Self-build - BSODs like clockwork

Associate
Joined
29 Apr 2020
Posts
9
Hi all,

First time self-build gaming rig here. Got my system all set up last week and everything was great, but a couple of days ago started getting BSODs on certain games (and sometimes just browsing on Chrome).

These are the stats:

MB: Z390 AORUS PRO WIFI
CPU: i5-9600K at factory clock speed
RAM: 4 x 8GB Corsair 2400Mhz RAM
GPU: GeForce GTX 1070
Storage: 2 x Crucial M.2 1TB SSD
PSU: Corsair TX650M 80+ Gold Semi Modular Power

I've updated all drivers and the BIOS and run memtest which passed without exception. I also run a userbenchmark test and it passed with flying colours.

It may be a coincidence, but this only started happening after I had problems with my integrated wifi controller, which I eventually fixed.

The system has never really run under stress, and temps are all good from what I can see.

The BSOD code is MACHINE_CHECK_EXCEPTION, so assuming it's a hardware problem. Regardless, I've tried a fresh installation of Windows, but can still reproduce the problem consistently. I've also tried many different RAM combinations, but each has the same outcome.

The GPU is second hand from a mate of mine who never had any problems with it.

Since I started the troubleshooting process, I checked my ************ list again, and noticed a compatibility note which I'd previously missed. Apparently my MB has an additional 4-pin ATX socket in order to recieve an additional 12v, but my PSU doesn't have that connector. According to another forum, that's unlikely to be a problem unless I start trying to massively overclock, which I'm not planning on doing.

Weirdly, I've also notice my second M.2 SSD isn't showing in Windows Explorer, but it's definitely present when I open up diskpart. Is there something I need to do to enable it?

Any thoughts?

Thanks,
Kohota
 
Man of Honour
Joined
16 Mar 2005
Posts
8,060
Location
Clevedon , Bristol
Additional 4 pin connector - No problem as you have already found out.

If you can recreate the problem consistantly - Have you tried totally disabling the wi-fi controller to see if this is the issue ? ( as the issue seems to be linked to that problem )

Missing SSD probably just needs initialising - Go to Disk Managment ( in Windows ) then see if the ssd is listed there . then just initialse/allocate
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Thanks Brizzles.

Just tried that, and the same crash at the same time.

What's infuriating is that I'm able to play a game like Mount & Blade: Bannerlord for hours at a time without a problem. But then I boot up Europa Universalis 4 - not the most graphically demanding game by any stretch - and it crashes within a minute or so.

And thanks for the heads up on the drive. I've managed to get the partition set up now.
 
Associate
Joined
1 Dec 2015
Posts
1,194
The missing SSD would be my first thing to sort out.
Do you get the same problems with it unplugged?

Also check your chipset drivers are correct. Try direct from Intel and Gigabyte (may be different versions) This might also sort the SSD problem.

Try some benchmarks to try and narrow it down a bit.
Its also worth putting a screen overlay of GPU,CPU and memory usage. You may see that Europa uses one resource more heavily that Mount+Blade.

You could also try downclocking the CPU RAM and GPU to narrow it down.

Also try an alternative PSU (borrow one from your mate?)
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Thanks guys, I'll try some different games to see if it's specific to EU4 - I'm sceptical as I've been getting BSODs when simply browsing the web on Chrome.

The SSD is fine, I just hadn't initialised it properly.

With regards to drivers, that was the first thing I did after installing windows.

All hardware is running at factory clock speeds, and as I said before, temp is absolutely fine. I will see if I can get an alternative PSU, but again, I'd be surprised by this given the crashes can be reproduced by following specific steps. I'll try a resource overlay to see if I can pin it down.

I have a suspicion that there is some kind of specific hardware resource that is only accessed in specific scenarios which is where the fault is, but identifying it seems next to impossible.
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Ok, so no idea what's going on here. Just played four hours of Rainbox Six Siege using ultra-high settings, and the system barely started sweating. CPU/RAM load and utilisation looks absolutely fine, and the hottest any component got was my GPU at 71c.

Seriously stumped on this one.
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Rinbow 6 Siege is about as old as your GPU, maybe an issue with new demands from games

COD: Warzone isn't, and I've just managed three hours of that on high graphics settings with silky-smooth FPS.

Looks like it's a problem with the EUIV specifically. Sorry guys, it's clearly sent me on a bit of wild goose chase.
 
Associate
Joined
23 May 2011
Posts
124
Have you tried setting the voltage to manual in the bios instead of auto, I had a similar issue with an older Aorus motherboard, once I set the voltage to manual no more BSOD and could play games with no interruptions
 
Associate
OP
Joined
29 Apr 2020
Posts
9
So I'm still getting the occasional crash. Again, weirdly only on games that would usually run on a potato - Kerbal Space Program, Tabletop Simulator etc.

I also just had a crash when I was just surfing the web on Chrome - but weirdly, no BSOD this time, just an instant restart.

Have you tried setting the voltage to manual in the bios instead of auto, I had a similar issue with an older Aorus motherboard, once I set the voltage to manual no more BSOD and could play games with no interruptions

Thanks Tinkers, I've gone into the BIOS and had a look, and yes, all the voltages are set to auto. The problem is, there's about 12 different voltage options and I have no clue what I'm doing.

I changed the CPUVcore and DDR CH(A/B) to the 'normal' setting, but I'm not sure if this is a manual override or simply a slightly different version of the 'auto' setting.
 
Associate
Joined
25 May 2014
Posts
166
Location
Cornwall
Hi, i've got the same board and had endless issues with crash's in games, temps ect were fine while benchmarking and looked ok in games, I think i've finally sorted mine now,turned out i think/hope it was dodgey thermal paste on a new CPU cooler I bought.

I was getting constant crashes and endless different event viewer reports. Since changing the pre applied paste one crash in 3 days...so far.

I don't know if it will help but you could try,

Are you running RGB Fusion or the Aorus engine? You could try uninstalling them and trying again if you are, seems to cause a lot of instability for some people.

Have you installed a game called valorant by any chance? as i've read with that and rgb installed has caused people to have issues.

Reapply some thermal paste if you havent already. Worth a try, was for me anyway.

Have you got XMP running for the ram? you could try disabling it and see if it makes any difference.

Memtest86? Maybe give that a go and test your memory to rule that out..

Good luck, ive just had a month of crashing myself, so feel ya pain :)
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Yup, have run memtest 86 - passed with flying colours.

I had RGB fusion installed on my previous windows install, but didn't bother this time round.

Interesting that you mention the thermal paste... I made a complete hash of attaching the CoolerMaster, the first time round, so there might be something in that...
 
Soldato
Joined
28 Aug 2017
Posts
2,797
Location
United Kingdom
MACHINE_CHECK_EXCEPTION

may need to update your bios, looked on a few sites with the above error and it's drivers out of date, windows not fully updated hardware issues (from above i'd say unlikely) and finally a bios update.

check your manual on how to do it, these days it's pretty easy, if you get stuck pop on and someone will help no doubt :)
 
Associate
Joined
24 Jun 2003
Posts
1,642
Thanks Tinkers, I've gone into the BIOS and had a look, and yes, all the voltages are set to auto. The problem is, there's about 12 different voltage options and I have no clue what I'm doing.

I changed the CPUVcore and DDR CH(A/B) to the 'normal' setting, but I'm not sure if this is a manual override or simply a slightly different version of the 'auto' setting.
If you have changed something in the BIOS and are unsure I would try loading the optimised defaults.
As for in windows , open device manager and make sure you have no yellow question marks, and every thing is installed ok.
Also did you try running just 2 sticks of ram , yes I know you've ran memtest , but give 2 sticks a try.
 
Associate
Joined
23 May 2011
Posts
124
use a program like cpu z to display the voltage when you put it under load (use something like aida64 so put it under load), and set it in the bios to a value just above that example if it is showing 1.22v set it at 1.25v, don't go above 1.3v unless you know what you are doing in the bios and have suitable cooling, also disable xmp settings for your ram and run that at standard and see if you are still getting issues. Hope this helps
 
Associate
OP
Joined
29 Apr 2020
Posts
9
Thanks Tinkers, I've updated the BIOS, but not I'm getting CLOCK_OVERWATCH_TIMEOUT on some CPU heavy games where I wasn't previously.

I've just upped the voltage to 1.25 to see if that helps.

I've also forced the RAM clock speed to 2400Mhz which matches the actual stick speed, as it was defaulting to 2133Mhz. Let's see how it goes...
 
Last edited:
Associate
OP
Joined
29 Apr 2020
Posts
9
Hi all, thought I'd tie this one off in case anybody with a similar problem stumbles upon the thread.

Flashing the bios and jacking the voltages seems to have worked, touch wood. No new BSODs since the changes, and managed to play a good 6 hours of EUIV the other day with zero problems, and excellent performance.

Thanks all for your advice.
 
Back
Top Bottom