Frustrating issue

[Inferno1295]

So first time ever posting here, hoping someone here can give me a solution 

I have an issue with my current PC whenever I play a more graphics intensive game the system just shuts off and restarts randomly with no error screen/ no error codes . I have replaced mobo, Graphics card and power supply with no significant changes. I don't think it's an overheating issue but I recorded an instance of it happening below just in case someone sees something I don't. This is driving me up the wall and I REALLY don't want to (can't) spend the money on buying a new cpu or building a new computer. I'll list all the troubleshooting steps I've taken below. 

 

Video: https://photos.app.goo.gl/nmMLWByMJebbRx4g8

 

I've switched gpu and ram slots 

replaced mobo,gpu and power supply 

ran furmark for 8 hours (no restart)

ran CPU tester for 12 hours ( no restart) 

ran both furmark and cpu tester for 30 min (restart occurred) 

The game that seems to trigger it fastest is Monster Hunter Worlds if that is of any help. 

 

Mobo:BaseBoard Product    ROG STRIX X299-XE GAMING

 

CPU: Processor    Intel(R) Core(TM) i7-7800X CPU @ 3.50GHz, 3504 Mhz, 6 Core(s), 12 Logical Processor(s)

 

GPU:Name    NVIDIA GeForce RTX 2080
 

PSU:EVGA 220-G3-0750-X1 SuperNOVA 750 G3, 80 Plus Gold 750W

 

RAM: 16 gig 

[Guest]

One thing you could keep an eye on is VRM temps. 

Generally an over heat shuts the board off and not a restart though.....

[Bombastinator]

howdy and welcome

 

in which I ramble on about your issue for a bit and offer up some ways to test memory (which may not be the problem)

Spoiler

This one comes up with fair frequency though the culprit changes.  Suspects are generally PSU, memory, or gpu. (Because it’s got memory in it) a rarer one might be thermals.  If your case happens to have really terrible airflow something could be throttling which causes issues.  It’s a lot less likely though.  My personal favorite is memory.  Not because it’s necessarily the most likely (I’m actually thinking maybe gpu because furmark triggered the problem but it should have done it faster) because it can be tested without additional equipment (even though it’s slow to do) and because if it IS the problem the fix costs the least money.

 

Two ways to test memory.  One is faster, the other is more accurate. 

The faster (but still slow) way:

run memtest86.  Several full passes.  The only problem here is it can throw false negatives.  It doesn’t always catch subtle stuff though.

the old fashioned and even slower but more accurate way:

put a single stick of memory in slot one and hit it with something hard for a couple hours, If it pases Then swap sticks and do it again.  The other one should fail.  If  The other one passes it’s memory.  RMA the kit.  If it doesn’t pass either it’s something else.  If they both pass then the problem went away by itself or the test wasn’t strenuous enough. 

 

[Inferno1295]

32 minutes ago, ShrimpBrime said:

One thing you could keep an eye on is VRM temps. 

Generally an over heat shuts the board off and not a restart though.....

Yeah I ran some gpu thermal tests, never hits above 80 if I remember correctly but it's in the video. The fact that it immediately auto restarts makes me almost positive it's not thermal. Thanks for the input tho! 

[Inferno1295]

20 minutes ago, Bombastinator said:

howdy and welcome

 

in which I ramble on about your issue for a bit and offer up some ways to test memory (which may not be the problem)

  Hide contents

This one comes up with fair frequency thought culprit changes.  Suspects are generally PSU, memory, or gpu. (Because it’s got memory in it) a rarer one might be thermals.  If your case happens to have really terrible airflow something could be throttling which causes issues.  It’s a lot less likely though.  My personal favorite is memory.  Not because it’s necessarily the most likely (I’m actually thinking maybe gpu because furmark triggered the problem but it should have done it faster) because it can be tested without additional equipment (even though it’s slow to do) and because if it IS the problem the fix costs the least money.

 

Two ways to test memory.  One is faster, the other is more accurate. 

The faster (but still slow) way:

run memtest86.  Several full passes.  The only problem here is it can throw false negatives.  It doesn’t always catch subtle stuff though.

the old fashioned and even slower but more accurate way:

put a single stick of memory in slot one and hit it with something hard for a couple hours, If it pases Then swap sticks and do it again.  The other one should fail.  If  The other one passes it’s memory.  RMA the kit.  If it doesn’t pass either it’s something else.  If they both pass then the problem went away by itself or the test wasn’t strenuous enough. 

 

Nice! I think the shop I brought it to tested the vram and ram. Their theory was that it was th vram in the gpu, but I got an RMA and asus replaced my 1080ti with a 2080, still the same issue which makes me think it's not the gpu, same thing with the psu. I'll test the RAM tonight. I have switched the sticks around and I think I remember only going on single sticks (I've been trouble shooting this issue since April) but I can try it again. 

[Bombastinator]

3 minutes ago, Inferno1295 said:

Nice! I think the shop I brought it to tested the vram and ram. Their theory was that it was th vram in the gpu, but I got an RMA and asus replaced my 1080ti with a 2080, still the same issue which makes me think it's not the gpu, same thing with the psu. I'll test the RAM tonight. I have switched the sticks around and I think I remember only going on single sticks (I've been trouble shooting this issue since April) but I can try it again. 

Re Gratz on gpu RMA:

Spoiler

Not a bad swap.  If you want to really put the screws to them and are willing to roll the dice you could tell them you’re a user of CUDA cores and this new one has less of them.  So they might send you a 2080ti or they might send you a 1080ti. Or they might tell you too bad.

Re: motherboard possibility

Spoiler

 Uhoh. If the gpu is known good and the memory turns out good it could be the connection to the motherboard that isn’t which means a possible motherboard rma. PSU still possible though.  

 

[Guest]

22 minutes ago, Inferno1295 said:

Yeah I ran some gpu thermal tests, never hits above 80 if I remember correctly but it's in the video. The fact that it immediately auto restarts makes me almost positive it's not thermal. Thanks for the input tho! 

Not the GPU the motherboard vrms.

[Inferno1295]

1 minute ago, Bombastinator said:

Re Gratz on gpu RMA:

  Hide contents

Not a bad swap.  If you want to really put the screws to them and are willing to roll the dice you could tell them you’re a user of CUDA cores and this new one has less of them.  So they might send you a 2080ti or they might send you a 1080ti. Or they might tell you too bad.

I thought about it but I don't wanna get too greedy, asus has done right by me for the most part and I was grateful for the upgrade. 

 

Re: motherboard possibility

  Hide contents

 Uhoh. If the gpu is known good and the memory turns out good it could be the connection to the motherboard that isn’t which means a possible motherboard rma. PSU still possible though.  

I got a motherboard RMA already and a PSU RMA. Thoughts on any issues with pcie lanes and cpu? The only parts I haven't replaced are CPU and RAM. 

 

[Inferno1295]

Just now, ShrimpBrime said:

Not the GPU the motherboard vrms.

Ah sorry yeah I didn't think of that, I think my airflow is pretty good and the fact that the machine turns back on immediately makes me think it's not thermal. 

[Guest]

5 minutes ago, Inferno1295 said:

Ah sorry yeah I didn't think of that, I think my airflow is pretty good and the fact that the machine turns back on immediately makes me think it's not thermal. 

Right, but without knowing actual figures, we dont really know. Guessing won't help diagnose anything ever.

 

But if you think it's a software issue, perhaps a fresh windows install on a different drive for testing could help.

[Bombastinator]

11 minutes ago, Inferno1295 said:

 

So PSU and motherboard also known good.  Yep if it’s not memory I got nuttin outside of vague references to win10 stuff (about which I know very little)  maybe check driver version numbers against what is current.  All I got.  (would suck if it was just a driver upset after all that RMA though lol.

 

there are people around here better than me at this though.  Someone else might have something.

[Guest]

1 minute ago, Bombastinator said:

So PSU and motherboard also known good.  Yep if it’s not memory I got nuttin outside of vague references to win10 stuff (about which I know very little)  maybe check driver version numbers against what is current.  All I got.  (would suck if it was just a driver upset after all that RMA though lol.

 

there are people around here better than me at this though.  Someone else might have something.

It's a weird issue. I've had DVD burners cause this issue. It litterally could be anything....

[Bombastinator]

A sudden thought:  what wattage is that new PSU? 2080 is a thirsty girl.

[Inferno1295]

1 minute ago, Bombastinator said:

So PSU and motherboard also known good.  Yep if it’s not memory I got nuttin outside of vague references to win10 stuff (about which I know very little)  maybe check driver version numbers against what is current.  All I got.  (would suck if it was just a driver upset after all that RMA though lol.

 

there are people around here better than me at this though.  Someone else might have something.

I really appreciate it, I've updated every driver I can think of. Thanks for all your help!

[Inferno1295]

3 minutes ago, Bombastinator said:

A sudden thought:  what wattage is that new PSU? 2080 is a thirsty girl.

750, but this issue was happening with my 1080ti and continued with my 2080 so I think I'm good there. 

[Deco_tc]

5 minutes ago, ShrimpBrime said:

Right, but without knowing actual figures, we dont really know. Guessing won't help diagnose anything ever.

 

But if you think it's a software issue, perhaps a fresh windows install on a different drive for testing could help.

I mean since you ran a Cpu stress test without fail, but it crashes in games that should be less cpu intensive, this is unlikely to be the issue.

Your motherboard is also made for overclocking so i don'T think this should be the issue.

[Guest]

1 minute ago, Deco_tc said:

I mean since you ran a Cpu stress test without fail, but it crashes in games that should be less cpu intensive, this is unlikely to be the issue.

Your motherboard is also made for overclocking so i don'T think this should be the issue.

Well a cpu test isnt gaming. Cant even compare.

 

Gaming utilizing everything from the SSD, memory, cpu, NB, power supply. All of it. This is a system load and not a single cpu running the same calculation over and over.....

 

So you start trying different hardware. Unplug the SSDs and use a different drive for testing. A fresh install. Test the game then.

 

Could be even be usb device causing a restart. 

 

But with little to no symptoms and no error reports to go by, diagnoses for us helping is just a bunch of guessing.

 

 

 

[Deco_tc]

Every driver update, so chipset drivers and bios too then? Then it seems like either a memory or windows issue.

Other than that no externals should cause ONLY crashes in games. If it was a weird compatability issue it would probably crash in normal windows applications aswell.

System reset is allways an option and maybe changing some windows power states to high performance aswell, although its rare that that causes straight up black screens and nothing else.

[Deco_tc]

1 minute ago, ShrimpBrime said:

Well a cpu test isnt gaming. Cant even compare.

 

Gaming utilizing everything from the SSD, memory, cpu, NB, power supply. All of it. This is a system load and not a single cpu running the same calculation over and over.....

 

So you start trying different hardware. Unplug the SSDs and use a different drive for testing. A fresh install. Test the game then.

 

Could be even be usb device causing a restart. 

 

But with little to no symptoms and no error reports to go by, diagnoses for us helping is just a bunch of guessing.

 

 

 

Well it does tell you that it is probably not the cpu vrms which i meant.

[Bombastinator]

7 minutes ago, Inferno1295 said:

750, but this issue was happening with my 1080ti and continued with my 2080 so I think I'm good there. 

750 more than enough.  Not enough power isn’t the problem.

[Guest]

Just now, Deco_tc said:

Well it does tell you that it is probably not the cpu vrms which i meant.

Yeah, I get that. Already acknowledged that a few posts back.

[Inferno1295]

8 minutes ago, Deco_tc said:

Every driver update, so chipset drivers and bios too then? Then it seems like either a memory or windows issue.

Other than that no externals should cause ONLY crashes in games. If it was a weird compatability issue it would probably crash in normal windows applications aswell.

System reset is allways an option and maybe changing some windows power states to high performance aswell, although its rare that that causes straight up black screens and nothing else.

I haven't changed any of my power states but I have done a windows reset. Maybe I just need to reformat everything and start from scratch? I'd rather not do that but I just want a working computer. 

[Bombastinator]

4 minutes ago, Inferno1295 said:

I haven't changed any of my power states but I have done a windows reset. Maybe I just need to reformat everything and start from scratch? I'd rather not do that but I just want a working computer. 

*crosses fingers for memory errors*

[Deco_tc]

And 

1 minute ago, Inferno1295 said:

I haven't changed any of my power states but I have done a windows reset. Maybe I just need to reformat everything and start from scratch? I'd rather not do that but I just want a working computer. 

And the windows reset also didn't fix anything? Well then there are only a few things left. You can give the power settings a try, setting it all to high performance, although it would suprise me if that worked (you never know). Other than that there is still the memory or you got really unlucky and they shipped you a broken part twice.

I would run memtest86 next.

[Inferno1295]

2 minutes ago, Deco_tc said:

And 

And the windows reset also didn't fix anything? Well then there are only a few things left. You can give the power settings a try, setting it all to high performance, although it would suprise me if that worked (you never know). Other than that there is still the memory or you got really unlucky and they shipped you a broken part twice.

I would run memtest86 next.

I'm on my 2nd GPU RMA so it would be shipped a bad product 3 times. And I'll do the power settings now then run memtest. And no reset didn't fix anything.