wesbuck

Distinguished
May 21, 2008
55
0
18,630
I am having some problems with my new build that i put together the other week, just seeing if anyone has a solution to this problems as i have bin racking my head for a while now.

Here are the specs (although i ran out of money for a new processor so reused an old one):
Antec 900 gaming case
Gigabyte GA-EP35-DS3 Socket 775
Seagate 500gb 7200.11 32mb
Pentium D 930 3GHz
Muskin geforce 8800GT 512MB over clocked edition
Cosair HX520 modular PSU
OCZ DDR2 PC2-6400 5-5-5-12 2x1GB
Dell E198FP 19-inch LCD Flat Panel Monitor
XP SP2

It is concerning playing games on my new rig, for example with counter strike source I load the game up as normal, and to test my pc was working properly i ran the video stress test, and after about 5 seconds the screen goes blank with no output being read from the computer, the computer remains on and the only thing i can do is switch the computer off manually and restart. This also happened when i downloaded and tryed the demo for crysis and lasted about 5 min before doin the same, also supreme commander did this after about 15mins. Now the odd thing is i also tryed bioshock and this works perfectly at medium settings. I updated all drivers and have installed everything correctly as far as i know.
Now what i can't work out is why it is doing this, does anyone have any idea?? The only thing i can think of is some issue with the old cpu maybe overheating, although i check the bios after this kind of crash and it reads about 30ºC although i can never read the cpu temp with any of my programs in windows like everest.
Thanks for any help anyone can provide.
 

Andrius

Distinguished
Aug 9, 2004
1,354
0
19,280
First get some proper software for monitoring temperatures and stressing components.
http://www.cpuid.com/hwmonitor.php

Prime95 for CPU load tests : http://mersenne.org/gimps/p95v256.zip

FurMark for graphics load test : http://www.ozone3d.net/benchmarks/fur/
It's free and runs in a window to help you monitor temperatures.
If you have 3D Mark 06 you can loop that as well.

This should help you isolate the problem.
Check to se if either of them by them self cause your system to crash.
A CPU stress crash would likely mean a poorly mounted CPU cooler.
A GPU stress crash could mean a faulty graphics card.

Test and report back.
PS: a screenshot of HW Monitor (alt+printscreen, paste to paint and save as an image, uploaded to www.imageshack.us) would be helpful.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Ok here is the CPUID HW screenshot, running all the test now sorry for the delay, i live in southern spain uno what its like mañana mañana and didn't help with the power cut lol
hw.jpeg


Thanks again for ur help.
 

Andrius

Distinguished
Aug 9, 2004
1,354
0
19,280
No problem. I just had lunch at work too (Europe/Slovenia).

These are idle temperatures I gather and they look very good (Antec Nine Hundred was a good pick). It's a bit annoying you kept the old D 930 as it doesn't have onboard temperature sensors.

You'll need to figure out (Prime95 is the proper tool for it) which temperature rises the most.
You can also try Speedfan 4.34 but it needs to be configured to be accurate. http://www.almico.com/sfdownload.php
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Yes it was a good pick keeps it very cool indeed, added an extra fan to the side so it blows onto the gpu.

Yes they are the idle temps, i've been running prime 95 for about an hour now and the TMPIN0 is max at 33ºC and TMPIN1 is max at 37ºC. Although i do not know how this exactly corresponds to the actual cpu temp, i think the TMPIN1 is the motherboard temp and TMPIN0 is Aux but i dunno what this is cos it says so in everest. Also i've heard that pentium D's do run pretty hot, so i bought a cpu cooler locally for 12€ because the old dell case it came from had a cooler that could not be changed over.

How long should i run Prime95 for or will it stop by itself and how long will this take??

Thanks again for the help.
 

Andrius

Distinguished
Aug 9, 2004
1,354
0
19,280
1. Those temperatures are likely not the CPU yes.

2. Prime95 : (if it ran for 1 hour and didn't crash you are likely safe)
Make sure you use the following settings :
Advanced -> Round Off Checking (enables error reports)
Options -> Torture Test (blend test, 2 threads)

3. Run FurMark in a window.
Enable "Stability test". Remove the "Fullscreen" checkmark and apply a resolution that will fit on your screen, 1024x768 should be good. Then click "Go!".

It's a good idea if you leave such tests running over night. Sometimes errors sneak up after several hours.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Ok thanks for the help i'll leave the prime95 as everything seems to be fine and try the furmark now, i'm beginning to think it is the graphics card as muskin seem new to the business and also my friend said that some people have bin having problems with this particular card.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
right ok i've ran furmark for an hour and a half with no failures with a max fps of 96 and average of 72, also my gpu stayed around 75ºC so i dunno if this is high or not for an 8800gt.

thanks.
 

Andrius

Distinguished
Aug 9, 2004
1,354
0
19,280
Looks like everything is stable on it's own.
What remains is a combined load test running both Prime95 and furmark at the same time.

But to me it looks like you have a software issue at work with the games.
Do you have any strange/older hardware except for the CPU ?
Any <?> icons under SystemProperties/DeviceManager ?
Latest BIOS?
Latest DirectX?
Drivers are latest you say.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Thanks for all the help andrius really apreciate it, i can't get to the computer tonight but will try it tomorrow morning with both.

The only things i have is the IDE dvd drive from my old dell although i play the games from the hd so don't see this being a problem and also a netgear wireless pci card to connect to the internet forget what model it is. And i presume its the latest bios as i only bought it the other week but i will check and yes i installed directx9c.

thanks again
 

Andrius

Distinguished
Aug 9, 2004
1,354
0
19,280
It's also a very new board so it could be a buggy BIOS.

I've also just thought of this when you said it's a new board. Have you installed DES (Dynamic Energy Saver)? If you have and it's running try running without it. It could be that the old D930 is a bit sensitive to it.

What remains after that is : going into BIOS and disabling any unnecessary devices like LPT/COM/USB(not if you have USB keyboard and mouse) and checking what happens. If the problem persists you should take it back to the shop and let them have a look at it.

Good luck.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Hmmm ye it could possibly be that, ye i installed DES but always close it when doin anything that requires a lot of power, as i presume it holds the system back, ye i'll do all of them things.

Ye taking anything back shouldn't be a problem a bought from http://www.pccomponentes.com/ and i have a friend who works for them so an easy job.

thanks again for the help i'll report back tomorrow.
 

wesbuck

Distinguished
May 21, 2008
55
0
18,630
Ok today my friend lent me his club hd 3850 graphics card to try in my computer and low and behold it work every gamed played perfectly, so i am sending the graphics card back to the manufaturer to see what comes back. Hopefully get a replacement or my money back to buy a different one.

Thanks for all the help andrius really apreciate it.