Question PC freezes restarts and sometimes even crashes during high loads

[Salocin]

My pc will randomly, during high load activities like gaming or training a neural network for school, freeze up, restart or crash. So far I've only had one hard crash and one hard restart and a bunch of random freezes. So far I've not had it happen during idling or other low stress activities. I've even tested by setting some game settings to very low and it seemed to run without any issues.
This is happening randomly after I came back from a month long vacation. Before the vacation, my pc worked completely fine and I could run pretty demanding games on it no problem (even in the summer). But after a month of literally doing nothing with it (not even turning it on or off) it has these problems.

During a freeze the following happens:

• My monitors will randomly turn black and say "no signal found"
• After a few seconds of turning black, my monitors will turn on again and show a freeze frame of whatever I was seeing just before it lost signal.
• This freeze frame usually looks "worse" as in more grainy, some slight rainbow patterns around the edges and overall less "crips" of an image.
• During these freezes I can still hear sound. I was even in a call with some friend and I could talk with them completely fine.
• No matter what I do, I need to hard reset my pc in order for the freeze frame to stop.
• I can't find any error logs whatsoever in event viewer except for an error that says that the system restart unexpectedly

During a restart virtually the same happens as above, except that my pc just randomly restarts when it happens instead of it freezing.
When my pc crashed, it just out of nowhere crashed, the screen became pixelated and had some weird rainbow images around the edges. The sound also got stuck/stuttered.

So far I've done the following with no luck:

• Re-seat GPU and RAM
• Re-paste CPU
• Monitor temperatures with gpu-z (before and after cpu repaste) (both cases I found that my gpu hotspot temp and cpu temp got pretty hot, around max 95 degrees to sometimes even 100+ for short moments). I should mention that my pc seemed to handle these temperatures just fine in the past few years (atleast I think?).
• Do a windows ram mem test (nothing was found)
• Clean my pc

I genuinely have not clue what it could be or where to look. I've found some results online that said it could be a psu but I don't have a spare psu lying around and I don't really know how I can methodically test which part of my pc it could be. Any help on how to troubleshoot this problem is welcome.

System specs:

• CPU: intel i9-9900K ~3 years old
• CPU cooler: looks like a Cryorig CR4S, but I'm not sure what it exactly was ~5+ years old?
• Motherboard: ASUS ROG Strix z370-H Gaming ~6 years old
• Ram: Corsair CMT16GX4M2C3200C16 ~ 3 years old
• SSD/HDD: 1x Crucial MX500 2,5" 1TB SSD and 1x Intel 660p 2TB NVME Both ~2 years old
• GPU: NVIDIA GeForce RTX 3070 TI, HWinfo says it has the following model: PNY RTX 3070 Ti REVEL EPIC-X
• ~4 years old
• PSU: Corsair HX1000 ~4 years old but bought brand new
• Chassis: No clue, sorry.
• OS: Windows 11
• Monitor: 1x LG UltraGear 27GP850P-B (slightly older than a year) 1xancor vs248(6+ years old)
• BIOS version: American Megatrends Inc. 2701, 7/13/2021
• GPU Driver version: 581.08 (I've pretty religiously kept the gpu driver up to date with NVIDIA GeForce experience)

I've had my first pc built in a shop at around 2018. Over the years I've slowly replaced almost all of the components except my motherboard. I've first replaced my GPU and PSU at the same time then I replaced my CPU roughly less then a year later. For my CPU cooler, I've basically not changed the cooler since 2018 or 2020 (don't quite remember).
My memory is a bit hazy due to an accident I had during Corona. So do take the ages of these componenents more in a broad manner than an exact age.

As requested, here are some images I took of my pc
Some images of around 4 years back when I had recently replaced my parts and my NVME SSD broke down and some images now:

Spoiler: Older pc Images

https://imgur.com/a/pc

View: https://imgur.com/a/pc-stuff-jLV7vkx?s=wa

and

Spoiler: Newer PC images

https://imgur.com/a/pls

View: https://imgur.com/a/pls-dont-judge-wiring-WKDV9mW?s=wa

And finally the HWINFO output screenshot

Spoiler: here

https://imgur.com/a/Lqa0LNM

View: https://imgur.com/a/Lqa0LNM

Thanks for being patient with me, as you might have guessed I'm not very technically inclined

 

[Lutfij]

Welcome to the forums, newcomer!

System specs:
GPU: RTX 3070-TI
CPU: intel i9-9900K
PSU: corsair HX1000
MotherBoard: Asus Rog Strix z370-H Gaming
Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
include the age of the PSU apart from it's make and model. BIOS version for your motherboard at this moment of time.

What driver version are you on for your GPU?

Monitor temperatures with gpu-z (before and after cpu repaste) (both cases I found that my gpu hotspot temp and cpu temp got pretty hot, around max 95 degrees to sometimes even 100+ for short moments). I should mention that my pc seemed to handle these temperatures just fine in the past few years (atleast I think?).
Abrupt reboots when the system is taxed will happen due to one of two reasons;
1| an overheating issue
pr
2| your PSU is incapable of powering your entire system when taxed.

To also ask, what number are we looking at when you say the past few years?

 

[Aeacus]

I genuinely have not clue what it could be or where to look.

Plop out your GPU, connect monitor to MoBo and do your level best to get BSoD/freeze.

Now, you can't do GPU intensive tasks, but you can bench your CPU. E.g AIDA64, Prime95 and the like.

Also, when able, do test with 2nd, known to work GPU. Because symptoms show GPU issue (namely due the artifacts you've seen).

PSU: corsair HX1000

How old the PSU is, and was the PSU bought new or used/refurbished?

Better yet: Full system specs?
You're missing RAM, CPU cooler, storage dives. Knowing full system specs is critical. E.g you listed RTX 3070 Ti. Is it FE or AIB? If AIB, which model?

I should mention that my pc seemed to handle these temperatures just fine in the past few years (atleast I think?).

As PC ages, hardware also ages. Where it can't keep the same level of performance and low temps (efficiency), as when everything was brand new.

Monitor temperatures with gpu-z

Best use HWinfo64 for temps, since it easily shows max, average and current temps.
Link: https://www.hwinfo.com/download/

To test GPU, run Unigine Superposition in one of the Presets. E.g 1080p Extreme is torture test.
Link: https://benchmark.unigine.com/superposition

Monitor (or use logging feature) with HWinfo64 to see how temps/voltage/frequency etc change during bench.

IF the bench completes successfully, you can share your score. This way, your GPU can be compared to other same SKU ones, to see if it performs on average or not.

Do a windows ram mem test (nothing was found)

In terms of RAM, download and run memtest86,
link: https://www.memtest86.com/

Guide to install and use it: https://www.memtest86.com/tech_creating-window.html

1 full pass (all 15 tests) is bare minimum. 2 full passes are better while 4 full passes is considered acceptable.

Since it takes a while, best to let it run overnight.
1x 8GB ~1h per 1 full pass. ~4h for 4 full passes.
2x 8GB ~2.5h per 1 full pass. ~10h for 4 full passes.
2x 16GB ~5h per 1 full pass. ~20h for 4 full passes.

If there are no errors - RAM is sound.
If there are errors - replace the RAM.

For absolute testing of RAM, 32 full passes are needed (due to test #7).
Further reading: https://www.memtest86.com/tech_individual-test-descr.html
But no-one in their right mind is going to do 32 full passes, since it takes forever + then some. Especially when you have a lot of RAM in the system. Consensus in the tech support is that 4 full passes are enough.

 

[Salocin]

Plop out your GPU, connect monitor to MoBo and do your level best to get BSoD/freeze.

Now, you can't do GPU intensive tasks, but you can bench your CPU. E.g AIDA64, Prime95 and the like.

Also, when able, do test with 2nd, known to work GPU. Because symptoms show GPU issue (namely due the artifacts you've seen).


How old the PSU is, and was the PSU bought new or used/refurbished?

Better yet: Full system specs?
You're missing RAM, CPU cooler, storage dives. Knowing full system specs is critical. E.g you listed RTX 3070 Ti. Is it FE or AIB? If AIB, which model?


As PC ages, hardware also ages. Where it can't keep the same level of performance and low temps (efficiency), as when everything was brand new.


Best use HWinfo64 for temps, since it easily shows max, average and current temps.
Link: https://www.hwinfo.com/download/

To test GPU, run Unigine Superposition in one of the Presets. E.g 1080p Extreme is torture test.
Link: https://benchmark.unigine.com/superposition

Monitor (or use logging feature) with HWinfo64 to see how temps/voltage/frequency etc change during bench.

IF the bench completes successfully, you can share your score. This way, your GPU can be compared to other same SKU ones, to see if it performs on average or not.


In terms of RAM, download and run memtest86,
link: https://www.memtest86.com/

Guide to install and use it: https://www.memtest86.com/tech_creating-window.html

1 full pass (all 15 tests) is bare minimum. 2 full passes are better while 4 full passes is considered acceptable.

Since it takes a while, best to let it run overnight.
1x 8GB ~1h per 1 full pass. ~4h for 4 full passes.
2x 8GB ~2.5h per 1 full pass. ~10h for 4 full passes.
2x 16GB ~5h per 1 full pass. ~20h for 4 full passes.

If there are no errors - RAM is sound.
If there are errors - replace the RAM.

For absolute testing of RAM, 32 full passes are needed (due to test #7).
Further reading: https://www.memtest86.com/tech_individual-test-descr.html
But no-one in their right mind is going to do 32 full passes, since it takes forever + then some. Especially when you have a lot of RAM in the system. Consensus in the tech support is that 4 full passes are enough.

Thanks for the answer, I will do all of that. How do I find out my exact specs? As in, how do I find out my gpu model, what cooler I'm using etc. A lot of the parts are quite old, so I don't have the original box they came in anymore.

 

[Aeacus]

How do I find out my exact specs? As in, how do I find out my gpu model, what cooler I'm using etc.

Have them memorized? 🤔

E.g i have 3x desktop PCs and when i bought individual items, i put it all down in the list. Then again, i have all components memorized.
But i digress.

Is your PC prebuilt or DIY?

HWinfo64 Full mode tells what you have.

CPU cooler wise, well, i guess you have PC case with transparent side panel? If not, open up the PC and take a pic of the innards. Afterwards, upload pics to the net, e.g www.imgur.com and share them here.

From the pics, we can also see and identify GPU better. Especially when you take a pic where GPU fans are seen.
Same with PC case and fans. If you take pics of the fans. Now, most fans have label at the back, on the hub. But to see that, one needs to unscrew the fan, since back end of the fan is usually blocked by the PC case. Here, i don't think fan exact model is needed. Knowing fan size (120/140mm), their location and orientation will do. This would tell us if you have proper airflow or do you have fans blowing into each other.

E.g here's the pic of my main build, with side panel off. Easy to identify components in it, including airflow path:

Spoiler: Click here to view

Skylake build. Full specs with more pics in my sig.

Now, PSU age isn't listed anywhere, and for that, you need to tell us when you bought it and was it brand new or used/refurbished. Since else-ways, PSU age is considered maximum it can be, which is 8 years old (HX-series came out in 2018). Also, do you have regular HX-series or HXi-series? To confirm that, you need to look at the PSU's label.

 

[Salocin]

Welcome to the forums, newcomer!

System specs:
GPU: RTX 3070-TI
CPU: intel i9-9900K
PSU: corsair HX1000
MotherBoard: Asus Rog Strix z370-H Gaming
Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
include the age of the PSU apart from it's make and model. BIOS version for your motherboard at this moment of time.

What driver version are you on for your GPU?

Monitor temperatures with gpu-z (before and after cpu repaste) (both cases I found that my gpu hotspot temp and cpu temp got pretty hot, around max 95 degrees to sometimes even 100+ for short moments). I should mention that my pc seemed to handle these temperatures just fine in the past few years (atleast I think?).
Abrupt reboots when the system is taxed will happen due to one of two reasons;
1| an overheating issue
pr
2| your PSU is incapable of powering your entire system when taxed.

To also ask, what number are we looking at when you say the past few years?

Ah my bad, new to the forums so I have no clue what I should do, thanks for the headsup! Do you maybe have some tips or software I could use to figure out my more detailed specs like exact gpu model? I can't quite figure out where or how to find that. I'll update the main post with the updated specs.

 

[Lutfij]

GPU-Z and CPU-Z are two that come to mind.

 

[Aeacus]

Do you maybe have some tips or software I could use to figure out my more detailed specs like exact gpu model?

HWinfo64,
link: https://www.hwinfo.com/download/

 

[Salocin]

Welcome to the forums, newcomer!

System specs:
GPU: RTX 3070-TI
CPU: intel i9-9900K
PSU: corsair HX1000
MotherBoard: Asus Rog Strix z370-H Gaming
Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
include the age of the PSU apart from it's make and model. BIOS version for your motherboard at this moment of time.

What driver version are you on for your GPU?

Monitor temperatures with gpu-z (before and after cpu repaste) (both cases I found that my gpu hotspot temp and cpu temp got pretty hot, around max 95 degrees to sometimes even 100+ for short moments). I should mention that my pc seemed to handle these temperatures just fine in the past few years (atleast I think?).
Abrupt reboots when the system is taxed will happen due to one of two reasons;
1| an overheating issue
pr
2| your PSU is incapable of powering your entire system when taxed.

To also ask, what number are we looking at when you say the past few years?

I've never really paid attention to it in the past, but I've been noticing these high temps ever since I bought RDR2 for PC on release.
I've also updated my my post with more detailed specs and some images to make up for some specs/model numbers I just don't really know anymore.

 

[Lutfij]

Chassis: No clue, sorry.
Take pictures, host them on a site akin to Imgur and then pass on a link on this thread. If you're limited us with info, we'll be limited in how much helpful suggestions we can convey to you.
Just noticed you shoehorned links to imgur on your initial post. Looks like a Corsair Crystal 570x chassis. Your CPU cooler is on backwards, it should be blowing air out the back of the case...which is to the left of the picture. The cooler is also a Cryorig H7 Quad Lumi. I have one on my shelf + I also have my reservations about it's capability to cool an i9-9900K.

If I were you, I'd tear down the entire system and give it a good cleaning...like deep cleaning and then reassemble breadboarded and see if the issue persists.

You state having an HX1000, what is the color of the stickered side of the PSU?

 

[Aeacus]

I've had my first pc built in a shop at around 2018.

Whomever assembled it for you, did a botched job.

As stated by Lutfij, CPU cooler is wrong way. It needs to be turned 180 degrees. Not just the fan on it, entire heatsink.

Also, you only have 3x 120mm front intake fans, creating extreme positive pressure. And since CPU cooler is backwards, CPU cooler is currently blowing into front intake fans. This will lead into far higher internal temps.

Moreover, there is 0 reason why your RAM needs active cooling. And other than looking pretty, it is a waste of money.
For DDR4 RAM, even the heatsinks on individual DIMMs are just for the show.

I'd remove the most bottom 120mm fan and put it way back, as exhaust fan. This at least creates proper airflow path inside the PC case.

E.g like seen here, where there is fan at the back:

Spoiler: Pic

Direct link: https://assets.corsair.com/image/upload/f_auto,q_auto/content/cc-9011098-ww-570x-rgb-blk-panels.png

Also, you need to remove the RAM cooler since else-ways, you can't turn around the CPU cooler.
Whomever built it, most likely deliberately put the CPU cooler backwards, else-ways the RAM cooler would not have fitted.

CPU cooler: looks like a Cryorig CR4S, but I'm not sure what it exactly was ~5+ years old? HWinfo says it has the following model: PNY RTX 3070 Ti REVEL EPIC-X

PNY RTX is your dedicated GPU and not your CPU cooler.
Easy to read specs: https://www.techpowerup.com/gpu-specs/pny-xlr8-rtx-3070-ti-revel-epic-x-triple-fan.b8947

If I were you, I'd tear down the entire system and give it a good cleaning...like deep cleaning and then reassemble breadboarded and see if the issue persists.

👍

You state having an HX1000, what is the color of the stickered side of the PSU?

White label with black lettering.

If you look the old pics of the build, there you can see the PSU side.

I also have my reservations about it's capability to cool an i9-9900K.

It did well with i7-6700K,
review: https://www.techpowerup.com/review/cryorig-h7-quad-lumi/6.html

But i9-9900K is completely different beast. Most likely, not enough to cool it properly.

I'd replace CPU cooler with Peerless Assassin 120 or Phantom Spirit 120 EVO.

Edit:

Ram: Corsair CMT16GX4M2C3200C16

I just noticed in the HWinfo64 summary page that your 3200 MT/s RAM is actually running at 2133 MT/s.

You need to go to BIOS and enable RAM XMP to set it run at 3200 MT/s. This will greatly increase RAM performance.

 

[Salocin]

Whomever assembled it for you, did a botched job.

As stated by Lutfij, CPU cooler is wrong way. It needs to be turned 180 degrees. Not just the fan on it, entire heatsink.

Also, you only have 3x 120mm front intake fans, creating extreme positive pressure. And since CPU cooler is backwards, CPU cooler is currently blowing into front intake fans. This will lead into far higher internal temps.

Moreover, there is 0 reason why your RAM needs active cooling. And other than looking pretty, it is a waste of money.
For DDR4 RAM, even the heatsinks on individual DIMMs are just for the show.

I'd remove the most bottom 120mm fan and put it way back, as exhaust fan. This at least creates proper airflow path inside the PC case.

E.g like seen here, where there is fan at the back:

Spoiler: Pic

Direct link: https://assets.corsair.com/image/upload/f_auto,q_auto/content/cc-9011098-ww-570x-rgb-blk-panels.png

Also, you need to remove the RAM cooler since else-ways, you can't turn around the CPU cooler.
Whomever built it, most likely deliberately put the CPU cooler backwards, else-ways the RAM cooler would not have fitted.


PNY RTX is your dedicated GPU and not your CPU cooler.
Easy to read specs: https://www.techpowerup.com/gpu-specs/pny-xlr8-rtx-3070-ti-revel-epic-x-triple-fan.b8947


👍


White label with black lettering.

If you look the old pics of the build, there you can see the PSU side.


It did well with i7-6700K,
review: https://www.techpowerup.com/review/cryorig-h7-quad-lumi/6.html

But i9-9900K is completely different beast. Most likely, not enough to cool it properly.

I'd replace CPU cooler with Peerless Assassin 120 or Phantom Spirit 120 EVO.

Edit:

I just noticed in the HWinfo64 summary page that your 3200 MT/s RAM is actually running at 2133 MT/s.

You need to go to BIOS and enable RAM XMP to set it run at 3200 MT/s. This will greatly increase RAM performance.

Thank you so much for the answer and the patience. I will reset my cpu cooler and fan for now and buy a better cooler. Just so I understand: In caveman terms I have to individually turn both my fan and cpu cooler 180 degrees? And I have to remove the bottom pc case fan and install it somewhere in the back in such a way that it blows the air OUT of the case?
I’ve also fixed my rather stupid mistake of setting gpu specs as my cpu in my main post.

 

[Aeacus]

Just so I understand: In caveman terms I have to individually turn both my fan and cpu cooler 180 degrees? And I have to remove the bottom pc case fan and install it somewhere in the back in such a way that it blows the air OUT of the case?

Yes.

Note: you don't have to remove fan on the CPU cooler. All you need to to, is remove CPU cooler, remove RAM cooler, flip the CPU cooler around and put it back. With thermal paste repaste of course.

Here's how airflow goes for fans:

and buy a better cooler.

Take your pick,
pcpp: https://pcpartpicker.com/products/compare/BRMMnQ,66cgXL,ZMvscf/

Can't go wrong with either of the three.
Royal Pretor 130 is the best of the three in terms of cooling.
Phantom Spirit 120 Evo is good middle ground. Has some touches of RGB.
Peerless Assassin 120 is weakest of the three. But it has plenty of eyecandy (there is non-RGB model as well).

Since i'm in the process of upgrading my builds, these three coolers were what i considered. But i ended up going with Royal Pretor 130, since it is both better cooling and cheaper than Phantom Spirit. (My cooler is currently in shipping and arrives next week.)