Crysis makes PC shut down after a while. (Sapphire 3870x2)

[redhawtcar]

OS: Win XP SP.
PC:

Videocard: Sapphire HD 3870 X2
Processor: Q6600 @ 2.40GHz
Mainboard: GIGABYTE DQ6 X38 (Latest BIOS on Gigabyte website)
Memory: GeiL BlackDragon DDR2 2GB @ 800MHz
Soundcard: SB Live 5.1
PSU: Thermaltake Toughpower 750W Modular

Everything at default clocks.


When playing Crysis on very high (DX9) settings my PC shuts down at some point between 5 minutes and 10 (random).

Things checked:

-I have the 6pin + the 8pin connected. (btw, why the Sapphire has only one 6pin while the rest have two 6pin and one 8pin connectors?)

-I have the molex cable that supplies extra power to the PCI slots on the motherboard connected.

-I tried CAT 8.2, 8.3, 8.3+hotfix (using Driver Cleaner when deleting).

-Temp goes at 81ºC max when auto, but I tried with the fan blowing at 50% constantly with RivaTuner and it still shuts down.

The rest of games I've played plays just fine (for example I played Oblivion for an hour straight and nothing weird. Everything maxed out with the QTP3).

I remember I played Crysis on very high (DX9) settings with my 8800gt for hours and it never made my PC shut down just like if someone pulled off the power cable out.

I refuse to belive a Thermaltake 750W is not suppling enought power to a system that has nothing overclocked.

Please, help ASAP.

 

[EXT64]

Well, it certainly sounds like power. But I agree that 750 should be plenty. Weird

 

[Steven Bancroft]

Multiple things could be wrong here. 750W is good and Thermaltake is a good brand, but no one has 100% quality contontrol. The Entire system shutting down is usually a thermal event to prevent damage or the PSU giving out. Swap video cards if possible and see if it still happens. A cheaper power supply tester can tell you if you're getting good voltage. 81C is pretty darn hot! I would start with the video card.

 

[dagger]

What about cpu temp and vcore? Crysis is cpu intensive too. Do you have an aftermarket cooler? If not, and it runs hot, get one.

 

[redhawtcar]

Multiple things could be wrong here. 750W is good and Thermaltake is a good brand, but no one has 100% quality contontrol. The Entire system shutting down is usually a thermal event to prevent damage or the PSU giving out. Swap video cards if possible and see if it still happens. A cheaper power supply tester can tell you if you're getting good voltage. 81C is pretty darn hot! I would start with the video card.

As I said:

I remember I played Crysis on very high (DX9) settings with my 8800gt for hours and it never made my PC shut down just like if someone pulled off the power cable out.

What about cpu temp and vcore? Crysis is cpu intensive too. Do you have an aftermarket cooler? If not, and it runs hot, get one.

I have an ASUS Silent Square Pro, and I don't remember temps getting weird values. Will get HWMonitor running in the background for the CPU and RivaTuner for the GPU monitorin. BRB with them.

 

[EXT64]

Did running the fan at 50% lower the temps? You could try it even higher or with the case side off. It sounds like it is intentionally shutting down for some reason.

 

[redhawtcar]

Here we go:

Catalyst 8.2, Catalyst A.I: Standard. Played for 7 minutes and had a shutdown, so these are the temps after playing 2 minutes (Afaik 2 minutes is enought to get max temps, and I don't want more shut downs in case I **** up my computer).

 

[cisco]

Usually random reboot especially under stress are caused by the PSU. I agree that 750 should be adequate, but as others have said it might not be functioning properly. I would start by having TT replace that PSU under warranty. I assume your not overclocking by the numbers and running stock timings on the ram. My money is on the PSU, if you have a friend with a good computer try swapping supplies and see what happens. Or do like I do, and use it as an excuse to start upgrading everything until you find the problem. There are only 4 things that could even cause the issue: PSU, Heat, Motherboard, Ram.

 

[redhawtcar]

Usually random reboot especially under stress are caused by the PSU. I agree that 750 should be adequate, but as others have said it might not be functioning properly. I would start by having TT replace that PSU under warranty. I assume your not overclocking by the numbers and running stock timings on the ram. My money is on the PSU, if you have a friend with a good computer try swapping supplies and see what happens. Or do like I do, and use it as an excuse to start upgrading everything until you find the problem. There are only 4 things that could even cause the issue: PSU, Heat, Motherboard, Ram.

Unfortunately I came from "an excuse" which was exchanging a faulty 8800gt for the 3870X2 (I paid the 100€ difference), and I really don't want to spend more money on hardware for a long time now. I'll keep testing a bit more things but I see myself RMA'ing the goddamned PSU. Lets see.

 

[redhawtcar]

P.D: I can't exchange PSU since noone I know needs such ammounts of W to wank arround at msn and stuff.

 

[Jpain]

1. Do you have 2x6pin or 1x6pin and 1x8pin connectors?

 

[redhawtcar]

1x6pin and 1x8pin connectors, both connected.

 

[redhawtcar]

Ok, fan at 100% (RIDICULOUSLY HIGH) and I had a shutdown at 9minutes 20 secs of gameplay. I think we can conclude and say this is not an overheating problem.

 

[stoner133]

How do you figure running the fan at 100% would eliminate a heat problem? You keep posting how long it ran instead of how high the temp got on the card. You last posted a temp of 81 at 50%. You never posted what case your using and if there isn't the proper air movement through the case that could be keeping the card to hot, after all it exhausts air out the back so it draws are in from inside the case. Play it with the fan at 100% for a few minutes and then check the temp before it crashes.

 

[firebird]

I was having similar issues a while back, except any 3d app would quit after x min. I did a few things at once so I'm not really sure what fixed it.

I had Asus mobo software installed that controlled ACPI, clock settings, etc. This wasn't working properly as I would randomly see high CPU usage from the exe related to the software. So I removed and re-installed that software.

I also turned on load line calibration for my CPU in the BIOS.

After that everything was fine. Se my system specs below.

And BTW, I tried everything from new CCC drivers to running the system at stock settings with no avail.

 

[redhawtcar]

How do you figure running the fan at 100% would eliminate a heat problem? You keep posting how long it ran instead of how high the temp got on the card. You last posted a temp of 81 at 50%. You never posted what case your using and if there isn't the proper air movement through the case that could be keeping the card to hot, after all it exhausts air out the back so it draws are in from inside the case. Play it with the fan at 100% for a few minutes and then check the temp before it crashes.

Dude I post a screenshot, look up. That was 50%, and as you can see max temp was 60. Max temp at 100% after 3 minutes was 59º.

 

[redhawtcar]

Btw, I got this response from another forum:

Quote:
The 3870x2 draws the same power as 2 3870s. The 3870 will not run stably on a PSU with less than 20A SUSTAINED on the +12v rails. SUSTAINED A (amperage) is found by multiplying the rated output by the rated efficiency. Your PSU offers 18A (rated) or 14.5A under load (assuming 80% efficiency). Not a good start.

I wouldnt run my 3870 on the PSU simply because there isnt a single +12v line with enough juice.

With Catalyst AI deactivated is the card actually using both cores? If not then you you can conclude that using both cores is unstable...likely because the power draw is too much. Other games cant load your system the way Crysis does.
What you guys have to say about that?

 

[redhawtcar]

I've been looking in the mobo's manual for temperature related options and found this:

CPU Enhanced Halt (C1E):
Enables or disables Intel CPU Enhanced Halt (C1E) function, a CPU power-saving function in system halt state. When enabled, the CPU core frequency and voltage will be reduced during system halt state to decrease power consumption. (Default: Enabled)


CPU Thermal Monitor 2 (TM2):
Enables or disables Intel CPU Thermal Monitor (TM2) function, a CPU overheating protection function. When enabled, the CPU core frequency and voltage will be reduced when the CPU is overheated. (Default: Enabled)

CPU EIST Function:
Enables or disables Enhanced Intel SpeedStep Technology (EIST). Depending on CPU loading Intel EIST technology can dynamically and effectively lower the CPU voltage and core frequency to drecrease average power consumption and heat production. (Default: Enabled)

Should I try to disable something? (I've search about these options and when overclocking people disables a couple of those).

 

[Shadow703793]

^ Those settings shouldn't effect the GPU. They only effect CPS. And yes, I would have to agree with the guy on the other Forum about the PSU.

The 3870x2 draws the same power as 2 3870s. The 3870 will not run stably on a PSU with less than 20A SUSTAINED on the +12v rails. SUSTAINED A (amperage) is found by multiplying the rated output by the rated efficiency. Your PSU offers 18A (rated) or 14.5A under load (assuming 80% efficiency). Not a good start.

He is pretty right. I say you PM Zorg and ask him his opinion, he knows best about PSUs.

 

[firebird]

My system is running on 18A max rails for 12v. I know it's not the same PSU, but specs look similar.

 

[Maximus_Delta]

I don't think it's the graphics card overheating... normally you get artifacts, screens freezes, juddering etc when it overheats - not a reboot on the basis of my experiences.

If I had to bet, I'd say it's the PSU. Crysis is the most demanding game out there... it causes machines otherwise stable to have problems... nothing else will load the rig like Crysis.

 

[dagger]

Right now, this stuff is too murky for me to take a guess. But you should not rule out psu as cause. Any psu, no matter how powerful, can possibly be defective. It's a matter of bad luck.

 

[EXT64]

Like I mentioned before, those were the exact symptoms I had when my PSU was insufficient for my 2900XT. How many rails does that PSU have?

 

[redhawtcar]

Goddamn the same happened while playing COD4

 

[redhawtcar]

Like I mentioned before, those were the exact symptoms I had when my PSU was insufficient for my 2900XT. How many rails does that PSU have?

+3,3V 30 A
+5Vsb 3 A
+5V 28 A
+12V1 18 A
+12V2 18 A
+12V3 18 A
+12V4 18 A
+12V Total 60 A
-12V 0,8 A