Nvidia performance article

[Ray Proudfoot 8,949]

PROBLEM IDENTIFIED AND FIXED.

After exhausting all other options I decided to close MSI Afterburner / RTSS as I've always had those running so I can monitor VRAM usage and other important aspects of P3Dv5.

Closing both programs has cured the problem. I can switch to other aircraft inside the sim without the DXGI error being generated. But delving deeper if I have both running but turn off the RTSS display before switching vehicles there is no crash.

So for those running Afterburner / RTSS make sure you turn off the screen display in RTSS before changing vehicles. Hopefully a fix can be found either by RTSS developers or LM.

Thanks @simbol for your help. As with many things the solution is very simple.

 

[kevinfirth 1,393]

36 minutes ago, Ray Proudfoot said:

PROBLEM IDENTIFIED AND FIXED.

After exhausting all other options I decided to close MSI Afterburner / RTSS as I've always had those running so I can monitor VRAM usage and other important aspects of P3Dv5.

Closing both programs has cured the problem. I can switch to other aircraft inside the sim without the DXGI error being generated. But delving deeper if I have both running but turn off the RTSS display before switching vehicles there is no crash.

So for those running Afterburner / RTSS make sure you turn off the screen display in RTSS before changing vehicles. Hopefully a fix can be found either by RTSS developers or LM.

Thanks @simbol for your help. As with many things the solution is very simple.

 

Good call Ray, I'll try that too!

[SteveW 1,792]

My advice with any problems is to eliminate in the first instance as much as half the possibilities. So on a typical FSX/P3D simulator I would turn off any external influences with a system first such as Nvidia profile changes, RTSS type programs, external programs that change any aspect of the simulator, leaving just the simulator. If the problem is cured I would investigate those items. If not cured next I would turn off half the addons and first include the ones I felt might be more suspicious of causing problems.

Say I suspect an addon causing an issue and I have 16 addons. Rather than trying 16 times eliminating one at a time, I can eliminate half (8), then half the remainder (4), then half the remainder (2), then that leaves one item. That is 3 tests to find one item in 16.

 

[Ray Proudfoot 8,949]

@SteveW, that involves a huge amount of work. I know my v5.1 install is stable. It had to be something simple because so few others were reporting the problem.

So I thought about what I was running that was different. The answer is Afterburner / RTSS. But delving deeper it's just the RTSS display causing it, not those programs running.

[SteveW 1,792]

6 minutes ago, Ray Proudfoot said:

@SteveW, that involves a huge amount of work. I know my v5.1 install is stable. It had to be something simple because so few others were reporting the problem.

So I thought about what I was running that was different. The answer is Afterburner / RTSS. But delving deeper it's just the RTSS display causing it, not those programs running.

Actually if you re-read me Ray, I was attempting to show how to do the least work Ray. Disabling RTSS and Afterburner in the first instance seems to be next to no work. Never mind, at least you found it and it's good to know that there is a problem with that.

[Ray Proudfoot 8,949]

2 minutes ago, SteveW said:

Actually if you re-read me Ray, I was attempting to show how to do the least work Ray. Disabling RTSS and Afterburner in the first instance seems to be next to no work. Never mind, at least you found it and it's good to know that there is a problem with that.

I have a vague recollection of turning off Afterburner when I first had this problem. But it didn't fix it probably because RTSS was perhaps still running or active in some way. So I never persisted with it until today.

Your advice is appreciated and I'm familiar with the binary chop process. Pete D uses it.

[SteveW 1,792]

1 hour ago, Ray Proudfoot said:

I have a vague recollection of turning off Afterburner when I first had this problem. But it didn't fix it probably because RTSS was perhaps still running or active in some way. So I never persisted with it until today.

Your advice is appreciated and I'm familiar with the binary chop process. Pete D uses it.

Sorry Ray, i didn't mean to seem perhaps like I felt you might have been wasting time on it. My intention was for the benefit of anyone coming along to these posts. I'm sure you are an expert problem solver. These things do happen.

[Pete Dowson 558]

5 hours ago, Ray Proudfoot said:

@simbol, I ran it with those 3 lines pasted into my cfg but no reports have been generated. I chose Yes when the DXGI_ACCESS_DENIED message appeared. I then tried Cancel to continue but the display remained black.

This is my windows and P3D version as reported by FSUIPC6...

Windows 10 Pro 64 Bit reported as Build 19041, Release ID: 2004 (OS 10.0)
Prepar3D.exe version = 5.1.12.26829

Nvidia drivers are 461.72

Baffling why the logs aren't being generated.

I think those three lines are ALTERNATIVE values for the same parameter. ie reporting level 1 2 or 3.  Only the first is probably recognised.

At least this is the way most CFG and INI files work, certainly FSUIPC ones using standard Windows APIs.

Pete

 

[simbol 5,819]

4 minutes ago, Pete Dowson said:

I think those three lines are ALTERNATIVE values for the same parameter. ie reporting level 1 2 or 3.  Only the first is probably recognised.

At least this is the way most CFG and INI files work, certainly FSUIPC ones using standard Windows APIs.

Pete

 

Pete is right, you had to enable only one of those settings.. My intention was to  show you what each setting does.. I didn't make it clear.. sorry Ray!.. being a tough week for me.. and I am having now brain farts lol

In any case, you found the cause.. so time to enjoy your simulator!!!

S.

[Ray Proudfoot 8,949]

39 minutes ago, Pete Dowson said:

I think those three lines are ALTERNATIVE values for the same parameter. ie reporting level 1 2 or 3.  Only the first is probably recognised.

At least this is the way most CFG and INI files work, certainly FSUIPC ones using standard Windows APIs.

Pete

 

I did wonder about that and commented out 1 and 3. But still no log. Strange why not but as the problem is solved it’s of no consequence now.

33 minutes ago, simbol said:

Pete is right, you had to enable only one of those settings.. My intention was to  show you what each setting does.. I didn't make it clear.. sorry Ray!.. being a tough week for me.. and I am having now brain farts lol

In any case, you found the cause.. so time to enjoy your simulator!!!

S.

Don’t worry Simbol. I appreciate the help. I managed to fit in a quick flight in the US and pleased to report all was well.

Apparently LM do read all the posts on their forums so I hope it’s something that can be fixed. I’m bound to forget to turn off RTSS one day being an old git. 😀

I hope you’re got your problem sorted.

[Bob Scott 7,720]

9 hours ago, Ray Proudfoot said:

PROBLEM IDENTIFIED AND FIXED.

After exhausting all other options I decided to close MSI Afterburner / RTSS as I've always had those running so I can monitor VRAM usage and other important aspects of P3Dv5.

Closing both programs has cured the problem. I can switch to other aircraft inside the sim without the DXGI error being generated. But delving deeper if I have both running but turn off the RTSS display before switching vehicles there is no crash.

So for those running Afterburner / RTSS make sure you turn off the screen display in RTSS before changing vehicles. Hopefully a fix can be found either by RTSS developers or LM.

Thanks @simbol for your help. As with many things the solution is very simple.

 

I couldda sworn I had already mentioned the problems DX12 on-screen displays were having with P3Dv5.

Ah...I did...but it was in a different forum/thread.

https://www.avsim.com/forums/topic/597550-at-my-wits-end-dxgi_error_device_hung/page/2/?tab=comments#comment-4501624

[Ray Proudfoot 8,949]

49 minutes ago, w6kd said:

I couldda sworn I had already mentioned the problems DX12 on-screen displays were having with P3Dv5.

Ah...I did...but it was in a different forum/thread.

https://www.avsim.com/forums/topic/597550-at-my-wits-end-dxgi_error_device_hung/page/2/?tab=comments#comment-4501624

Indeed you did a Bob but most errors seemed to relate to running out of VRAM and despite searching for the ACCESS_DENIED one very little came back.

Inflight things were fine and it’s such a bizarre error where you can load a completely different scenario with a change of vehicle, airport and weather and it’s fine. But change the vehicle to an avatar and it causes the error.

That’s why it was difficult to link it to the RTSS display. So innocuous but so deadly.  

[CanAmRyder 30]

Quick question guys. I've been following this thread and wanted to ask, what would you guys consider the best NVIDIA driver to install for P3D right now? The new drivers 465.89 has played havoc on my system and my sim so I need to reinstall an older driver.

 

Thanks,

Keith

[Ray Proudfoot 8,949]

@Grand_Pubah, 461.72 seem fine to me.

[CanAmRyder 30]

That's the one I installed. Thanks