|
|
#1 | |
|
Registered User
Join Date: Nov 2009
Posts: 1
|
Hardware/software configuration:
Quadro NVS140M in Dell D830 Latitude, 1920x1200 screen external LCD Soyo Topaz S (1920x1200), 24", driven by the Dsub connector Fedora Core 10, kernel 2.6.27.38-170.2.113.fc10.x86_64 Nvidia driver bundle: nvidia-180.60-1.fc10.x86_64 (from rpmfusion.org) Log Xorg.0.log says that both displays are seen, and that 1920x1200@60Hz is validated for both displays. It then adds: (WW) NVIDIA(0): Due to hardware capability constraints, disabling display (WW) NVIDIA(0): device Seiko (DFP-0) in MetaMode (WW) NVIDIA(0): "CRT:1920x1200+0+0,DFP:1920x1200+0+1200". Random commentary: I have not been able to use both displays for any 180.xx driver. The previous driver series, 177.82, worked as expected: both displays enabled, both with 1920x1200 resolution. One possible issue: the Soyo monitor returns a bad name descriptor ("BIP @ $A?0? ??"), so if the nvidia driver is doing checksums on the EDID data, that might be triggering the problem. On the other hand, if it's the Soyo monitor that's returning an EDID error, why is it turning off the laptop screen? My Dell 8500 Inspiron with a Go FX5200 card (linux driver 173.14.18) drives both its internal display and the Soyo LCD at 1920x1200. I assume that what the Go FX5200 hardware can do, the Quadro NVS140M can also do. Windows on the Dell D830 (same NVS140M card) drives both displays at 1920x1200, so the problem seems to be that the 180.xx driver series is checking (against what??) in a way that the Windows driver, the 173.14.xx driver, and the 177.82.xx drivers did/do not. I have found exactly two other instances of this error message online: http://osdir.com/ml/linux.redhat.fed.../msg00058.html http://bbs.archlinux.org/viewtopic.php?id=65734 Not clear that the first is relevant, but the 2nd link - especially the next-to-last post - is close to my problem. Possibly significant that webframp has an NVS140M as I do. It seems likely (in the absence of evidence) that the driver is performing a check that tells it that the NVS140M can't drive both displays. I'll speculate wildly that the Soyo is returning a bad parameter (which nonetheless allows the nvidia driver to construct a valid mode for the Soyo) and - the external display being primary - the driver decides that if it has to turn off a display, it will turn off the secondary display. Suggestions? Is there such a check? if so, can it be turned off? Will the 190/195 driver series fix this? Thanks for any help. |
|
|
|
|
|
|
#2 | |
|
Registered User
Join Date: Dec 2009
Posts: 1
|
I had the exact same issue and finally got it working by adding:
Code:
Option "ModeValidation" "NoExtendedGpuCapabilitiesCheck" I hope it can help. |
|
|
|
|
![]() |
| Thread Tools | |
|
|