Question new rtx 3080, crashes when gaming only

[tshefi]

CPU: Intel core i5, 10400 @ 2.9 ghz
mobo: Asus TUF gaming B460 Plus
Ram: 32 gb 2xCorsair Vengeance LPX 16GB DDR4 2400Mhz CL16
SSD: Sandisk SSD Plus 480 gb
HHD: Seagate barracuda 3TB, SATA
GPU: ASUS TUF 3080 10gb OC
PSU : Corsair CX650M
Monitor: gigabyte M32QC (2k, 31.5 inch curved) on DP, + old 24' philips monitor on HDMI
OS : Win 10


after years of using asus gtx 1060, i decided it was time for an upgrade, recently purchsed the Asus TUF 3080 RTX 10 GB OC,

ever since installing it, the PC freezes after several minutes of playing time, with the addition of static noise coming from the speakers/headphones
doesnt respond to anything, until it crashes and resets about 10 seconds later

timing is not consistent but it does seem to happen faster on graphic intensive games

does not happen with casual browsing/office work

temps on the card are anywhere from 50 C to 60 C when crashing


i tried DDU + reinstaling new gpu drivers (from either asus or nvidia)
mem test went fine
updated mobo's bios to latest


This never happened on the old 1060 so i can only conclude it has something to do with either a compatibility issue between the new card and my components, or perhaps a faulty card.
seeing as this pc was built up with "patches" along the years, I fear it may have more to do with the former

Any suggestions as to how to continue troubleshooting this ?

Thanks!

 

[Lutfij]

Welcome to the forums, newcomer!

How old is the PSU in your build? FYI, you're underpowered for that GPU, you're in need of a reliably built, branded PSU that can deliver at least 750W to the entire system. The power draw on the GTX1060 and the RTX 3080 are like day and night.

 

[tshefi]

Welcome to the forums, newcomer!

How old is the PSU in your build? FYI, you're underpowered for that GPU, you're in need of a reliably built, branded PSU that can deliver at least 750W to the entire system. The power draw on the GTX1060 and the RTX 3080 are like day and night.

i figured the same thing, so together with the GPU, i bought a new corsair tx850M

the issue happened initially with the new PSU, i went back to the old one to see if perhaps the new PSU was the source, and it actually was able to last longer on games before crashing, so i stayed with that, the new one is right at hand if you think it might somehow help address the issue

the 650 is about two years old


and thank you for the welcome, i was a silent stalker for quite some time now 😛

 

[azed3000]

Hey mate, im having the exact same issue. I'm certainly sure its PSU related. Need 750watt minimum with your build. i9 10900k needs 850watt minimum. I've pre-emptively purchased a power supply to see if it will resolve my issue. I have the exact symptoms as you, graphic intensive scenes the gpu carcks it. Also just make sure you are running 2 SEPARATE pcie cables from the PSU to the gpu. As the GPU is 350watts while the cables can only push up to 150 watts. 75 watts comes from the PCIE slot (150+150+75 = 375 watt)

Oh if its a triple pcie powered card, then first 2 cables need to be seperate, the third can be daisy chained from the second.

 

[tshefi]

Hey mate, im having the exact same issue. I'm certainly sure its PSU related. Need 750watt minimum with your build. i9 10900k needs 850watt minimum. I've pre-emptively purchased a power supply to see if it will resolve my issue. I have the exact symptoms as you, graphic intensive scenes the gpu carcks it. Also just make sure you are running 2 SEPARATE pcie cables from the PSU to the gpu. As the GPU is 350watts while the cables can only push up to 150 watts. 75 watts comes from the PCIE slot (150+150+75 = 375 watt)

Oh if its a triple pcie powered card, then first 2 cables need to be seperate, the third can be daisy chained from the second.

i also preemptively bought a 850 wat PSU,

on both the old 650 and the new 850, i ran separate cables to the card and did not daisychain

freezes on both accounts, just somehow it lasts longer on the 650 before crashing

 

[azed3000]

i also preemptively bought a 850 wat PSU,

on both the old 650 and the new 850, i ran separate cables to the card and did not daisychain

freezes on both accounts, just somehow it lasts longer on the 650 before crashing

Do you have assignable rail settings on the 850watt? Might need to switch it from Multi-rail to single rail? I know some of the corsair PSUs you can change between rails. Multi-rail would cause issues with 3080s and above

 

[tshefi]

Do you have assignable rail settings on the 850watt? Might need to switch it from Multi-rail to single rail? I know some of the corsair PSUs you can change between rails. Multi-rail would cause issues with 3080s and above

this is something i did not consider, thanks!!


for the life of me, i couldn't find any mention of single\multi rail system on the official user manual
https://www.corsair.com/corsairmedia/sys_master/productcontent/TXM_Manual.pdf


I did however find a store mentioning that it is a non single/multi adjustable PSU

How can i check the rail settings with out physically opening the psu and voiding the warranty ?

 

[azed3000]

How can i check the rail settings with out physically opening the psu and voiding the warranty ?

Just did a quick search, I think its a single rail cause i only see 12v1 in the table chart:

Review: Corsair TX-850M (850W) - PSU - HEXUS.net

Usually multi rails would be like 12v1 12v2 12v3 @ 25amp. But not 100%

 

[tshefi]

Just did a quick search, I think its a single rail cause i only see 12v1 in the table chart:

Review: Corsair TX-850M (850W) - PSU - HEXUS.net

Usually multi rails would be like 12v1 12v2 12v3 @ 25amp. But not 100%

which means that this is not the case 🙁


im gonna plug the 850 PSU to the GPU alone, just by itself, while have the 650 run the other components,

if this was the case, this should solve the issue, shouldn't it ?

 

[geofelt]

3000 class graphics cards can have some abnormal power spikes and can send signals that some power supplies can not handle.
Replacing the modular psu cables may be a fix.

Contact Corsair customer support, they can likely help.

 

[tshefi]

Update

i tried again this morning, got 1 game crash back to desktop and one bsod rather than the usual freeze


I put the 850 watt psu in, cleaned out the case with a blower, reconnected ram cards, replugged the card
underclocked the card at 90% power and 60 mghz underclock

tested it on a game and it worked fine for about an hour
went out for an hour or two while the pc was on sleep mode, when i came back i found the PC waiting for me to login as if it has crashed,
i found a critical error on the system event viewer

"The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."
sourced from the kernel-power

so it appears the PC crashes when on idle as well


ideas ?

 

[faalin]

Kernel power 41 errors dont give a reason, its just a generic computer was shut down unexpectedly. you need to look in the error section in the logs to see why it crashed.


Make sure all drivers and bios are up to date and see if that fixes it. I had a server at work that would crash, found out it had an outdated video card driver that was causing the issue. Updated it and crashes went away.

 

[tshefi]

Kernel power 41 errors dont give a reason, its just a generic computer was shut down unexpectedly. you need to look in the error section in the logs to see why it crashed.


Make sure all drivers and bios are up to date and see if that fixes it. I had a server at work that would crash, found out it had an outdated video card driver that was causing the issue. Updated it and crashes went away.

Flash updated bios drivers and ddu+installed newest gpu drivers two days ago and the problem persists through


Will try again with either nvidia or asus drivers

Thanks

 

[tshefi]

Update:
Had a hunch removed one of the ram cards and the problem is resolved

Currently only on ome 16gb corsair
No more crashes, 165 fps on most games

Thanks for the help!

 

[geofelt]

Update:
Had a hunch removed one of the ram cards and the problem is resolved

Currently only on ome 16gb corsair
No more crashes, 165 fps on most games

Thanks for the help!

Your Corsair ram will have a lifetime warranty.
To confirm a valid RMA,
Run memtest86 or memtest86+
They boot from a usb stick and do not use windows.
You can download them here:

MemTest86 - Official Site of the x86 Memory Testing Tool

MemTest86 is the original self booting memory testing software for x86 and ARM computers. Supporting both BIOS and UEFI, with options to boot from USB.

www.memtest86.com

Memtest86+ | The Open-Source Memory Testing Tool

Memtest86+ is an advanced, free, open-source, stand-alone memory tester for 32- and 64-bit computers (UEFI & BIOS supported)

www.memtest.org

If you can run a full pass with NO errors, your ram should be ok.

Running several more passes will sometimes uncover an issue, but it takes more time.
Probably not worth it .

 

[Third-Eye]

Update:
Had a hunch removed one of the ram cards and the problem is resolved

Currently only on ome 16gb corsair
No more crashes, 165 fps on most games

Thanks for the help!

You should run a memory tester on the module you removed. Place it in your working slot and then run something like memtest86 Free edition from a usb drive. If it comes out clean with 0 errors, it may mean the CPU pins associated with that ram slot are shifted out of position and the CPU may need to be reseated in the socket after checking the CPU socket pins for damage.

https://www.memtest86.com/download.htm