intermittent failure to detect both screens on fx4800 with recent drivers

chris.cooper · 04-15-10, 08:19 PM

Hi,

We are having trouble on machines with fx4800 that have the 195.17 and 195.36.15 drivers installed. On system startup/restart often one of the displays will fail to be detected and will remain in standby mode. A manual scan via a button on the display fails to detect a signal. Many restarts later the screen will become active again (ie the chance of a successful detection seems to be a lot less than an unsuccessful one). Older drivers (eg 180.60) work fine.

thanks,
Chris

luke bewsell · 04-27-10, 01:36 AM

anyone?

**danix** · 04-27-10, 02:45 PM

Thanks for reporting this. This sounds similar to an issue we are currently investigating.

The Quadro FX 4800 has one DVI, and two DisplayPort connectors. The HP LP2475w has both DVI and DP.

How are these monitors connected? One is showing up as dual-link TDMS, and one single link. I'm assuming you're using a dual-link DVI cable for one, and a passive DVI single link DP-DVI dongle for the other.

Is the monitor that fails always the one on the DP dongle, if a dongle is used?

Does disabling hot plug detect support on the monitor(s) make the issue go away?

luke bewsell · 04-28-10, 06:39 PM

Quote:

Originally Posted by danix

Thanks for reporting this. This sounds similar to an issue we are currently investigating.

Can you provide me some links to the similar issues you are getting?

Quote:

The Quadro FX 4800 has one DVI, and two DisplayPort connectors. The HP LP2475w has both DVI and DP.

How are these monitors connected? One is showing up as dual-link TDMS, and one single link. I'm assuming you're using a dual-link DVI cable for one, and a passive DVI single link DP-DVI dongle for the other.

The secondary monitor is running over a normal DVI cable to port 1 of the monitor.
The primary monitor is running a display port -> dual link dvi converter. From the dual-link connection we are running a normal dvi cable to port 1 of the other monitor.

Quote:

Is the monitor that fails always the one on the DP dongle, if a dongle is used?

yes

Quote:

Does disabling hot plug detect support on the monitor(s) make the issue go away?

No, I've turned off all auto select feature and power off features on both monitors. It appears that once the monitor goes into standby mode and a logout(restart x session) occurs. That monitor will fail on the adapter.

**danix** · 05-03-10, 09:25 PM

Quote:

Originally Posted by luke bewsell

Can you provide me some links to the similar issues you are getting?

Unfortunately, this issue is logged in NVIDIA's internal bug tracking system, which is not visible outside of NVIDIA.

Quote:

No, I've turned off all auto select feature and power off features on both monitors. It appears that once the monitor goes into standby mode and a logout(restart x session) occurs. That monitor will fail on the adapter.

The specific feature which I'm asking about is called "Hot Plug Detect Support" and it's separate from the auto input select or the power management. On the 2480, which is similar to the 2475, I found this option in the menu under "Video Input Control".

It would also be useful to know what firmware revision you have on your display. You can get this from the menu under "Information"; the firmware revision is listed as "Version".

luke bewsell · 05-04-10, 01:54 AM

Quote:

Originally Posted by danix

Unfortunately, this issue is logged in NVIDIA's internal bug tracking system, which is not visible outside of NVIDIA..

Chris copper currently has developer access to your forums and limited visibility to your bug tracking system. Is there anyway to make that information in your bug tracking system available to either chris or myself?

Quote:

Originally Posted by danix

The specific feature which I'm asking about is called "Hot Plug Detect Support" and it's separate from the auto input select or the power management. On the 2480, which is similar to the 2475, I found this option in the menu under "Video Input Control".

We don't have hot plug detect support on the HP LP2475w. Under Video Input Control the following are set:
default Video Input -> DVI 1
Autodetect input -> off
auto switch input -> off

I take it you would also like the most current firmware running on these monitor as well. I'll get it installed once I get time.

Quote:

Originally Posted by danix

It would also be useful to know what firmware revision you have on your display. You can get this from the menu under "Information"; the firmware revision is listed as "Version".

I've ran around on got the firmware for computers having this issue. Firmware versions we are running :
GIG 034
GIG 045
GIG 078

luke bewsell · 05-10-10, 07:08 PM

does anyone have any other information relating to this issue?

**danix** · 05-17-10, 04:02 PM

Hi Luke,

This is a complex issue and we're still root-causing it. Certain combinations of hardware appear to lead to situations where display hardware won't be detected when X starts, and under the right circumstances, detection will continue to fail on subsequent attempts. These situations appear to be the result of interactions between the display, the passive DP-DVI dongle, and the DP controller on our GPU boards.

From what we understand so far, the hot plug detection probe that is performed by some displays will shut off DDC on some passive DP-DVI dongles momentarily. If the DP controller on the GPU board tries to initialize that display at just the right moment, DDC reads will fail, and the display won't come up. Simply restarting the X server will usually clear this up at first, but eventually, we get locked into a state where DDC reads will always fail, until we disconnect and reconnect the display. In some cases on certain displays, the display gets frozen up, and needs to have its power disconnected before it will work again.

We have found that other dongles with a different design, notably the Hosiden passive DP-DVI dongles, don't appear to produce this same behavior. We have also made a few changes in the driver that should reduce the rate at which bringing up the display fails when using one of the affected display/dongle/GPU board configurations, and prevent the complete lockups which would eventually happen, so that if the display fails to come up when starting X, simply restarting X should be enough to bring it back.

We also addressed an unrelated issue where EDID reads going through the DP controller were sometimes offset, which resulted in an invalid EDID.

Our next driver releases will include the above improvements. We hope to deliver additional improvements in future releases, as the problem is better understood.

luke bewsell · 05-18-10, 11:46 PM

ok great, thanks heaps danix. I'll now hunt down those hoisden dongles

I've also noticed the blurring of the edid information when I was debugging this issue.

james-p · 06-23-10, 06:43 AM

We are also seeing similar issues with FX1800 and FX3800 cards

For example, on a machine with an FX3800, using the 190.42 driver with an Apple Cinema HD on the DVI port and a Sony SDM-P234 on the DisplayPort, using a DP-DVI 'dongle', which was supplied with the card and is manufactured by Hosiden. After a reboot, X sometimes fails to find the Sony monitor.

Quote:

Originally Posted by danix

Our next driver releases will include the above improvements. We hope to deliver additional improvements in future releases, as the problem is better understood.

Which driver release is this referring to? Is this the current 195.36.31, or some future driver version?

Thanks

James

**danix** · 06-28-10, 06:07 PM

Hi james-p,

At the time I wrote that comment, 195.36.31 had not yet been released. It does include improvement work for detection of DVI displays connected with passive DP-DVI dongles. Field reports from customers who had been affected by this and similar issues after upgrading to 195.36.31 have been positive so far.

If you are experiencing similar issues with 195.36.31 and Hosiden dongles, then display detection may be failing for other reasons. Please test with 195.36.31, and if you find you do have problems, start a new thread and include an nvidia-bug-report. Thanks.

james-p · 06-29-10, 04:25 AM

Thanks - since my original posting, the 256.35 drivers have been released - I assume these also have the passive DP-DVI dongles improvements?

However, I couldn't find any mention of the DP-DVI dongle improvements in the release notes for either 256.35 or 195.36.31 ...

04-27-10, 01:36 AM	#2
luke bewsell Registered User Join Date: Apr 2010 Posts: 12	Re: intermittent failure to detect both screens on fx4800 with recent drivers anyone?

04-27-10, 02:45 PM	#3
danix NVIDIA Corporation Join Date: Feb 2010 Location: Santa Clara, CA Posts: 237	Re: intermittent failure to detect both screens on fx4800 with recent drivers Thanks for reporting this. This sounds similar to an issue we are currently investigating. The Quadro FX 4800 has one DVI, and two DisplayPort connectors. The HP LP2475w has both DVI and DP. How are these monitors connected? One is showing up as dual-link TDMS, and one single link. I'm assuming you're using a dual-link DVI cable for one, and a passive DVI single link DP-DVI dongle for the other. Is the monitor that fails always the one on the DP dongle, if a dongle is used? Does disabling hot plug detect support on the monitor(s) make the issue go away?

05-10-10, 07:08 PM	#7
luke bewsell Registered User Join Date: Apr 2010 Posts: 12	Re: intermittent failure to detect both screens on fx4800 with recent drivers does anyone have any other information relating to this issue?

05-17-10, 04:02 PM	#8
danix NVIDIA Corporation Join Date: Feb 2010 Location: Santa Clara, CA Posts: 237	Re: intermittent failure to detect both screens on fx4800 with recent drivers Hi Luke, This is a complex issue and we're still root-causing it. Certain combinations of hardware appear to lead to situations where display hardware won't be detected when X starts, and under the right circumstances, detection will continue to fail on subsequent attempts. These situations appear to be the result of interactions between the display, the passive DP-DVI dongle, and the DP controller on our GPU boards. From what we understand so far, the hot plug detection probe that is performed by some displays will shut off DDC on some passive DP-DVI dongles momentarily. If the DP controller on the GPU board tries to initialize that display at just the right moment, DDC reads will fail, and the display won't come up. Simply restarting the X server will usually clear this up at first, but eventually, we get locked into a state where DDC reads will always fail, until we disconnect and reconnect the display. In some cases on certain displays, the display gets frozen up, and needs to have its power disconnected before it will work again. We have found that other dongles with a different design, notably the Hosiden passive DP-DVI dongles, don't appear to produce this same behavior. We have also made a few changes in the driver that should reduce the rate at which bringing up the display fails when using one of the affected display/dongle/GPU board configurations, and prevent the complete lockups which would eventually happen, so that if the display fails to come up when starting X, simply restarting X should be enough to bring it back. We also addressed an unrelated issue where EDID reads going through the DP controller were sometimes offset, which resulted in an invalid EDID. Our next driver releases will include the above improvements. We hope to deliver additional improvements in future releases, as the problem is better understood.

05-18-10, 11:46 PM	#9
luke bewsell Registered User Join Date: Apr 2010 Posts: 12	Re: intermittent failure to detect both screens on fx4800 with recent drivers ok great, thanks heaps danix. I'll now hunt down those hoisden dongles I've also noticed the blurring of the edid information when I was debugging this issue.

06-28-10, 06:07 PM	#11
danix NVIDIA Corporation Join Date: Feb 2010 Location: Santa Clara, CA Posts: 237	Re: intermittent failure to detect both screens on fx4800 with recent drivers Hi james-p, At the time I wrote that comment, 195.36.31 had not yet been released. It does include improvement work for detection of DVI displays connected with passive DP-DVI dongles. Field reports from customers who had been affected by this and similar issues after upgrading to 195.36.31 have been positive so far. If you are experiencing similar issues with 195.36.31 and Hosiden dongles, then display detection may be failing for other reasons. Please test with 195.36.31, and if you find you do have problems, start a new thread and include an nvidia-bug-report. Thanks.

06-29-10, 04:25 AM	#12
james-p Registered User Join Date: Jun 2010 Posts: 13	Re: intermittent failure to detect both screens on fx4800 with recent drivers Thanks - since my original posting, the 256.35 drivers have been released - I assume these also have the passive DP-DVI dongles improvements? However, I couldn't find any mention of the DP-DVI dongle improvements in the release notes for either 256.35 or 195.36.31 ...