[SOLVED] Help ! 4090 FE restarts under stress test after 1 min ?

[Sangeeth Sivan]

Hi, i have a corsair HX1200W connected to a 4090 with corsair 12hvpwr pcie cable. I ran aida64 for gpu stress test and just after 1 min , upon reaching 360 plus watts my pc restarts. When checking windows event viewer the under critical it says “Kernel power” . Cpu stress test and all works fine. Could this be a problem with gpu? Or psu? Thanks.

EDIT: Some more details, the psu is in single rail mode, Also this is how the cables are connected, the one in center is the gpu cable corsair 12hvpwr one, top is the cpu cable, and then atx 24pin connector.

System Specs:
Ryzen 7600x
ASUS TUF X670E
Gskill Trident Z5 Neo DDR5
Corsair 1200W platinum

 

[shengton]

Based on the information you've provided, it's possible that the problem is with either the power supply unit (PSU) or the graphics card. The PSU is responsible for providing power to all the components in your system, including the graphics card. If the PSU is not providing enough power to the graphics card, or if it is failing to deliver power consistently, this can cause the system to crash.

 

[Sangeeth Sivan]

Based on the information you've provided, it's possible that the problem is with either the power supply unit (PSU) or the graphics card. The PSU is responsible for providing power to all the components in your system, including the graphics card. If the PSU is not providing enough power to the graphics card, or if it is failing to deliver power consistently, this can cause the system to crash.

Hey thanks for replying, now my PC restarts even when its idle, just after login. And when I checked the Mobo Q-LEDs it showed DRAM light in orange, when it was restarting. Also the RGB on my Memory module is kinda pulsating and not smooth. Could my memory be the problem then ?

 

[shengton]

Hey thanks for replying, now my PC restarts even when its idle, just after login. And when I checked the Mobo Q-LEDs it showed DRAM light in orange, when it was restarting. Also the RGB on my Memory module is kinda pulsating and not smooth. Could my memory be the problem then ?

It's possible that your memory is the cause of your computer restarting. Memory (or RAM) can sometimes fail, which can cause all sorts of problems, including random restarts. If you have more than one memory module, I would try removing one at a time and see if that helps. If the problem persists with only one memory module installed, then it's likely that the module itself is the cause. You may need to replace it.

 

[ak47jar3d]

I would watch the Temps of the card and the power consumption. I also try changing the 12 pin adapter if possible as it's a high probability of being a psu issue....

 

[Sangeeth Sivan]

It's possible that your memory is the cause of your computer restarting. Memory (or RAM) can sometimes fail, which can cause all sorts of problems, including random restarts. If you have more than one memory module, I would try removing one at a time and see if that helps. If the problem persists with only one memory module installed, then it's likely that the module itself is the cause. You may need to replace it.

Hey, So yesterday I switched my 4090 with a 3070ti from my prev build, the problem still persists. Also removed both RAM and tested by having only one, the problem still persists. But when I had only one RAM installed and ran UNIGINE Superposition benchmark for the first time, it was able to complete. The other times around 11th test or something the system restarts. When having both RAMs installed the system restarts around the same time aswell. I ran memtest86 4 passes, no errors. Is 4 passes with no errors enough to leave out RAM out of the possible faulty hardware ? This just leaves me with PSU, Mobo. Any suggestions on what I could do next to narrow it down ?

 

[DRagor]

So yesterday I switched my 4090 with a 3070ti from my prev build, the problem still persists

3070ti was connected with regular 8 pin cables? That would rule out both bad 12HPWR cable and GPU.

I ran memtest86 4 passes, no errors.

If there is a problem it should usually appear in first pass. No specific reason to suspect RAM, especially if the system was passing CPU benchmarks without problems.

How old is your HX?

 

[Unolocogringo]

What does your case cooling look like?
Ram could be overheating. Seen it before.
Is your ram a matched pair from one package or 2 separate sticks from 2 different packages?
Are they in the correct slots on the motherboard? 2nd and 4th slots from the cpu?

 

[Sangeeth Sivan]

3070ti was connected with regular 8 pin cables? That would rule out both bad 12HPWR cable and GPU.

If there is a problem it should usually appear in first pass. No specific reason to suspect RAM, especially if the system was passing CPU benchmarks without problems.

How old is your HX?

Hi yes they are connected with 8 pin cables that comes with the PSU, and I just bought my PSU on Dec 5th. Could it be a MOBO problem then ? I also tried installing only one RAM module , switched slots with RAM module and the problem still persists. Sometimes it just happens right away after login. And it happens all the time when I play RDR 2 for 10 mins, However I was able to play Cyberpunk for more than an hour and had no crash. But with RDR2 it always crashes in the first mission

View: https://www.youtube.com/watch?v=GU5vd5Z28UQ
at 13:42 right when that cutscene was about to happen. Always crashes around this time. Happened on both 4090 and 3070ti.

 

[Sangeeth Sivan]

What does your case cooling look like?
Ram could be overheating. Seen it before.
Is your ram a matched pair from one package or 2 separate sticks from 2 different packages?
Are they in the correct slots on the motherboard? 2nd and 4th slots from the cpu?

Case cooling is fine, I monitored the temps when I was playing games and also when running stress tests, temps are always within limit. And yes they are installed on the correct slots.

 

[DRagor]

Could it be a MOBO problem then ?

Impossible to say for sure, but I would say there is a good chance. Next in line of suspects is PSU (even if its new - I assume here the problem appears consistently since you built the PC).

 

[Sangeeth Sivan]

Impossible to say for sure, but I would say there is a good chance. Next in line of suspects is PSU (even if its new - I assume here the problem appears consistently since you built the PC).

Yeah I just built it like few days ago, been experiencing it from day one. I will try and get a replacement for the PSU and check. I already had my MOBO replaced because when I unboxed it , I found some kinda leaks behind VRMs. The replaced one didn't had any such things. So even I am thinking it could be either PSU or Mobo. Thanks.

 

[shengton]

Hey thanks for replying, now my PC restarts even when its idle, just after login. And when I checked the Mobo Q-LEDs it showed DRAM light in orange, when it was restarting. Also the RGB on my Memory module is kinda pulsating and not smooth. Could my memory be the problem then ?

It's possible that your memory (RAM) could be the cause of the issue you're experiencing. A pulsating RGB light on the memory module and an orange DRAM light on the motherboard could indicate a problem with the RAM.
There are a few steps you can try to troubleshoot the issue:

1. Check to make sure that the RAM is properly seated in the memory slots on the motherboard. Sometimes, if the RAM is not seated correctly, it can cause issues like the ones you're describing.
2. Try running a memory diagnostic tool to check for any issues with the RAM. You can use a tool like Windows Memory Diagnostic (on Windows) or MemTest86 (on Linux) to test the RAM.
3. Try removing one or more of the RAM sticks and see if the issue persists. This can help you narrow down whether the issue is with a specific stick of RAM or with the RAM as a whole.
4. If you have multiple sticks of RAM, try swapping them around and see if the issue persists. This can help you determine if a specific stick of RAM is causing the issue.

If none of these steps helps, it's possible that the RAM itself is faulty and may need to be replaced. In any case, it's always a good idea to have a backup plan in case something goes wrong, such as having a bootable USB drive or a bootable CD/DVD on hand in case you need to reinstall the operating system

 

[Sangeeth Sivan]

It's possible that your memory (RAM) could be the cause of the issue you're experiencing. A pulsating RGB light on the memory module and an orange DRAM light on the motherboard could indicate a problem with the RAM.
There are a few steps you can try to troubleshoot the issue:

1. Check to make sure that the RAM is properly seated in the memory slots on the motherboard. Sometimes, if the RAM is not seated correctly, it can cause issues like the ones you're describing.
2. Try running a memory diagnostic tool to check for any issues with the RAM. You can use a tool like Windows Memory Diagnostic (on Windows) or MemTest86 (on Linux) to test the RAM.
3. Try removing one or more of the RAM sticks and see if the issue persists. This can help you narrow down whether the issue is with a specific stick of RAM or with the RAM as a whole.
4. If you have multiple sticks of RAM, try swapping them around and see if the issue persists. This can help you determine if a specific stick of RAM is causing the issue.

If none of these steps helps, it's possible that the RAM itself is faulty and may need to be replaced. In any case, it's always a good idea to have a backup plan in case something goes wrong, such as having a bootable USB drive or a bootable CD/DVD on hand in case you need to reinstall the operating system

Hey, tried all that, ran memtest etc. The problem still happened. However, I switched to AMD Ryzen Eco mode,and its been stable for the longest time. I played RDR 2 for more than an hour at 4k TV, 80 + plus FPS, with GPU drawing as much as 370+ watts. But just like few hours back I was playing overwatch training range, and 5 mins in, the problem happened again. Launched OW again , played for 5 mins boom again system restarted. But I then optimized overwatch with nvidia settings, its been 20 minutes now and stable. When I played overwatch I set the fps to be 300 for both desired and maximum, and I was getting 300 fps for sure, looks like its related to that. With this finding can we determine what could be the issue ? Could it be like my cpu is faulty and it can't handle heavy load ? I was monitoring the temps, the temps were fine just under 75. I know its not my GPU because I switched with 3070ti it was still happening. hard to find the root cause of failure ;/

 

[Sangeeth Sivan]

UPDATE: It was my processor. RMAed and got a replacement, everything works great now ! I am now even able to have my 64gb 4 stick ram run at 6000Mhz on Expo 2.

 

[Unolocogringo]

Thanks for the update.
Processors ,unless overvolted are usually the last thing to die.
So it does happen.