Code: Section 'Device' Identifier 'Radeon' Driver 'radeon' #Option 'DRI' '2' #Option 'AccelMethod' 'exa' Option 'NoAccel' 'True' EndSectionWhat these config options do is drive the card at maximum AGP-speed but turn off the radeon hardware specific acceleration methods. The boost of bandwith from 133 MB/s (PCI mode, as in radeon.agpmode=-1) to 2133 MB/s (AGP-x8 mode, as in radeon.agpmode=8) outweighs the loss of specific acceleration a lot, at least on my G5 machine with its 'ATI Radeon 9600 AP (AGP)' (ChipID = 0x4150).

You have to check first what maximum AGP-speed your card and your AGP-Northbride support. Then adjust the radeon.agpmode=x boot-parameter accordingly. On my G5 it's radeon.agpmode=8, on my PowerBook G4s it would be radeon.agpmode=4.

If this wors correctly you will get following messages from your kernel. Code: $ dmesg grep -i agp 0.000000 Found U3-AGP PCI host bridge. @rican-linux: Glad to hear that this works with your setup too! What graphic cards are in your iBook & Powerbook? I only got Radeon r300-class cards, so I can't make any statements about different hardware. I can only suppose this can improve performance on other rx00-card systems too if the machine has at least AGPx2 and an otherwise hardware acceleration is not. The only machine where I get.

full. HW-acceleratio n is my PowerBook 5,8. There I don't need the 'NoAccel'-option in xorg.confg and can drive the card with AGPx4. This does not work on a PB 5,6 or 5,4 which I had. There must be some subtile differences between the 5,8 model and the other ones.

I will continue playing around with some xorg.conf options, e.g. 'AccelMethod', 'ColorTiling', 'ColorTiling2D' look interesting. Let's see what leads to a GPU-lockup and if not to more performance. Code: rican-linux@Lubuntu-G5:$ dmesg grep -e radeon -e drm -e agp 5.725462 Linux agpgart interface v0.103 7.366602 agpgart-uninorth 0000:f0:0b.0: Apple U3 chipset 7.390331 drm Initialized drm 1.1.0 20060810 7.438792 agpgart-uninorth 0000:f0:0b.0: configuring for size idx: 64 7.438943 agpgart-uninorth 0000:f0:0b.0: AGP aperture is 256M @ 0x0 7.525634 drm radeon kernel modesetting enabled.

7.525700 fb: switching to radeondrmfb from OFfb ATY,Simone 7.528397 fb: switching to radeondrmfb from OFfb ATY,Simone 7.529583 radeon 0000:f0:10.0: enabling device (0006 - 0007) 7.530140 drm initializing kernel modesetting (RV350 0x1002:0x4150 0x1002:0x4150 0x00). 7.530164 drm register mmio base: 0xA0000000 7.530166 drm register mmio size: 65536 7.530210 radeon 0000:f0:10.0: Invalid PCI ROM header signature: expecting 0xaa55, got 0x0000 7.622759 drm Not an x86 BIOS ROM, not using. Code: $ cat /var/log/Xorg.0.log grep -i render 40.154 (WW) RADEON(0): Direct rendering disabled 40.236 (EE) AIGLX: reverting to software renderingMeans the DRI2-infrastucture is not used to display graphics but the framebuffer. But with DRI2 turned off the Radeon cards no longer freeze when using the intended AGP-x4 or AGP-x8 mode. That's what is causing the 'speed boost', because the maximum bandwith for accessing the graphics card is so much faster than plain PCI-mode.

Ideally DRI2 or DRI3 can be used AND the appropriate AGP-mode is autodeteced. Which is normally the case on Radeon cards on x86 PC-boards. And which magically works on your PowerMac G5 7,2 and my PowerBook G4 5,8 but not on other hardware-combiations. Don't know if variations of the Apple U3/U4 northbridge in the different PPC-Macs or specific Radeon cards are to blame but enabling DRI seems a very delicate matter.

Apple Footer. This site contains user submitted content, comments and opinions and is for informational purposes only. Apple may provide or recommend responses as a possible solution based on the information provided; every potential issue may involve several factors not detailed in the conversations captured in an electronic forum and Apple can therefore provide no guarantee as to the efficacy of any proposed solutions on the community forums.

Ati Radeon 2600 Pro Agp Drivers For Mac Pro

Apple disclaims any and all liability for the acts, omissions and conduct of any third parties in connection with or related to your use of the site. All postings and use of the content on this site are subject to the.