PC Restarting, GPU or PSU?

benjo

benjo

benjo

Soldato
Joined
4 Sep 2005
Posts
12,830
Location
London
My machine is constantly restarting under load. Originally I thought it was temps but they're OK. CPU sits at around 60c and the Fury is at 65c.

One thing I've found is that my PC instantly restarts when I run Furmark.
If I run any CPU tests with AIDA64 it's fine, the OC is stable.

Specs:
Ryzen 1600 @ 3.6 with a Noctua dh15
R9 Fury Nitro
EVGA 850 P2

I think it's either the Fury or the PSU, but if anyone has any other guesses I'm all ears :)
 
It's an EVGA Supernova 850w P2, which is why I'm surprised if it is the PSU. I'll run 3dmark and see what happens!
 
What is your motherboard and are you using multiple programs at the same time that access the smbus (hardware monitor programs do this)?
 
BoomAM said:
What is your motherboard and are you using multiple programs at the same time that access the smbus (hardware monitor programs do this)?
Click to expand...

Using a Taichi x370. I've got HWiNFO64 running, and occasionally have the AMD Relive overlay on.

tamzzy said:
unstable gpu overclock?
Click to expand...

GPU is running at stock, the only thing I've changed is setting the target temperature to 65c
 
I've just done a full 3dmark run with no crashes.

If I run Furmark at 1080p, Fullscreen and 8x MSAA it runs fine. Although it's at 25fps~

If I run Furmark at 1080p, Fullscreen and 0x MSAA it crashes the PC instantly.

Can it be related to the framerate the card is pushing?
 
benjo said:
Using a Taichi x370. I've got HWiNFO64 running, and occasionally have the AMD Relive overlay on.
Click to expand...
ok, for a test, disable/dont load/stop the services on boot, all but one of those apps and retest.
I've recently had an issue on my build that im 90% sure is due to a bug in my Asus UEFI caused by too many apps using the smbus at the same time.
Could be related.
 
freddie64 said:
Could this be a unstable RAM problem?

What Ram and motherboard have you got? Maybe a compatibility issue, or needs a bios update.
Click to expand...

I'm using some GSkill Ripjaw V 3600, and a Taichi x370. My ram is running at 3000 at the moment, and it survives memtest and Aida64.
 
benjo said:
I'm using some GSkill Ripjaw V 3600, and a Taichi x370. My ram is running at 3000 at the moment, and it survives memtest and Aida64.
Click to expand...
Are you configuring the memory manually, via DOCP or (if the UEFI/BIOS) has the option, XMP?
 
BoomAM said:
Are you configuring the memory manually, via DOCP or (if the UEFI/BIOS) has the option, XMP?
Click to expand...

It never liked the XMP profiles so I've always done it it manually, I'm actually running loose timings at 3000~ mhz so didn't think it'd be ram related.
 
pandem0nium said:
Best to put everything to stock inc the gpu and memory when testing.
Click to expand...

GPU is at stock, and Memory is running a lower than stock.

The only thing that's overclocked is the CPU and that's only at 3.6 (Usually stable at 3.8). I'll do some more testing tonight!
 
benjo said:
GPU is at stock, and Memory is running a lower than stock.

The only thing that's overclocked is the CPU and that's only at 3.6 (Usually stable at 3.8). I'll do some more testing tonight!
Click to expand...

Memory is overclocked, even if it's lower than XMP advertised speeds.

You might have complete confidence in the stability but you want to keep things as simple as possible when troubleshooting and rule out as much as you can.
 
I spoke to soon, still rebooting during Furmark/Gaming.

I've underclocked (below stock) the GPU by 200~ mhz and it's been ok so far.
 
You must log in or register to reply here.
Back
Top Bottom