Jump to content
Sign in to follow this  
Branimir

DXGI_ERROR_DEVICE_HUNG error: Is there a solution?

Recommended Posts

I've developed a test setup for this bug thanks to one of the guys on this forum. Slew any plane at dusk at 160kts 5000ft over populated areas and let the sim run on its own. Wait for the DEVICE HUNG to happen. Make sure that multi-monitor or VR headset software is running and note the load on the GPU and CPU making sure they are roughly 50%. Let the setup run until the error happens. Compare things that were changed for each test.

Not having to actually fly the plane is a huge step forward for finding this error.

Share this post


Link to post
Share on other sites
5 hours ago, Pete Dowson said:

Or that something in the Video Driver actions for the 1080 are liable to reveal the error, whilst those for the 2080 are not. Also, of course, you have to consider that you are probably driving the 2080 to a lot less than its performance capacity compared to the 1080.

I thought I had it narrowed down to a specific 1080 card (one out of three) but after removing that I got it on one of the others.

Pete

 

Perhaps but before I had the 1080ti I had a 1070 which never had a error like this. And considering that I consistently could not run 3D Mark's default DX11 test without the card locking up (the 2080 works fine in it), I think it may be a hardware problem. Also, I have a test flight from PAJN (Juneau) to PANC (Anchorage), full overcast. Surprisingly, the GPU was pegged at 100% (but was still 30 FPS and smooth). I would have thought it would be less than that. I never was able to get the 1080ti to work on this flight plan without the DXGI fault. That was in 4k. When I had my 1070 (resolution 1080P), with the same settings as the 2080, my GPU usage was significantly less, probably at 50% or so. One could say that this is due to DX11 vs DX12 but my 2080 synthetic benchmarks are about equal to those published for the 2080. And considering the price of the card is $849 at the big box store (or $749 Black Friday sale), I am not too impressed with the value proposition of this card.

I haven't had a chance to thank you for your work in FSUIPC. Very much appreciated!

Happy Holidays!

The Other Peter

Edited by pgde

Gigabyte x670 Aorus Elite AX MB; AMD 7800X3D CPU; Deepcool LT520 AIO Cooler; 64 Gb G.Skill Trident Z5 NEO DDR5 6000; Win11 Pro; P3D V5.4; 1 Samsung 990 2Tb NVMe SSD: 1 Crucial 4Tb MX500 SATA SSD; 1 Samsung 860 1Tb SSD; Gigabyte Aorus Extreme 1080ti 11Gb VRAM; Toshiba 43" LED TV @ 4k; Honeycomb Bravo.

 

Share this post


Link to post
Share on other sites

Sorry for bring up an old experience but i had this issue all the time with P3D4.3 in SLI GTX970´s when i had ticked "high resolution textures" inside the P3D UI. 4GB VRAM was the limit on those cards.

Crossing my fingers now with my EVGA 1080ti FTW3 . Havent seen the error yet in 4.4 (also with exp=10)

Thanks

Michael Moe

  • Like 1

Michael Moe

 

fs2crew_747_banner1.png

Banner_FS2Crew_Emergency.png

Share this post


Link to post
Share on other sites
7 hours ago, Pete Dowson said:

So do you mean that has to be stopped somehow? Some dep -rooted Windows services?

Maybe so, but I been testing all day with no recurrence.

That would be neat, but it has only ever occurred he in a particular scenery -- but one which I've been using in P3D4, rigorously (after all, it is EGLL) since first installing P3D4 (and before, in previouus scenery versions).

Before P3D4.4 is occurred just once, a month or two ago. With P3D 4.4. it just started occurring reproducibly -- but now, not since i installed newer nVidia drivers.

Or that something in the Video Driver actions for the 1080 are liable to reveal the error, whilst those for the 2080 are not. Also, of course, you have to consider that you are probably driving the 2080 to a lot less than its performance capacity compared to the 1080.

I thought I had it narrowed down to a specific 1080 card (one out of three) but after removing that I got it on one of the others.

Pete

 

occasionally get this error on a stock 1080ti non-shader mod single gpu system.

i was thinking it would be great if Pete could fix this like the g3d.dll trap fix back in the fsx days....

  • Like 2

i9-10900k @ 5.1GHz 32G XMP-3200 | RTX3090 | 3T m.2 | Win11 | vkb-gf ultimate & pedals | virpil cm3 throttle | 55" 4k UHDTV | HP R-G2 VR | DCS

 

 

 

Share this post


Link to post
Share on other sites
12 hours ago, Pete Dowson said:

So do you mean that has to be stopped somehow? Some dep -rooted Windows services?

Maybe so, but I been testing all day with no recurrence.

That would be neat, but it has only ever occurred he in a particular scenery -- but one which I've been using in P3D4, rigorously (after all, it is EGLL) since first installing P3D4 (and before, in previouus scenery versions).

Before P3D4.4 is occurred just once, a month or two ago. With P3D 4.4. it just started occurring reproducibly -- but now, not since i installed newer nVidia drivers.

Or that something in the Video Driver actions for the 1080 are liable to reveal the error, whilst those for the 2080 are not. Also, of course, you have to consider that you are probably driving the 2080 to a lot less than its performance capacity compared to the 1080.

I thought I had it narrowed down to a specific 1080 card (one out of three) but after removing that I got it on one of the others.

Pete

 

Exactly. Now the card is replaced with a different model you have a fresh factory default setup so could still be just that as I suggested. Factory defaulting any profile is not making a profile in the sense you have a personal file of settings its a null. The format of data arriving at the card is what the error suggest so can be more or less anything - including a faulty card. Rather than purchase new hardware I prefer to be sure I don't have a mess up in the system somewhere. Especially where P3D is concerned - put it this way -  the error may have gone because some order of function has changed since some features will be accelerated with the new card and others less so.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites
10 minutes ago, SteveW said:

Exactly. Now the card is replaced with a different model you have a fresh factory default setup so could still be just that as I suggested. Factory defaulting any profile is not making a profile in the sense you have a personal file of settings its a null.

No, I removed the third GPU of my 3-way SLI attempt. The two left were the ones with the majority of use before in any case. When I was informed (rather belatedly, as I had this since Sept 2017) that 3-way SLI wasn't supported, I re-jigged the SLI links first one way then the other (keeping in common the card with all the displays connected and disabling the disconnected one). This seemed to indicate card 2 but then it happened with 3 also.

The fresh defaut setup in NCP was by using DDU and installing a new version of the nVidia driver.

I'm not sure the error has really gone yet. I haven't actually tried flying.

Pete

 


Win10: 22H2 19045.2728
CPU: 9900KS at 5.5GHz
Memory: 32Gb at 3800 MHz.
GPU:  RTX 24Gb Titan
2 x 2160p projectors at 25Hz onto 200 FOV curved screen

Share this post


Link to post
Share on other sites

Did this error appear only after pulling out the third card? I believe the system should be OK with three cards, however many a particular app can use. So P3D shouldn't be affected, but sli is a bit sketchy on most systems. The new 20xx cards have a better system for linking which takes stress off the bus.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites

The factory default thing seems to have stretched a long way....No matter how unimportant it might seem to anyone, the Restore/Apply buttons are there for a purpose. May not be relevant to Pete's system, I don't know, but it cured a bunch of systems in the past. Usually I thought because the user had tried setting up in NPI (profile inspector). But then I found some claimed they had not been in there, and only updated a driver at some stage. So it is possible that the factory defaults do not get applied in some circumstances even with a new driver update. I would say ignore at one's peril.

Edited by SteveW

Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites
3 hours ago, SteveW said:

Exactly. Now the card is replaced with a different model you have a fresh factory default setup so could still be just that as I suggested. Factory defaulting any profile is not making a profile in the sense you have a personal file of settings its a null. The format of data arriving at the card is what the error suggest so can be more or less anything - including a faulty card. Rather than purchase new hardware I prefer to be sure I don't have a mess up in the system somewhere. Especially where P3D is concerned - put it this way -  the error may have gone because some order of function has changed since some features will be accelerated with the new card and others less so.

Hi,

Does anyone else find all this to be a little worrying? I know I do..lol!

It’s almost as if the sword of Damocles is hovering over systems, like mine, currently free from this showstopping issue.

When I update a graphics card I always return NCP settings to their defaults before uninstalling the drivers. That’s the only time I use DDU. I keep a record of every game/sim-specific ‘tweaked’ Profile, including that for Prepar3D, so these can be reinstated following the hardware upgrade/driver installation. It can be a bit of a pain if you have multiple games and sims installed, but at least ensures you are truly restarting afresh.

However, after reinstating everything, GPU Drivers are usually updated via GeForce Experience. This is a quick and painless exercise and I rarely go down the ‘clean install’ route. That way all my Profiles are preserved without any further interventions on my part. Can’t help feeling I must be doing something right as I haven’t been experiencing problems relating to the subject of this thread. If I had, by now my scalp certainly would be follicly challenged! To date, with this rig I have updated three times: 2xGTX980 in SLI, 2xGTX980 Ti in SLI and a single GTX1080 Ti.

Problem I see with all this is how do others across the world cope with a GPU or Driver Upgrade? Most are not endowed with our in-depth knowledge and understanding so I suspect most will simply rely on the operating system to handle everything and keep them out of trouble. I’m guessing that this mind set is usually successful and the hardware/software/operating system elements of the equation are far more fault tolerant than we give them credit. But then, what do I know?

None of this removes that undercurrent of unease as I await what may be an inevitable development. Wishing now that I had not come across this thread! 😥

Regards,

Mike

Share this post


Link to post
Share on other sites

Hi Mike, Not really a worry as 99% of everyone experience no problems at all whatever they do with NCP or NPI or Experience, all work as planned. If an error creeps into that system, it's difficult to understand what's going on. When there's a system like this with ongoing updates, problems go into that system and even clean up along the way with subsequent updates and go completely unnoticed. It's never safe to assume this stuff is always going to be 100% correct. For a few unlucky ones, problems show up if the set of conditions allows it. That is the nature of how problems appear in these systems.

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites
3 hours ago, Cruachan said:

 

None of this removes that undercurrent of unease as I await what may be an inevitable development. Wishing now that I had not come across this thread! 😥

 

Did you not see the recent P3D thread whereby the system crashes then runs fine so long as it crashes the DirectInput first? Anything can crop up. Assumption is the first point of failure. When investigating a problem, take care of possible randoms, reduce the system to the leanest error prone package and cross check (run similar software to invoke the system under investigation), reduce the system further to the stock known working setup and cross check. The problem cause should surface quick enough.

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites

DX utilises hardware that is shared precisely, not like mere files and folders. DX subsystems in operation are not intuitive and do not lend themselves to a natural way of thinking. These are specially shared resources and we might very well find a ghost in the system if things don't go on as expected.

.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Share on other sites
19 minutes ago, SteveW said:

Did you not see the recent P3D thread whereby the system crashes then runs fine so long as it crashes the DirectInput first?

Yes, I did, and it reminded me of an annoying period last year when I was having trouble accessing the NVidia Control Panel - kept crashing as it loaded. I noted I could get it to work if I ran it during a brief window after the desktop appeared and this success persisted throughout a session. I suspected the ASUS AISuite 3 and, indeed, delaying the loading of a couple of elements of the Suite via the Scheduler provided a workaround. In the end, the problem somehow resolved itself (? via an update) with no acceptance of responsibility from ASUS. Should I be surprised or was it a system-specific issue since I found no similar reports on the Net?

Regards,

Mike

Edited by Cruachan
  • Like 1

Share this post


Link to post
Share on other sites

I have yet to see any DXGI error since my re-install of windows and P3D. I used to get them constantly..

One thing I did do differently was,  to only install P3Dv4 compatible scenery. I left other FSX only scenery in my storage drive. Another thing I've noticed was WIn 10 1809 included newer drivers in the release package. It didn't take as long to update my default drivers to the latest releases. I'm wondering if the previous Windows update didn't overwrite the video drivers properly? A proper re-install may have corrected that.

I'm saying this was the solution but, I haven't seen any errors since doing a full re-install. 

Anyway, thought I'd contribute to the conversation, since it's such a difficult error to pinpoint.

Cheers.

Edited by Tino
  • Like 2

Sante Sottile
 

Share this post


Link to post
Share on other sites

Is this bug a 4.4 thing? Because with 4.3 and and earlier I never had a Device Hung error but having just installed 4.4 (client only because I have a system rebuild planned) I have had my first - at KSEA at the gate about to push back seemed to be precipitated by changing to an external view (EZDOCK).  Reset the scenario with the same settings and addons P3D4.4, LittleNav, AS, GSX2, EZDOK and the usual hi end Pacific NW scenery/airports but couldn't reproduce the error and my flight to KSFO proceeded without incident.  However if this keeps happening it is back to 4.3.  My current system is 4770 @ 4.4/980ti based.

Bruceb

  • Like 1

Bruce Bartlett

 

Frodo: "I wish none of this had happened." Gandalf: "So do all who live to see such times, but that is not for them to decide. All we have to decide is what to do with the time that is given to us."

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...