Testing Hardware Against the Video

[DasFox]

When I run F.E.A.R it constantly locks up. I'm running SLI with Evga 7900GT KO Superclock cards.

If you want to see my complete specs, it's my GameRg in my SIG.

Prime has been running right now on the box 7 hours with no errors, and when I ran memtest about this long before in the past I had no errors either.

I just installed a new PSU, a Corsair 520, and a new CPU, Opteron 150 I'm not sure if either one of these could be giving me the problems, or these new video cards.

I've tried 3 different video drivers. In the bios the PSU volts look good, and the Corsair 520 PSUs, everyone has been raving over them, and the box seems to run just fine, and I only see a problem when gaming, so I'm not really thinking the PSU, or Opteron is the problem, but how can I be sure?

Prime as I mentioned has been running 7 hours, and I'm at work right now, and it's still running, so let it run 24 hours to be sure? Then after Prime should I run memtest, or would prime be enough for the memory too?

I'm running Prime on the default "Torture Test"

I guess since I'm SLI I should run one card in both slots then try them SLI and see what happens.

But if I have any hardware problems, then prime or memtest will error. Also how can I make sure there is no problem with the motherboard, would Prime error if there where issues with it?

I spoke to Evga on these cards and they tell me running idle at 60c isn't a problem or gaming at 80c, one guy even told me they can idle at 75-80 and game at 90, I thought this seemed a bit high, but even when I stuck a room fan next to the case to blow on the cards and really cool them down, the game still locked up, so I didn't consider cooling to be an issue anymore causing this, because there was more then enough air blowing across the cards then.

ALOHA

 

[pkme2]

Would our higher humidity have a contributing factor to your problem? Just asking.

Have you got enough room to add card fans? or maybe something other than the side fan you're using now?

I know that ventilation was very important in my old Win 2000 workstation which used to overheat at times. Getting a better ventilated case solved that problem.

 

[DasFox]

Originally posted by: pkme2
Would our higher humidity have a contributing factor to your problem? Just asking.

Have you got enough room to add card fans? or maybe something other than the side fan you're using now?

I know that ventilation was very important in my old Win 2000 workstation which used to overheat at times. Getting a better ventilated case solved that problem.

This isn't Mississippi, or some bayou here in Maui, LOL, we hardly have any humidity where I live. It's very mild here.


Cooling isn't the issue right now. I have 4 case fans, but that wasn't really helping, and at this point in time, all I can do is put a fan blowing into the case to rid heat as the issue, so I have a room fan blowing in the case. At this point in time I wanted to rid out the card temps and with this big room fan blowing in I have, and Evga says my cards are within normal specs anyways, but none the least I've cooled them down further with this fan.

IF these cards aren't bad, and I stress IF, then I'll buy some Zalman fans for them, but since I have this BIG room fan laying on it's side, cooling isn't the issue anymore to be causing lockups, so now I'm trying to figure what then?

ALOHA

 

[Atheus]

FEAR used to crash a lot for me too. I passed hours of prime, superpi, loops of 3Dmark, but it still turned out to be the overclock. Ended up reverting back to stock and re-overclocking using FEAR itself as the stability test. Looks like this is just one of those games that is extremely sensitive to stability.

If I were you I'd downclock everything a few MHz, up the volts, and drop the temps as much as possible. Just to make sure.

 

[DasFox]

I was just told over at Evga's forum by a Mod, that Antialiasing and Soft Shadows on at the same time is a known FEAR bug.

1.08 supports widescreen now, and I'm running a NEC 20WMGX2 at 1680 x 1050, so when it's isn't locking up, LOL, it's looking great. MAN I really hope this is that bug, I sure hope there is nothing wrong with my cards.

So I'll turn off the soft shadows and see how this work.

ALOHA

 

[Maxspeed996]

I agree , it sounds like you've done your homework. I get what EVGA is telling you normal temps on these cards are but you may do some inspecting to make sure the heatsink isn't out of contact on one of the cards , you may have one card spiking in temperature and I don't know how that would affect your SLI setup. But I could imagine a cascading effect from an error generated on one card to the next could be the culprit.
The only other thing I can point a finger at is your Power Supply voltage dropping when your cards are in SLI and getting low voltage surges trying to keep up with the needs of the cards and the rest of the system. Out of curiosity , how many other drives , and fans are you running? Any lights ?

 

[DasFox]

Well at this point in time the simplest approach is looking at this supposed FEAR bug, and I'll turn off the soft shadows and see what happens, then take it from there.

Keep ya all posted.

ALOHA