it faster. =)
On Thu, 06 May 2010 01:46:11 +0100 Adam Williamson wrote:
There's no 'of course' about it, there's no intrinsic reason vesa shouldn't be capable of the native resolution
Then why doesn't it read the EDID info and just go ahead and do native resolution? I thought the vesa driver was restricted to the built-in list of vesa standard resolutions or maybe mode setting via the BIOS (the BIOS on this video card can't do 1920x1088 either, it fact if it tries to read the EDID info it picks a mode the monitor can't display at all).
That's actually what I'd much rather have than a fast driver: a driver that worked reliably in 2D at native resolution all the time. Worry about speed and 3D much, much later :-).