LaCie 321 DFP: mode misdetected - EDID misinterpreted?

Roland Scheidegger sroland at tungstengraphics.com
Thu Jun 14 12:14:19 PDT 2007


Mourad De Clerck wrote:
> Hi,
> 
> I've got 2 LaCie 321 DFPs, but can't seem to drive them at the native
> resolution (1600x1200) with the open "nv" driver. The maximum resolution
> I'm able to use is 1280x1024.
> 
> I don't have any problems driving both at the native resolution with the
> closed "nvidia" driver. In addition, I seem to remember having the same
> problem with the DVI output of an ATI Radeon 9200 and the corresponding
> open "radeon" driver - though I can't verify and confirm this at the moment.
> 
> Attached are the logfiles and conffiles of the "nv" and "nvidia" drivers
> resp., to highlight the difference.
> 
> 
> In the xorg-nv.log you'll notice:
> 
> (II) NV(0): Supported VESA Video Modes:
> ...<snipped lower resolutions>
> (II) NV(0): 1280x1024 at 75Hz
> (II) NV(0): 1152x870 at 75Hz
> (II) NV(0): Manufacturer's mask: 0
> (II) NV(0): Supported Future Video Modes:
> ...<snipped>
> (II) NV(0): #7: hsize: 1600  vsize 1200  refresh: 60  vid: 16553
> ...<snipped>
> (--) NV(0): CRTC 0 is currently programmed for DFP
> (II) NV(0): Using DFP on CRTC 0
> (--) NV(0): Panel size is 1280 x 1024
> ...<snipped>
> (II) NV(0): Not using driver mode "1600x1200" (exceeds panel dimensions)
> (WW) NV(0): Shrinking virtual size estimate from 1600x1200 to 1280x1024
> 
> 
> I don't have the logfiles of the Radeon 9200 with the same screen(s),
> but the symptom was the same: instead of displaying at 1600x1200, the
> maximum resolution was 1280x1024.
> 
> Does anyone have any idea how to solve this issue?

This rather looks to me like your panel lies about its size, I don't
think there is any "misinterpretation".
The radeon driver allows to override this (Option "PanelSize"), not sure
about nv. Maybe the driver could be made more intelligent and just
ignore it if the panel provides a prefered detailed timing with a larger
resolution (which is probably what the nvidia driver does).
This is certainly not the first monitor with bogus edid data.

Roland



More information about the xorg mailing list