[Solved] Aurora R8, 4 games crash, different reasons, cause?

Have you tried monitoring GPU temps and fan speed to see if the GPU is overheating during game play.  

@GTS81 Yes, never run with overclock. Warranty is till January. 

Only things purchased after was the 32 GB Ram from Ballistix which I check compatibility, as well as a Crucial MX500 500 GB SSD.

I ran the Heaven Unigine test before didn't find any problem there. 

The only thing I haven't try physically is to use my RTX 2080 on another PC. I will try it out when my friend got his PC later this month. 

@Covollel , in order to rule out issue with specific game(s) or the DirectX API they are using, you may want to use a 3D benchmark software and run their stability test for > 30 minutes to see if there's overheating/ circuit marginality issue related to your GPU. I presume you have never made any hardware mods to your R8 before and it is still within the 12 month warranty?

@r72019  Yes, I been monitoring the GPU temperature from AWCC, which the highest GPU temp was around 85°C, I thought that is normal? 

Given that you have several stack dump codes, I'd recommend someone like @Tesla1856 or @speedstep to chime in on this topic.

85C may or may not be ok. Personally for me, it's encroaching territory for marginal hardware instability when sustained over a long time.

You mentioned the games crashing and all but one is from Steam? I'd like to rule out Steam first so if you have one game you can run on that is not on Steam (I know, all but 1 of my games, FH4, are on Steam! ), you may want to turn on an FPS limiter through NVIDIA software, say cap it at 50 FPS, and then run the game on various resolutions and settings. The rationale to doing this is a "less-artificial" benchmark but at the same time creating a graded approach to stressing your GPU. I'd expect the temps to start out low for low settings and resolution, and then working your way up to the toasty 85C. If you can play at lower settings i.e. lower temps for an hour or so, then you would be able to narrow down the issue to hardware temperature related.

Also, you have another GPU on your system: the integrated graphics . You may want to try that out on the game (sorry for the headache running on single digit FPS for 30 minutes!).

I'd recommend you get a different GPU and plug it into your system than to take your GPU to plug into your friend's. This way you get to keep more variables static in the debug process.

In the unlikely case your SSD or RAM upgrade may be the issue, here are my recommendations:

1. RAM - revert to stock RAM shipped with your R8 OR use 1 stick of the DDR, assuming you bought 2x16GB.

2. Ensure that your SSD has more than 50% disk space. Best to run your system software from the fastest SSD but other software from a different SSD/ HDD.

Lots of games require DOTNET 2.0 and 3.5 as well as VISUAL C++ and Directx June 2010 and REMOVAL OF MCAFEE OR NORTON and many other pieces of nonsense.  

Older games like fallout 3 and 4 and 76 as well as CSS and others will require quite a bit of tweaking and installing.

NVIDIA GPUs are designed to operate reliably up to their maximum specified operating temperature. This maximum temperature varies by GPU. Refer to the nvidia.com product page for individual GPU specifications. If a GPU hits the maximum temperature, the driver will throttle down performance to attempt to bring temperature back underneath the maximum specification. If the GPU temperature continues to increase despite the performance throttling, the GPU will shutdown the system to prevent damage to the graphics card.  Up to 85°C is normal when playing games.  94 TO 105°C is absolute max.

@GTS81  All great ideas! Thank you so much!

I have two games I play a lot not from Steam. 

D3, which is the most stable out of all. Even though the LiveKernelEvent crashes won'st stop the game, just have a 2-3 second of pause then carry on as normal.

COD MW I just bought last weekend from Battle.net, nothing but crashes, both of the crashes are well known crashes so I am not saying it is due to the GPU... yet...

I will try them out over the weekend and get back to you guys for any findings.

@r72019  Also worth noting is that when I run the Heaven Unigine 4.0 on ultra setting, the screen would turn black a few times for 2-3 seconds, but the benchmark would continues after that, is that normal? When I check the reliable report, it also logged LiveKernelEvent 141 

Try running World of Warcraft using the free mode up to level 20.

It should not die and it will work your gpu if you set everything to ULTRA.

I was reading thread, but I think @GTS81 pretty much covered it (since it exhibits like hardware problem)

- Return to original factory ram config (DIMMs and slots)
- No Over-Clocks
- Try a different SSD (because you messed with factory drive config).
- Try a different video-card

However, I have seen my Aurora-R6 act like it had a hardware problem before (hard-lockup without BSOD) ... but it was just bad nvidia video-drivers.

https://www.dell.com/community/Alienware-General-Read-Only/Aurora-R6-Hard-Lockup-and-crash-while-gaming-SOLVED/td-p/5504111

To remove Steam and games from equation, I stress-test and burn-in system with Power-Supply test in OCCT . I've run it for 30-minutes straight on various systems ... pushes them hard but works fine.

Re-imaging from pre-existing Dell Recovery partition does not really count as a clean Windows install:

https://www.dell.com/community/Alienware-Desktops/Aurora-R7-M-2-NVMe-bootable-options/td-p/6073037

It really could be anything. Make system as lean and simple as possible. Once you find parts that are 100% stable together, work-forward from there.

Thank you all for the replies! @Tesla1856 @speedstep @GTS81 @Anonymous 

I think it may indeed been thermal issue.

I only got one hour to test yesterday, so here is what I did:

- Only using AWCC to limit the GPU temp cap at 80°C, it still go up to 83°C during my test.

- Lower the Core clock for -100 mhz; lower the memory clock for -50 mhz

- In COD MW, everything still at highest setting, but lower the rendering from 1440p to 1080p and lower the resolution from 1440p to 1080p

I was able to run the game for about 1 hour without any issue.  

I will do more test when I got a chance on another game, with different conditions. 

But I guess it lead to another problem. 

If I don't limit the clock speed or temp, my stock RTX2080 climbs to 85°C quite easily, everything else is stock except for my ram and additional ssd. Anyway to cool it down? 

Thank you!

If I don't limit the clock speed or temp, my stock RTX2080 climbs to 85°C quite easily, everything else is stock except for my ram and additional ssd. Anyway to cool it down? 

You'll have to tune those fan curves in AWCC until you get the best results. If it ends up with fans blasting like an airplane taking off, you would need to look at liquid cooling the GPU. Changing the RTX2080 card for another RTX2080 card from 3rd party is unlikely to help because open fan style cards will end up dumping more hot air into your R8's chassis.

If Dell states that 85C is normal operating temperature, and yet your games fail, then wouldn't that warrant a swap of the graphics card as it is considered to be defective under warranty?

@GTS81 Right, after I got another game into test I will call them up again see what kind of excuse they come up this time.... Also worth noting is whether I turn the fan to max or let it be "auto", the GPU temperature doesn't seem to change. 

My system already have liquid cooled CPU, I doubt I have room for another radiator for the GPU in the R8 chassis? 

Thank you so much!

Why open fan card doesn't work: https://www.dell.com/community/Alienware-Desktops/Aurora-R8-Experience-of-Buying/m-p/7351136/highlight/true#M11595

GPU radiator hack: https://www.dell.com/community/Alienware-Desktops/Aurora-R8-Experience-of-Buying/m-p/7352232/highlight/true#M11643