SOLVED Odd OCing issue (OC vid card, get HDD issues)

Recently got a GTX580 and have tried my hand at OCing a bit. Running stress tests stock gives no issues, after some mild OCing my computer starts acting up. Stock clocks are 832, if I bring it to around 890/900, it will run fine for a bit. No artifacts will appear. However, at random programs may start hanging up, or explorer will crash, etc. I end up getting a BSOD with code 0x000000F4, which I've gathered is an IDE issue. Incidentally, sometimes I'll restart and my boot options in UEFI will have changed to set my 250GB HDD as boot drive instead of my RAID array, other times I'll boot and my RAID array will say that my second OCZ Vertex has failed (which has given me major issues in the past), and most of the time after a reboot Windows will schedule an automatic chkdsk.

Since I'm not seeing artifacts, I'm wondering if something else is the issue with this?

SolidxSnake said:

Recently got a GTX580 and have tried my hand at OCing a bit. Running stress tests stock gives no issues, after some mild OCing my computer starts acting up. Stock clocks are 832, if I bring it to around 890/900, it will run fine for a bit. No artifacts will appear. However, at random programs may start hanging up, or explorer will crash, etc. I end up getting a BSOD with code 0x000000F4, which I've gathered is an IDE issue. Incidentally, sometimes I'll restart and my boot options in UEFI will have changed to set my 250GB HDD as boot drive instead of my RAID array, other times I'll boot and my RAID array will say that my second OCZ Vertex has failed (which has given me major issues in the past), and most of the time after a reboot Windows will schedule an automatic chkdsk.

Since I'm not seeing artifacts, I'm wondering if something else is the issue with this?

Click to expand...

Maybe overloading the chipset? When I used RAID 0 in the past my OC wasn't as stable on the high clocks it used to be. Going back to a RAID free system got my oc back to what it was. Now the video card oc causing this? Very strange, but that may be the issue. Try running everything at stock clocks on the cpu and everything else and see if it happens.

How many amps does that PSU run on 12V?
I don't think the PCIE lanes from the CPU will affect the PCH.
Is your BCLK altered? Is it @ 100?

Are you on AHCI mode? Have you tried increasing VCCIO/SA slightly ? I'd do that and also increase PCH voltage slightly, just to see if it increases stab. If not return to auto.

Okay, so it's not an OCing issue. If I run FAH SMP and GPU with everything at default (ram at XMP, rated DDR3-1600 @ 9-9-9-24 1.5v, CPU at all auto, GPU at all auto) I still get the same error.

Vertex drives are in RAID0, the HDDs are regular. BCLK was at 100 when OC'd, and remains there running stock.

Computer works fine at all times besides GPU stress tests and FAH. Haven't tried running IBT or Prime95 since, will do that and see the results.

PSU I'm using: http://www.hardwaresecrets.com/article/745

+12V1 = 22A
+12V2 = 22A
+12V3 = 36A
+12V4 = 36A

Update: running P95 or IBT has no issue. Will try Memtest next, but I think only GPU loads (OC'd or stock) are causing this issue. Also going to try running just FAH CPU without FAH GPU and see what happens. I'm willing to bet it will work fine.

Well then. From what I've been able to gather the card consumes ~20A at stock so if it is plugged into rail 1 or 2 only then I could see there being a problem, but I doubt that is the case.

It does seem to be power related though. What card did you have before?

GTX260, and before that a 9800GTX+. Never folded on either of those, though (well, on the 9800GTX+ I did, but that was with a C2D E8600 and an EP45-UD3P).

More update: FAH GPU on its own will lock up my rig. It seemed to be working fine, but then I tried to do something and explorer started acting up. Couldn't shut down, used the reset button, and the RAID ROM booted to tell me the second SSD had an "Error Occurred" msg. Booted into chkdsk as before.

Gonna try reinstalling GPU drivers (something I didn't do when I switched to the GTX580 since I was using the same driver package), but after that I'll be stumped. I was thinking power as well, but I feel like my PSU is running fine. Perhaps I should flip the fan on the PSU back (i swapped its orientation so it blows out instead of in).

SolidxSnake said:

GTX260, and before that a 9800GTX+. Never folded on either of those, though (well, on the 9800GTX+ I did, but that was with a C2D E8600 and an EP45-UD3P).

More update: FAH GPU on its own will lock up my rig. It seemed to be working fine, but then I tried to do something and explorer started acting up. Couldn't shut down, used the reset button, and the RAID ROM booted to tell me the second SSD had an "Error Occurred" msg. Booted into chkdsk as before.

Gonna try reinstalling GPU drivers (something I didn't do when I switched to the GTX580 since I was using the same driver package), but after that I'll be stumped. I was thinking power as well, but I feel like my PSU is running fine. Perhaps I should flip the fan on the PSU back (i swapped its orientation so it blows out instead of in).

Click to expand...

Does you psu tell you what pcie connectors are on what rails? Since you didn't start with a fresh driver install that may have something to do with it, but you'll find out shortly I guess.

PanteraGSTK said:

Does you psu tell you what pcie connectors are on what rails? Since you didn't start with a fresh driver install that may have something to do with it, but you'll find out shortly I guess.

Click to expand...

The 12v3 and 12v4 rails are the PCIE ones (they are labelled on the actual plug as +12v3 and +12v4)

Installed the new beta drivers off nVidia's site and am running FAH GPU no issues. I changed the configuration slightly to allow the GPU to be 100% loaded (i dropped it to around 95% previously since it slowed down web browsing, and currently is slowing down the response to my typing in this reply field). Haven't had issues yet, been going for a couple of minutes now.

As per usual, once the GPU starts running FAH there is a faint high-pitched squeal coming from my computer. This has always been the case as long as I've been running FAH GPU (with my old 9800GTX+ and the C2D rig). I haven't checked but my intuition tells me it's coming from the PSU. Hasn't ever posed an issue before, just been slightly annoying at times.


Edit: spoke too soon. After a couple more minutes the OS started acting up again.

edit2: Just tried running OCCT-GPU and the second I started it the OS acted up again. So I've narrowed it to GPU definitely, and it's having issues with my HDDs. I would suspect power, but that seems far fetched, as I'm sure it's plenty powerful to handle my rig. Hopefully not a mobo issue?

Any other ideas?

Tried upping PCH voltage. I had it at 1.301v, ran OCCT GPU. Seemed fine (ran for several minutes). The second I clicked stop I got a BSOD =_=

Set BIOS entirely to default, ran the test for a minute or two and had no issues.

I am SO confused =_=

Update: I pulled my PSU apart this morning and reversed the fan direction. I also swapped the ram to slots 1/3 instead of 2/4. In addition I used the first PCIe connectors on each branch to connect to my 580 (instead of the ones on the end).

Been running the SMP and GPU FAH clients with no issues all day. Weird fix, huh?