Re: radeon graphic speed regression 2.4 -> 2.6 test7 benh
On Wed, 2003-10-29 at 23:30, Soeren Sonnenburg wrote:
> On Wed, 2003-10-29 at 18:00, Michel Dänzer wrote:
> > On Wed, 2003-10-29 at 09:51, Soeren Sonnenburg wrote:
> > > On Wed, 2003-10-29 at 09:43, Benjamin Herrenschmidt wrote:
> > > >
> > > > You may be encountering a different issue ... (2.6 is usually overall
> > > > faster than 2.4 here, though I haven't timed gnome-terminal specifically)
> > >
> > > it is not gnome-terminal but multi-gnome-terminal which is more like
> > > xterm. So no antialiasing -> much faster.
> >
> > BTW, are you using AGP GART? You need to load the uninorth-agp module
> > for that with 2.6, agpgart is no longer enough. The pathetic performance
> > of PCI GART on newer machines might explain what you're seeing.
>
> Indeed not. But I just tried, no change in results...
So the server log showed it to use PCI GART before but AGP GART
afterwards?
> isn't agp needed for 3d stuff only anyway ?
When the DRI is enabled, the GART is also used for 2D acceleration, so
PCI GART will slow it down as well.
--
Earthling Michel Dänzer | Debian (powerpc), X and DRI developer
Software libre enthusiast | http://svcs.affero.net/rm.php?r=daenzer
Reply to: