Re: DVI on X1250 detected on radeonhd but not on radeon

2008-03-17 Thread Alex Deucher
On Sun, Mar 16, 2008 at 2:21 AM, Adam J. Richter [EMAIL PROTECTED] wrote:
 Hi everybody,

 radeonhd-1.1.0 correctly detects my DVI connection but
  radeon-6.8.0 always thinks is not connected.  With radeon-6.8.0, I can
  get correct DVI output on my 1920x1200 monitor by simultaneously
  connecting DVI and VGA cables from the same monitor, but, if I only
  connect the DVI cable, the video comes up in a much lower default
  resolution, perhaps 640x480.


This is fixed in ati git master.  6.8.0 is too old.

Alex
___
xorg-driver-ati mailing list
xorg-driver-ati@lists.x.org
http://lists.x.org/mailman/listinfo/xorg-driver-ati


Re: DVI on X1250 detected on radeonhd but not on radeon

2008-03-17 Thread Matthias Hopf
On Mar 15, 08 23:21:27 -0700, Adam J. Richter wrote:
   By the way, I should mention that while radeonhd normally
 works fine for me, it does have slightly corrupted video if I run it
 after I have run the radeon driver.  In particular, there is a
 horizonal line about 1/3rd of the way from the top of the screen where
 the default X weave seems to have a kink in it.

I think we have a bug open about that one. At least it sounds familiar.
Ah, right, #15022, though that one is worse.

   I would be happy to assist in further tracking this problem
 down in the limited time I have available.  I'd appreciate a point to
 a public source code control development tree if one exists, as I
 haven't found one public cvs, svn, git development tree in my perhaps
 too quick efforts at poking around the mail archives and the doing web
 searches.

radeonhd's git:

git://anongit.freedesktop.org/git/xorg/driver/xf86-video-radeonhd

   Also by the way, in case anyone wonders, I was trying
 radeon-6.8.0 instead of radeonhd-1.1.0 because I am considering
 getting a 2560x1600 monitor, and the term dual link at least appears
 in a few places in xf86-driver-ati-6.8.0, although I don't know if

Current git of radeonhd should support dual link DVI as well, though
this is untested ATM.

CU

Matthias

-- 
Matthias Hopf [EMAIL PROTECTED]  ____   __
Maxfeldstr. 5 / 90409 Nuernberg   (_   | |  (_   |__  [EMAIL PROTECTED]
Phone +49-911-74053-715   __)  |_|  __)  |__  R  D   www.mshopf.de
___
xorg-driver-ati mailing list
xorg-driver-ati@lists.x.org
http://lists.x.org/mailman/listinfo/xorg-driver-ati


Re: DVI on X1250 detected on radeonhd but not on radeon

2008-03-16 Thread Adam J. Richter
Hi everybody,

radeonhd-1.1.0 correctly detects my DVI connection but
radeon-6.8.0 always thinks is not connected.  With radeon-6.8.0, I can
get correct DVI output on my 1920x1200 monitor by simultaneously
connecting DVI and VGA cables from the same monitor, but, if I only
connect the DVI cable, the video comes up in a much lower default
resolution, perhaps 640x480.

I assume that what is happening when I have both cables
connected is that the DDC information is being read from VGA and the
DVI port is left on with the same signal as the VGA port is sending.
With both cables connected, xrandr will tell me that only VGA is
connected when I am using the radeon driver and it tells me that both
VGA and DVI are connected when I use radeonhd.

My hardware configuation is an AMD CPU + ATI X1250 (RS690
according to lspci) built into the motherboard.  The motherboard actually
has the following connectors or cable headers: 

1. VGA
2. S-Video
3. Component video
4. DVI
5. HDMI

Both radeon and radeonhd drivers claim to get a list of three
connections from BIOS:

1. VGA
2. S-Video
3. HDMI (radeon) / DVI (radeonhd)

I suspect that the there really are only three output
channels, with HDMI and DVI wired together and some similar
arrangement for S-Video and component video.

Trying the ReverseDDC option with radeon-6.8.0 and with only
the DVI cable connected still resulted in X coming up in low
resolution, as it did without ReverseDDC.  Likewise, attempts to
specify MonitorLayout and monitor-HDMI-0 also had did not change
this behavior.

By the way, I should mention that while radeonhd normally
works fine for me, it does have slightly corrupted video if I run it
after I have run the radeon driver.  In particular, there is a
horizonal line about 1/3rd of the way from the top of the screen where
the default X weave seems to have a kink in it.

I would be happy to assist in further tracking this problem
down in the limited time I have available.  I'd appreciate a point to
a public source code control development tree if one exists, as I
haven't found one public cvs, svn, git development tree in my perhaps
too quick efforts at poking around the mail archives and the doing web
searches.

Also by the way, in case anyone wonders, I was trying
radeon-6.8.0 instead of radeonhd-1.1.0 because I am considering
getting a 2560x1600 monitor, and the term dual link at least appears
in a few places in xf86-driver-ati-6.8.0, although I don't know if
radeon-6.8.0 actually supports dual link DVI.  The monitor I am
contemplating does not have a VGA port, so the workaround of
simultaneously connecting a VGA cable would not work.

Thanks to Timo Jyrinki for directing me to this mailing list.

Adam Richter
___
xorg-driver-ati mailing list
xorg-driver-ati@lists.x.org
http://lists.x.org/mailman/listinfo/xorg-driver-ati