Michel Dänzer wrote: > On Sun, 2007-11-25 at 14:42 -0800, [EMAIL PROTECTED] > wrote: >> http://bugs.freedesktop.org/show_bug.cgi?id=12934 >> >> ------- Comment #8 from [EMAIL PROTECTED] 2007-11-25 14:49 PST ------- >> (In reply to comment #7) >>> Does Option "AGPMode" "1"... >> No, unlike "AGPMode" "2", "1" does not freeze the system. > > I'm interested in people's thoughts on whether leaving the AGP transfer > mode unchanged from what was set up by firmware by default has been a > net win or loss. > > AFAIR, previously there was mainly one report about some AGPv3 cards > only working with what was set up by the firmware, but in the meantime > there have been several reports where 1x works but not higher rates, > even if set up by the firmware. Maybe we could try sticking to the > current scheme for AGPv3 but going back to 1x by default otherwise? > > Thoughts?
Might be a good idea. I'm a bit puzzled though why those higher transfer rates turned out to be problematic (even with somewhat new chipsets not known to have trouble with AGP). Not to mention the circumstances it happens can be very very weird (like in this bug). Maybe it's possible though that other drivers reprogram the agp cntl regs of the gpu a bit depending on the chipset. Roland _______________________________________________ xorg-driver-ati mailing list [email protected] http://lists.x.org/mailman/listinfo/xorg-driver-ati
