|
|
#1 | |
|
Registered User
Join Date: Aug 2005
Posts: 5
|
After installing the latest 100.14.09 driver, the X resolution went down to 1024x768.
I'm using an FX 5700 card, connected via DVI to a DFP with 1600x1200 native resolution. This worked well with the previous driver, 9755, until right before rebooting with the new one. Here is an extract of a relevant part of Xorg.0.log, and see below for a link to the generated bug report: II) NVIDIA(0): NVIDIA GPU GeForce FX 5700 (NV36) at PCI:1:0:0 (GPU-0) (--) NVIDIA(0): Memory: 262144 kBytes (--) NVIDIA(0): VideoBIOS: 04.36.20.23.00 (--) NVIDIA(0): Interlaced video modes are supported on this GPU (--) NVIDIA(0): Connected display device(s) on GeForce FX 5700 at PCI:1:0:0: (--) NVIDIA(0): Iiyama AU5131 (DFP-1) (--) NVIDIA(0): Iiyama AU5131 (DFP-1): 150.0 MHz maximum pixel clock (--) NVIDIA(0): Iiyama AU5131 (DFP-1): Internal Single Link TMDS (II) NVIDIA(0): Assigned Display Device: DFP-1 (WW) NVIDIA(0): No valid modes for "1600x1200"; removing. (WW) NVIDIA(0): No valid modes for "1280x960"; removing. (II) NVIDIA(0): Validated modes: (II) NVIDIA(0): "1024x768" (II) NVIDIA(0): "800x600" (II) NVIDIA(0): "640x480" (II) NVIDIA(0): Virtual screen size determined to be 1024 x 768 (--) NVIDIA(0): DPI set to (63, 62); computed from "UseEdidDpi" X config http://www.elanor.org/~laurent/nvidi...bug-report.log |
|
|
|
|
|
|
#2 | |
|
NVIDIA Corporation
Join Date: Aug 2002
Posts: 3,740
|
Please start X with the -logverbose 6 option and generate another nvidia-bug-report.log file.
|
|
|
|
|
|
|
#3 |
|
Registered User
Join Date: Aug 2005
Posts: 5
|
Done, here it is, updated at the same address:
http://www.elanor.org/~laurent/nvidi...bug-report.log Thanks! |
|
|
|
|
|
#4 | |
|
NVIDIA Corporation
Join Date: Aug 2002
Posts: 3,740
|
It looks like the driver rejects the various 1600x1200 modes due to display device and TMDS encoder limitations; if one of the 1600x1200 modes worked with an older driver, you could run X with -logverbose 6 and that driver, identify which mode was set, determine the differences in the driver's mode validation and (at your own risk), override the problematic parts of the 100.14.xx driver's mode validation (see the README for details).
|
|
|
|
|
|
|
#5 | |
|
Registered User
Join Date: Aug 2005
Posts: 5
|
Quote:
Code:
Option "ModeValidation" "NoMaxPClkCheck" So there's still a regression as compared to the previous driver. AFAICT, the values received from the DFP are the same in both versions, but they are interpreted and used in different ways by each drivers. Any other idea why? |
|
|
|
|
|
|
#6 | |
|
Registered User
Join Date: Nov 2007
Location: europe
Posts: 4
|
hello everybody!
i realize this thread is already a few weeks old but nevertheless this might be interesting for users with the same problem as well as for nvidia developers: i while back i came across the same problem as user lblume describes above with driver-version 100.14.11. after searching for quite some time i found that there are others who have the same issues with 1600x1200 resolution and after some further research and trying around many hours i came to a solution on how to fix the matter. everything is described in this long thread here which i kind of hijacked but which also brought me a lot closer to the source that might cause this problem. its quite a lot to read but i guess the central aspects are to be found in my posting #58 and follow-ups. i also translated my description into german and posted it on hwluxx-forums in a lengthy thread about my motherboard with nvidia 7050 onboard graphics chipset in postings #337 and #339. there seem to be not too many users which run a combination of a 100.14.xx - driver at dvi-port and a resolution of 1600x1200 since i could not find much about this issue around the web and i thought that this might have been fixed already but when installing ubuntu 7.10 which comes with the 100.14.19 driver i realized that this behaviour is in fact still there. luckily my fix still worked ... i didn't get much feedback from other users that have the same issue since its quite some effort to get this fixed but i think it should work for almost everybody. so hopefully this bug won't be around with further releases since from what i've seen this seems to be a bug in the drivers interpretation of the edid submitted over dvi by the display. regards, anti EDIT: omg, i've just seen that this thread is posted in the solaris section! (found it via the boards search) but i haven't found any threads coming closer the the issue than this in the board even though i am talking about linux not (only) with focus on the BSD-branch here... EDIT2: ok, i've now just seen that there are also threads in the linux-section which reference this topic! i'm sorry to have not seen these earlier but since they (as far as i could see) don't come up with a general solution i will link this thread in the ones i've found adressing the same problem... EDID resolution problem Cannot get native resolution oh, and sorry for the confusion! |
|
|
|
|
![]() |
| Thread Tools | |
|
|