PowerSpec G900 CTDs and BSOD while playing games

TangoCheesecake

Hi, I purchased a G900 last week and have been having some problems since then, primarily involving CTDs and BSODs. The BSODs have all been for various reasons (all have occured while playing games) and the game crashes seem to occur at random. I was initially thinking that maybe I was having a PSU issue, that the issues were specifically related to New World (but it has happened in various games now), and/or that the issues have been driver related--but I am honestly not sure at this point since I am not an expert. Crashes have been happening enough that it is becoming an annoyance, and I would like to try and resolve the issues sooner rather than later (and, although this may not be avoidable, without driving to a Micro Center since the nearest one is a 4 hour round trip).

I have been checking the reliability history and the event viewer, but I am not exactly sure how to read them. I do know that, with respect to the reliability history, not all crashes are showing up (at least not for New World). Again, however, this is not strictly a New World issue as the CTDs have been occurring in other games, though the BSODs have primarily been related to New World.

Finally, with respect to potential driver issues I have check as many drivers as I can and have made sure that they are updated. I have also made sure that Windows is updated and have numerous scans through the the Command Prompt (which initially found and fixed some apparent Windows issues, but this did not fix the crashing or BSODs). Finally, I tried the Windows Driver Verifier, which put me in a boot loop before I was finally able to get out of it, which suggests to me that there is a driver issue at least somewhere (again, however, I am definitely not an expert--when I used the Driver Verifier I only checked non-Microsoft drivers).

Any help with this would be greatly, greatly appreciated. I am sure I have planning of crash logs that I could post if that would help, although I might need a bit of guidance on accessing them and posting them correctly so that they are actually helpful. Thank you!

EDIT: I just thought to add this; at least two of the BSODs I believe were either directly memory or driver related (again, not an expert--sorry!): One was an IRQL_NOT_LESS_OR_EQUAL BSOD, and the other was a SYSTEM_SERVICE_EXCEPTION. There were a few others, but I do not remember what they were off the top of my head. I know one of my Windows crashes (which was not a BSOD, but when I got in the boot loop), was related to the Driver Verifier.

Ian

Greetings, we can look into this with you.

Please run Bluescreenview on your PC and post screenshot(s) / post the descriptions of what bluescreens you have seen: https://www.nirsoft.net/utils/bluescreenview-x64.zip

TangoCheesecake

Apologies for the delay, here Is what I screenshotted! Please let me know if this isn't what you are looking for. Thank you!

Ian

If all these crashes are while gaming, please try a clean install of your video card drivers, here is a step by step guide on how to do this: https://community.microcenter.com/kb/articles/133-how-to-clean-install-video-card-drivers-in-windows-10 

Once done with this process, use the system as normal and see if the issues persist.

TangoCheesecake

Hi Ian, thank you for that--I will try that this evening once I am at my desktop.

I have a few other quick things to add: I may have missed this in reading the guide that you posted, but is there any concern in doing a clean install over the fact that I have no integrated graphics card?

I also have a little bit of additional information--although I am not sure if it is useful at all:

1. My internet browsers (both Firefox and Edge) have crashed when watching Twitch streams on my second monitor while I was playing games on my first--to the point that I cannot even reopen either browser without restarting my computer.
2. In the event viewer and reliability history I am seeing that "lightingservice.exe" is crashing/failing to start whenever I start my computer. I know that this is an ASUS program related to my RGB (which is still working), but I unsure as to whether this is leading to any additional issues. I have not tried uninstalling and/or re-installing the program yet, but I also plan on doing that I figure out how to (I have read that other people have had problems completely uninstalling the program, which apparently requires uninstalling Lightingservice and Aura).
3. Finally, I read on the New World forums that one potential options to prevent crashes in that game is to manually increase the available virtual memory to 1.5 to 3x the amount of physical ram, so I did that as well. I don't have the exact number on-hand, but I know that the amount of virtual memory available seemed low to me, around 4,500mb or so (whereas now the minimum is 1024*32*1.5 = 49152 and the maximum is 1024*32*3 = 98304). Since I did this yesterday, the game has not crashed, but the webpages with Twitch streams on have continued to crash. As one final note, although the game itself did not crash, it did stall when exiting such that I had to restart my computer as I got a Windows error (but not a BSOD).

Thanks again! I apologize if I am just bombarding you with a bunch of useless information, I just want to make sure I provide you with any potentially relevant information. I will reply to this post later tonight once I have tried the clean install of the GPU and have let the PC run for a while.

TangoCheesecake

No dice after clean installing the GPU driver. Still getting issues with browsers crashing while watching streams (re-installed Firefox as well), but got another BSOD while playing Far Cry 6. Here is the screenshot from that one:

At this point I really have no idea if it is driver related or hardware related (RAM or the SSD according to cursory, amateur research). Thanks again for the help so far; I really hope we're able to get this sorted out sooner rather than later!

PowerSpec_MikeW

@TangoCheesecake

Random BSOD's generally indicate hardware. Start with Memtest86: https://www.memtest86.com/

Run four passes, it'll take a few hours. From the BSOD's, that's what this looks like to me.

TangoCheesecake

@TSMikeW

Thank you for the response. Below is the screenshot--no memory errors. I will do another run tomorrow while I am away from the computer, and I will try to configure it to do 8 passes. I have read that sometimes higher temps can cause the errors and that more passes is generally better anyways--so we'll see if another Memtest run comes with up anything that slipped through the first time.

Any additional tips or steps I should take at this point (update bios, anything to check the SSDs, etc.)? As much as I would like to avoid having to take the long drive to the nearest Micro Center, I am beginning to suspect that might be all I can do, though I would like to take any other steps necessary that might solve the problems that I am having before doing visiting the store (which I would likely try to do on Friday, if needed).

PowerSpec_MikeW

@TangoCheesecake

Lets try something else before we move off the RAM. Go into the BIOS(Delete) and to the AI Tweaker tab. Change AI Tuner from DOCP to Auto. Game for a while and see if the issue remains.

Have you upgraded this system to Windows 11? In search type "Core" and open "Core Isolation". Is Memory Integrity on or off?

TangoCheesecake

@TSMikeW I have just made the changes within BIOS--I will let you know if there are any different results under the Auto settings by making a separate reply to this post. As a side note, I ran another memtest and it passed (it was, however, only four passes). Finally, I am still using Windows 10, and "Memory Integrity" is turned off (I have not changed this setting at all).

Thanks again!

TangoCheesecake

@TSMikeW So far, with the BIOS/AI Tuner change you suggested, I have had no game crashes, BSODs, or browser crashes (about 8ish or so hours of use). Does this suggest that the ram is what was ultimately (or at least likely) causing the issues, and that it is not able to consistently, without crashing, run up 3200mhz? I am no expert by any means, just trying to make sure I am understanding things correctly--so please correct me if I am wrong.

PowerSpec_MikeW

@TangoCheesecake

It does, but there's another factor to consider. The IMC on the CPU itself. If we had some failures in Memtest86, based on the failing test we could more accurately ascertain which may be the culprit. You might try running memtest overnight for a few days, try to catch the failure. I would also try this. Set AI Tuner back to DOCP, but set your frequency to 2400. Give it a day or two. If you don't crash increase to 2666, 2933, 3000. We can try to figure out where you're getting unstable. The DOCP profile will loosen the timings, as you're probably defaulting to 2133 CL15.

TangoCheesecake

@TSMikeW Apologies for the last response, it's been a busy week and I have been progressing through the various frequencies. After running for several hours each at almost every frequency between 2400 and 3200 I have had no crashing issues. I am currently testing it out at 3133 and will move back to 3200 mhz if 3133 proves to be stable. In any case, I will update this thread further when I have more information, but I figured this information might help you. Thank you!

TangoCheesecake

PowerSpec_MikeW

@TangoCheesecake

Interesting, let me know what your results are. Sometimes you can have a system that'll fail to POST on XMP but you can prime it and get it to POST by basically doing what I told you to. Loading XMP and changing the frequency. It effects the way it trains secondary and tertiary timings. Seems like you're on the edge of stability with the defaults loaded. We test the RAM extensively on the platform, to double the capacity. It strikes me as an IMC issue.

What you can do to confirm. Is load defaults, load the XMP then test again. The crashing should return as it was before. Then you can prime as you've done and work it back up to stability.

TangoCheesecake

@TSMikeW Apologies again for the late response--3133 was stable (and that is what I am currently at). I only tested again at 3200 for an hour or two during which the game I was playing unexpectedly crashed twice and my browser crashed once. No BSOD, but I think I just didn't state at 3200 long enough for it to BSOD.

I apologize, but I do not think I am quite understanding the process that you suggest I try; are you saying change the settings back to DOCP (changing from auto) but then still manually setting the frequency like I was before--starting again at around 2400--and then work back up to 3200 again?

I also wanted to confirm, DOCP / XMP are essentially the same thing for what we discussing, right? It's just DOCP because it is an AMD CPU (probably oversimplying it, but I just want to make sure I am understanding it well-enough for what I will be doing)?

Thanks again!

PowerSpec_MikeW

@TangoCheesecake

If you're failing and seeing instability at 3200 after you primed it, this sounds like a defect. I would try to confirm that the system is unstable at 3200.

Yes, XMP is technically Intel, but almost everyone, except ASUS, will use it as an acronym on AMD. I believe you'll see A-XMP on Gigabyte boards, as an example.

TangoCheesecake

@TSMikeW Gotcha, thank you. I will try priming it (using the process I described in my previous post--I believe that is what I am supposed to do, but please correct me if I am wrong) and get back to you. Thank you again for all the help so far.

PowerSpec_MikeW

@TangoCheesecake

You're correct. Set DOCP, lower frequency to 2133. Save and exit. Walk it up to 3200 then test of stability.

TangoCheesecake

@TSMikeW Apologies for the delay--I did as instructed and it continued to crash at 3200

PowerSpec_MikeW

@TangoCheesecake

Unfortunately at this point I'd have to advise bring it in for service. You're on the edge of stability at 3200. This sounds like a memory controller issue to me.