pc shuts down during 3dmark vantage

S

snail

Expert Member
Joined
Feb 5, 2008
Messages
1,395
Reaction score
5
Location
Toti
Ok, so my pc keeps shutting down after only a few minutes when running 3dmark vantage. It does it no matter what test i run in the program.

Now i have run intel stress test to eliminate the cpu, and that passed successfully. Now i need to find another program to eliminate the gpu (other than 3dmark).

My setup is as follows,

cpu: i7-860
GPU: 5770 crossfire
motherboard: asus p7p55d-e pro
ram: 8GB mushkin (2 x 4GB)
HDD: 500GB WD caviar black
power: 600w omega
OS: windows 7 64bit

Could it possible be the version of 3dmark that i have that would be causing this, as it installed it in x86.
 
Have no idea, where can I find a log of that. as soon as the pc boots and gets back into windows then I go directly into the 1st gpu and it sitting at 66 degrees. Could it maybe be a setting under the profile that is causing it to shut down due to temperature.
 
Use one of the many free utilities like SpeedFan & ATI Tool, ATI Tray Tool etc to monitor temps of all components.

I don't think Speedfan monitors GPU so you might have to run two utils at once.
 
Last edited:
You must remember that temp is for my top card that is in crossfire. The bottom card runs several degrees cooler than that
 
under where in ati tools must i go to run a stress test?
 
Under the ati tool it states gpu mem bios is 1200 but the current mem I'd 600, and clock bios is 850 but the current clock is 157. And the asus smart doctor states its set at 850 and 1200. When I run the mem test in ati tool its fine at 600, but when it runs the test at 1200 it goes faulty.

Any ideas. I will upload screen shot soon
 
Snail, which PSU is that EXACTLY?

Is the PC shutting down or just turning off?

What kind of ventilation have you got in the case? I doubt it is cooling problem but best to ask... CPU Cooler, GPU cooler and Case fans, their location and how many, etc. would help. Which case are you using also, plox.
 
Last edited:
I'm using HWMonitor to monitor the temperatures of my CPU and graphics cards.

66'C isn't that hot, seeing that my one MSI Cyclone GTX 460 gets 68'C and I have no issues whatsoever :)

You can also use PassMark's BurnInTest to stress your PC, and it would allow you to select the PC components that you want to stress test.
To stress the CPU, I usually just use 7-zip's Benchmark option.
Please run the HWMonitor (or other temperature monitoring software) in the foreground until you've solved this issue!

My guess is that your PSU is a piece of crap. Your setup isn't supposed to use more than 400W.
 
Pada said:
My guess is that your PSU is a piece of crap. Your setup isn't supposed to use more than 400W.
Click to expand...
x2

If it suddenly turns off that is probably the motherboard/PSU/GPU (any of them) forcing the system to turn off. PSUs usually turn the system off in event of: Overload (under volt, over current, short circuit) or operating temperature exceeded. Motherboard may also include such safe guards in case of under volting. My vote is the motherboard is protecting the system. Usually cheap-o PSUs just explode (no joke, was at a LAN just 3 weeks ago, guy had a Thermaltake and it exploded, 3rd Thermaltake I've seen explode, worst PSUs ever bar none).
 
I still want to see a power supply explode :D

I've also read that the Huntkey power supplies also tend to explode if you try to pull their rated power! I find it quite funny that it's the only brand of power supplies that PCInt sells - LOL.

If it's the CPU that's running at too low voltage, then Windows should give you a BSOD with an error code.
Here are the BSOD codes that I know of which are relevant to overclockers:
0x00000101 vcore (CPU voltage)
0x00000124 QPI/VTT voltage
0x00000050 memory timing/frequency/uncore multi
Click to expand...
 
find a screen shot here, http://mybroadband.co.za/photos/showphoto.php?photo=20341&cat=500

notice the differant settings in the ati tools, it states that the bios settings are 850/1200 but the current actual settings are 157ish/600. When i run the artifact tester and click on "mem" it runs at 600 fine, and then changes to 1200, which then gives a problem after a couple of seconds.

When i run 3dmark the entire pc just switches off, and switches on after a couple of seconds, but when i run ati tool, the monitor stops working, but the pc continues to run.

Here is my psu,
http://www.systemshock.co.za/forums/index.php/topic/11295-omega-600w-psuantec-psu-tester/
 
Sorry, couldn't save it any bigger. Is there no way u can save it to pc and zoom in on paint.
 
snail said:
Sorry, couldn't save it any bigger. Is there no way u can save it to pc and zoom in on paint.
Click to expand...

That's kinda like the movies/CSI where the idiots zoom into a guys sunglasses to get an ID. You cannot turn schit into gold.
 
Sorry, will try upload it somewhere tomorrow as I'm currently running off my phone.

update: sorry guys, cant upload any bigger than that.
 
Last edited:
alright, i have tested each card individually in each pci-e socket, and i still get the same error. So i can presume i have ruled out gpu fault, and power supply (seen as though both cards are giving same problem). so i went into my bios and disabled the tprobe setting (designed to optimize power to areas, google asus tprobe), and now when i run ati tool find max for mem i get either a light blue screen, no not the blue screen of death, just a light blue screen, but pc still running, or i get light blue and white stripes running vertical down the screen.
 
So you have tested with another PSU?

Not 100% sure the PSU can be discounted just yet. I just found this thread online: http://www.tomshardware.co.uk/forum/279351-15-600w

Please tell me this part:
Tomshardware said:
One rail at 16a and the other at 17a from what I can tell
Click to expand...
Is not true.

Just to put that into perspective, I have a Seasonic S12II 430W here at my house in an older computer, 12v rails are rated: 17A each (that is 34A). And the Seasonic CAN actually provide that much power, my guess is that PSU you have is around 350-400w unit in reality. My 600W Cooler Master Silent Pro supplies 40A on the 12v rail.

Can't you still return that PSU and get something decent? (no offense). Just try with a decent PSU first. That SystemShock forum makes this PSU sound like a real winner but I'll level with you here: That PSU is about as bad as it gets. It is the no-name of no-name brands. I especially like the part in the SystemShock forum where the guy talked about his current ThermalTake PSU not hacking it anymore and he will be upgrading to a Omega unit, eish.

Pada said:
I still want to see a power supply explode :D
Click to expand...
LOL, we were all gaming and there was a little bit of rain outside and some light thunder (you know really light, not the kind you power down for). And suddenly there was a bright flash, and *BAM*, inside the room, I thought it was lightning at first. Not much smoke but it did smell like burn. When I touched it tho, bugger burned my hand. I kid you not the PSU was as hot as an oven. 10 Minutes later it was still hot to the touch (tough not burning your hand).

Same happened to my own TT PSU, also while gaming, my case was closed tho (friends was open @ the LAN so you could see the flash). My other friends also died but long time ago. All of them different TT units.
 
Last edited:
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X