Jump to content

PC shuts down immediately on Prime95 with graphics card.

ybriK

I have a spare PC that if I run Prime95 or any other CPU stress testing software it will immediately shut down when a graphics card is installed which is weird because running a GPU stress software like Valley, FurMark or 3DMark (Fire Strike) will run fine no problems whatsoever. I have swapped out the CPU from an i5-3470 to an i5-2500 thinking it was a faulty CPU but that wasn't the case. Even with a fresh install of Windows 7 it still happens. When I use the iGPU instead (graphics card removed) I can run Prime95/AIDA64 any mode for a long time, longest was 7 hours on Prime95 blend and then I stopped it.

 

Temperatures aren't the issue here (I have been monitoring temperatures using AIDA64 and they are well within the thermal limits) as I have mentioned before the PC only crashes when a graphics card (power hungry in particular) is installed while the CPU being stressed. I'm suspecting it's something to do with the motherboard and PCIe its' power delivery. It wouldn't make sense if it was the PSU as I can run a GPU stress test as mentioned before on these power hungry just GPUs fine.

 

pcpartpicker.com/list/wtgrPs

I have the latest motherboard BIOS

 

List of GPUs I have tested with that have immediately shutdown. I have the latest drivers.

HD 6990

HD 5970

GTX 590

GTX 480

R9 270X

 

List of GPUs I have tested with that DOESN'T immediately shutdown. I have the latest drivers.

HD 6670

iGPU - Intel HD 2000, Intel HD 2500

Link to comment
Share on other sites

Link to post
Share on other sites

Have you tried not using Prime95? From what I hear, P95 is what you use when your hardware has been bad and you want to punish it. AKA it puts such a high load on the hardware that overheating can occur even when you have adequate cooling. Using pretty much any other benchmarking software will be better.

Link to comment
Share on other sites

Link to post
Share on other sites

P95 is excessively hard on computers, I personally have had it kill motherboards.

 

The problem that you are describing makes it sound like your power supply is too weak and you are triggering an over current shutdown protection. What are your full system specs? At this point it is only speculation.

Intel Xeon 1650 V0 (4.4GHz @1.4V), ASRock X79 Extreme6, 32GB of HyperX 1866, Sapphire Nitro+ 5700XT, Silverstone Redline (black) RL05BB-W, Crucial MX500 500GB SSD, TeamGroup GX2 512GB SSD, WD AV-25 1TB 2.5" HDD with generic Chinese 120GB SSD as cache, x2 Seagate 2TB SSHD(RAID 0) with generic Chinese 240GB SSD as cache, SeaSonic Focus Plus Gold 850, x2 Acer H236HL, Acer V277U be quiet! Dark Rock Pro 4, Logitech K120, Tecknet "Gaming" mouse, Creative Inspire T2900, HyperX Cloud Flight Wireless headset, Windows 10 Pro 64 bit
Link to comment
Share on other sites

Link to post
Share on other sites

Only thing Prime95 does is stress the CPU, if it's overclocked it might BSOD or freeze, never simply shutdown. 1st thing that comes to mind is the PSU getting pushed past it's limits or faulty. Does this happen only with Prime95 ? Have you tried stressing with other programs like RealBench or OCCT ?

 

Hardware 1st then Software.

 

Ryzen 3800x 4.4ghz 1.294v all core Noctua NH-D15 x570 Aorus Elite 16gb G.Skill Flare X 3600mhz 15-15-15-30 1T

XFX 5700 XT Thicc III Ultra Seasonic Focus+ 650w Platinum

Link to comment
Share on other sites

Link to post
Share on other sites

After re-reading the post I see that the OP has tried two different CPUs. This further supports the theory that it is a PSU issue. The problem can also point to a faulty motherboard. If the GPU doesn't trigger the over current then it is most likely a motherboard problem.

 

Neither one of these CPUs support overclocking right?

 

In any case, I would suggest never using P95 again and I would suggest not using Aida64 anymore either as further stress testing will just make the problem worse for the motherboard.

Intel Xeon 1650 V0 (4.4GHz @1.4V), ASRock X79 Extreme6, 32GB of HyperX 1866, Sapphire Nitro+ 5700XT, Silverstone Redline (black) RL05BB-W, Crucial MX500 500GB SSD, TeamGroup GX2 512GB SSD, WD AV-25 1TB 2.5" HDD with generic Chinese 120GB SSD as cache, x2 Seagate 2TB SSHD(RAID 0) with generic Chinese 240GB SSD as cache, SeaSonic Focus Plus Gold 850, x2 Acer H236HL, Acer V277U be quiet! Dark Rock Pro 4, Logitech K120, Tecknet "Gaming" mouse, Creative Inspire T2900, HyperX Cloud Flight Wireless headset, Windows 10 Pro 64 bit
Link to comment
Share on other sites

Link to post
Share on other sites

I still don't buy it's the PSU as this was the PSU I used with an overclock 2700K at 4.5GHz and that never crashed which used to be on my main PC that's using a different PSU now. If it really is then it's a shame it couldn't even live for at least 5 years as this thing is only 2 years old. But if I can get my hands on a cheap 1155 motherboard then I might get one to see if the situation will still occur.

 

No I've used 3DMark Physics, AIDA64 SST, Cinebench all will immediately shutdown (less than 10 seconds as soon as I start any of these) the system only when a graphics card is connected to the motherboard.

 

Currently have my HD 5970 (290W+ TDP) running for about 7 hours now on FurMark no issues but if I ran a CPU stress tool right now I guarantee this would immediately crash.

Link to comment
Share on other sites

Link to post
Share on other sites

Then it would be the motherboard causing the issue methinks.

Ryzen 3800x 4.4ghz 1.294v all core Noctua NH-D15 x570 Aorus Elite 16gb G.Skill Flare X 3600mhz 15-15-15-30 1T

XFX 5700 XT Thicc III Ultra Seasonic Focus+ 650w Platinum

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×