Question New PC keeps crashing frequently when gaming ?

[Wouter_U]

Hi,

Recently I bought and built my new pc. But ever since it's been nothing but trouble.
Whenever I'm playing a game that really puts a load on the system, the pc very often just crashes out of nowhere. It does not give a blue screen. it either freezes and does nothing anymore or it spontaniously restarts without any warning or error.

My specs:
CPU: AMD 7800X3D
GPU: Nvidia RTX 4080 Super
Motherboard: Asrock A620I
RAM: 32Gb Corsair Vengeance at 6000hz
PSU: Cooler Master V SFX Gold ATX 3.0 850W
Case: Fractal Ridge

I've tried a lot to try to find the problem and fix it, but I'm still no closer then when I started.
Things tried:
- change PCI mode to gen3
- take side panels off to see if PSU was choking
- checked temps (are all fine)
- made sure the Nvidia power connector has enough room (even with panels off it still crashes)
- updated everything. From drivers to Windows itself.
- reinstalled enitre pc
- Checked Windows error log, could only find kernel error which said the power cut off unexpectedly.

I don't know what to do anymore and I really need to find out what the issue is exactly because I don't want to have to send the entire pc back.

 

[kanewolf]

Hi,

Recently I bought and built my new pc. But ever since it's been nothing but trouble.
Whenever I'm playing a game that really puts a load on the system, the pc very often just crashes out of nowhere. It does not give a blue screen. it either freezes and does nothing anymore or it spontaniously restarts without any warning or error.
My specs:
CPU: AMD 7800X3D
GPU: Nvidia RTX 4080 Super
Motherboard: Asrock A620I
RAM: 32Gb Corsair Vengeance at 6000hz
PSU: Cooler Master V SFX Gold ATX 3.0 850W
Case: Fractal Ridge
I've tried a lot to try to find the problem and fix it, but I'm still no closer then when I started.
Things tried:
- change PCI mode to gen3
- take side panels off to see if PSU was choking
- checked temps (are all fine)
- made sure the Nvidia power connector has enough room (even with panels off it still crashes)
- updated everything. From drivers to windows
- reinstalled enitre pc
- Checked windows error log (could only find kernel error which said the power cutoff unexpectedly)
I don't know what to do anymore and I really need to find out what the issue is exactly because I don't want to have to send the entire pc back.

I don't see "BIOS update" in your list of attempted fixes.
Personally, I would never choose an "A" motherboard for a gaming PC. Yes, the 7800X3D is supported, but the "A" level motherboards are the lowest level of chipset for AMD. Maybe that is just my personal bias.

 

[Wouter_U]

I don't see "BIOS update" in your list of attempted fixes.
Personally, I would never choose an "A" motherboard for a gaming PC. Yes, the 7800X3D is supported, but the "A" level motherboards are the lowest level of chipset for AMD. Maybe that is just my personal bias.

That's the first time I've heard that. I've heard good things about A series motherboards. No idea how to test if that would be the issue though.
Bios updates are also done. Could not remember everything I've done so far.

 

[Wouter_U]

Right now I'm testing it out with all the sides and the top off So that the GPU power cable (Nvidia 12VHPWR) gets as much room as possible. It get's quite cramped in the case for the power cable, but I've never heard of issues with it in the Fractal Ridge. People even put 4090's in the thing.
So far no crashing, but as the problem is so inconsistent I will keep testing the upcoming days.

 

[kurdtnz]

Have you set the EXPO profile for the RAM in BIOS? Have you tried running it with the EXPO turned off? I was having the same issue when I first built my AM5 pc, it wasnt until I updated the BIOS that EXPO worked. AM5 is REALLY fussy with RAM speeds/timings.

 

[Wouter_U]

Have you set the EXPO profile for the RAM in BIOS? Have you tried running it with the EXPO turned off? I was having the same issue when I first built my AM5 pc, it wasnt until I updated the BIOS that EXPO worked. AM5 is REALLY fussy with RAM speeds/timings.

I will test that out after I've had more time testing out if the problem persists when the GPU power cable is less restricted.

 

[Wouter_U]

This week I've hjad the panels off to give the gpu cable more room. the entire time there was no crash. No I tried it with the panels on again and within 10 minutes there was a crash.
Looks like the Nvidia power cable (12VHPWR cable) is indeed the culprit, but it's hard to know for 100% certain.

Any advice on custom cables? I know you have to be very carefull with power cables from the PSU, but I don't know if it's the same with the special Nvidia connector as the gpu and psu both have it.

 

[kurdtnz]

This week I've hjad the panels off to give the gpu cable more room. the entire time there was no crash. No I tried it with the panels on again and within 10 minutes there was a crash.
Looks like the Nvidia power cable (12VHPWR cable) is indeed the culprit, but it's hard to know for 100% certain.

Any advice on custom cables? I know you have to be very carefull with power cables from the PSU, but I don't know if it's the same with the special Nvidia connector as the gpu and psu both have it.

Is the side panel pressing on the cable? Personally, I would avoid 3rd party 12VHPWR cables, I hope someone can advise you further.

 

[DaleH]

This week I've hjad the panels off to give the gpu cable more room. the entire time there was no crash. No I tried it with the panels on again and within 10 minutes there was a crash.
Looks like the Nvidia power cable (12VHPWR cable) is indeed the culprit, but it's hard to know for 100% certain.

Any advice on custom cables? I know you have to be very carefull with power cables from the PSU, but I don't know if it's the same with the special Nvidia connector as the gpu and psu both have it.

When you took the side panels off, you likely had more cooling also.

 

[Wouter_U]

When you took the side panels off, you likely had more cooling also.

Yes, but cooling is not an issue. Temps stay below 80 on the cpu (which is fine for a 7800x3d) and below 70 on the gpu

 

[kanewolf]

Yes, but cooling is not an issue. Temps stay below 80 on the cpu (which is fine for a 7800x3d) and below 70 on the gpu

Could be a VRM cooling issue. Especially if you have liquid cooling on the CPU.

 

[Wouter_U]

I've received a replacement PCI-E riser card from Fractal Design just to make sure that that is not the problem. So far I've been gaming without problem with the case fully closed. But I've seen it before running well for a week and then suddenly starting to frequently crash again so I can't say it's been solved yet.

 

[Wouter_U]

Could be a VRM cooling issue. Especially if you have liquid cooling on the CPU.

I don't know how to check the VRM temps. I don't have liquid cooling. I use the Noctua NH-L12S since that's the best that can fit in this case.

 

[kanewolf]

I don't know how to check the VRM temps. I don't have liquid cooling. I use the Noctua NH-L12S since that's the best that can fit in this case.

An A620 motherboard, a 7800X3D, a 4080 and a tiny case. Take the side panel off and point a desk fan at the motherboard. See if that takes care of your problems. That will cool everything. If it cures the problem then you have to work on WHAT is overheating.

 

[Roland Of Gilead]

I don't know how to check the VRM temps. I don't have liquid cooling. I use the Noctua NH-L12S since that's the best that can fit in this case.

That's the issue with the A series mobo's. They have weak VRM/power phase set up. Technically they can handle the 7800x3d or higher core count CPU's. But do you really want to limit the CPU potentially out of the box? No, you don't.

Your A620 has an 8+2+1 set up. Not very robust for the best CPU's. As an example the ASRock B650e Steel Legend has a much better set up with 16+2+1 phases. This may be your issue.

Also, you mentioned you updated the bios. Did you clear CMOS afterward? If not, do that before proceeding. Gotta rule that out.

IMO, you need a higher end board for your components.

 

[Wouter_U]

That's the issue with the A series mobo's. They have weak VRM/power phase set up. Technically they can handle the 7800x3d or higher core count CPU's. But do you really want to limit the CPU potentially out of the box? No, you don't.

Your A620 has an 8+2+1 set up. Not very robust for the best CPU's. As an example the ASRock B650e Steel Legend has a much better set up with 16+2+1 phases. This may be your issue.

Also, you mentioned you updated the bios. Did you clear CMOS afterward? If not, do that before proceeding. Gotta rule that out.

IMO, you need a higher end board for your components.

That really sucks if that really is the issue here. If I'd known this I would probably have gone for a bit better motherboard.
How is anyone who is not a complete computer expert supposed to know this beforehand. And it's far too late to return the current motherboard, so that's wasted money then if it is indeed the issue.

Another update on my troubleshooting: The PCIE riser card I received from Fractal Design did not help the issue. I experienced another crash just now.