weird display issues using radeon HDMI output
Alex Deucher
alexdeucher at gmail.com
Tue Apr 26 22:44:30 PDT 2011
On Mon, Apr 25, 2011 at 9:13 PM, Shane O'Connell <shane at oconnell.cc> wrote:
> Hi all,
> I'm having problems getting the HDMI output of my motherboard working
> with a Samsung TV. The motherboard is an MSI 785GM-P45 with an AMD
> 785G chipset with an onboard Radeon HD 4200. I can see the desktop on
> the TV, but the edges are cut off and the colours are weird.
Your TV is overscanning hdmi. You can either disable overscan in your
TV's configuration, or enable underscan in the radeon driver (xrandr
--output HDMI-0 --set underscan on). As for the color problems, try
disabling hdmi audio (using kernel 2.6.38 or newer, boot with
radeon.audio=0 on the kernel command line in grub.
> Strangely, another computer connected to the same TV using VGA is
> working fine.
> I've looked over the Xorg.0.log file for the computer using HDMI and
> the one using VGA, and it looks like they get slightly different EDID
> information from the TV. I don't know enough about it though to know
> if it's related to my problem. Here's a couple of key lines from the
> logs where they seem to differ (I've attached the complete log files
> as well, there's another few places where they don't match but these
> seem the most notable to me):
The TV treats the VGA and HDMI inputs differently.
Alex
>
> VGA:
> [ 10.808] (II) intel(0): EDID for output VGA1
> [ 10.808] (II) intel(0): Manufacturer: SAM Model: 666 Serial#: 0
> [ 10.808] (II) intel(0): Year: 2009 Week: 47
>
> HDMI:
> [ 1026.636] (II) RADEON(0): EDID for output HDMI-0
> [ 1026.636] (II) RADEON(0): Manufacturer: SAM Model: 667 Serial#: 1
> [ 1026.636] (II) RADEON(0): Year: 2009 Week: 47
>
>
> VGA:
> [ 10.809] (II) intel(0): Ranges: V min: 60 V max: 75 Hz, H min: 30
> H max: 81 kHz, PixClock max 175 MHz
>
> HDMI:
> [ 1026.636] (II) RADEON(0): Ranges: V min: 24 V max: 75 Hz, H min: 26
> H max: 81 kHz, PixClock max 235 MHz
>
>
> VGA:
> [ 10.809] (II) intel(0): EDID (in hex):
> [ 10.809] (II) intel(0): 00ffffffffffff004c2d660600000000
> [ 10.809] (II) intel(0): 2f130103685832782aee91a3544c9926
> [ 10.809] (II) intel(0): 0f5054bdef80714f8100814081809500
> [ 10.809] (II) intel(0): 950fb300a940023a801871382d40582c
> [ 10.809] (II) intel(0): 450076f23100001e662150b051001b30
> [ 10.809] (II) intel(0): 4070360076f23100001e000000fd003c
> [ 10.809] (II) intel(0): 4b1e5111000a202020202020000000fc
> [ 10.809] (II) intel(0): 0053414d53554e470a20202020200053
> [ 10.809] (II) intel(0): EDID vendor "SAM", prod id 1638
> [ 10.809] (II) intel(0): Using EDID range info for horizontal sync
> [ 10.809] (II) intel(0): Using EDID range info for vertical refresh
>
> HDMI:
> [ 1026.636] (II) RADEON(0): EDID (in hex):
> [ 1026.636] (II) RADEON(0): 00ffffffffffff004c2d670601000000
> [ 1026.636] (II) RADEON(0): 2f130103801009780aee91a3544c9926
> [ 1026.636] (II) RADEON(0): 0f5054bdef80714f8100814081809500
> [ 1026.636] (II) RADEON(0): 950fb300a940023a801871382d40582c
> [ 1026.636] (II) RADEON(0): 4500a05a0000001e662150b051001b30
> [ 1026.636] (II) RADEON(0): 40703600a05a0000001e000000fd0018
> [ 1026.636] (II) RADEON(0): 4b1a5117000a202020202020000000fc
> [ 1026.636] (II) RADEON(0): 0053414d53554e470a20202020200129
> [ 1026.636] (II) RADEON(0): 020322f1469004050320222309070783
> [ 1026.636] (II) RADEON(0): 010000e2000fe305030167030c002000
> [ 1026.636] (II) RADEON(0): b82d011d007251d01e206e285500a05a
> [ 1026.636] (II) RADEON(0): 0000001e011d8018711c1620582c2500
> [ 1026.636] (II) RADEON(0): a05a0000009e8c0ad08a20e02d10103e
> [ 1026.636] (II) RADEON(0): 9600a05a000000180000000000000000
> [ 1026.636] (II) RADEON(0): 00000000000000000000000000000000
> [ 1026.636] (II) RADEON(0): 000000000000000000000000000000df
>
>
> Does anybody know where I should go from here? Is there anything else
> I can try? My thinking is somehow maybe the EDID information received
> over HDMI is incorrect, if so does that mean a quirk can be added for
> this TV so that it works properly?
>
> Thanks!
> -Shane O'Connell
>
> _______________________________________________
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
>
>
More information about the dri-devel
mailing list