Mark,
You might want to update the CVS version of XFree86. I've just committed
a patch from ATI which has some mods for Rage 128 Pro PF and Flat Panels.
Alan.
On Tue, Sep 25, 2001 at 05:26:05PM +0100, Mark Shinwell wrote:
> As mentioned in a previous mail, I am having trouble with a
> Rage 128 Pro PF chipset and a Samsung SyncMaster 170T flatpanel.
>
> The problems arise when trying to drive the panel through the DVI output.
>
> I am now of the opinion that the card itself is a Rage Fury Pro (it does
> not say on the card, and the box was not purchased by myself) -- I had
> originally been informed that it was an Xpert 2000.
>
> Back on 7th May 2001 there was a post explaining how the card appears to lack
> a flatpanel BIOS, which causes trouble in various places in r128_driver.c.
> By using ProgramFPRegs and patching this file as suggested in that post
> I have been able to obtain a display which is partially OK. I am working
> on XFree86 4.1.99.1.
>
> Having got that far I am determined to make the thing work properly :)
>
> Upon starting X (with no clients) the display is fine at 1280x1024, with the
> exception of a pair of vertical lines about 3/4 of the way across the screen,
> and corruption of a few pixels at the bottom between the two vertical lines.
> Attempting to start any clients at all causes strange effects -- the left-hand
> half of the screen is fine but the right-hand half has corruption and some
> mirroring from the left half.
>
> I note that the one parameter which is read from the FP bios which cannot
> be overridden is "PanelPwrDelay" (line 596 of r128_driver.c). This
> will contain garbage on my system due to that line never being executed --
> could someone enlighten me as to whether this is likely to cause display
> corruption?
>
> My main question is whether it is possible for me to lay my hands on
> any ATI documentation which might help in sorting this out -- how does
> the Rage Fury Pro DVI version work under Windows, for example? I presume
> there must be some method of getting the required information from the
> card, as the parameters themselves (panel width/height, etc) will depend
> on the individual panel connected.
>
> Any suggestions would be appreciated -- this is my first attempt at
> hacking video drivers!
>
> Mark
>
> --
> Mark Shinwell -- http://mrs30.quns.cam.ac.uk/ -- [EMAIL PROTECTED]
> Theory and Semantics Group, University of Cambridge Computer Laboratory
> _______________________________________________
> Xpert mailing list
> [EMAIL PROTECTED]
> http://XFree86.Org/mailman/listinfo/xpert
_______________________________________________
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
