Problems with new hardware (games)

[koellaa]

Hello everyone!

I've been searching for a good HW-forum, and I think I finally found it =)

I recently bought myself a new computer, but I keep having this problem with gaming.. First I ran Win XP Pro x86 (just temporary), and when playing games like FM2010 and Assassin's Creed (especially the last one), my computer kept freezing and rebooting after 10-20 minutes.. I tried un- and -installing all drivers so they were up to date, but with no luck.. I also tried some programs like SpeedFan and SmartDoctor, but none of them showed signs of overheating (CPU/MB: 30-40C, GPU: 40-55C).. Then yesterday I finally got my Vista x64, and thought that might help.. Installed all drivers, installed the games etc.. FM2010 seems to work fine (it's not very GPU-heavy), but Assassin's Creed (and LotR:Conquest) still keeps rebooting and freezing after a short time of playing.. I really don't know what to do with this, so I would be ever thankful if someone could help me find a way to fix this =)

Specs:
OS: Vista Business x64
PSU: MIST 650W
MB: ASUS M4A79XTD EVO
CPU: AMD Phenom II X4 955 Black Edition
RAM: OCZ Platinum Dual Channel 4GB i5
GPU: ASUS Radeon HD4870 1GB DDR5 (single VGA, both power lines connected)

If any other info is necessary, just ask =)


Thanks in advance!

 

[koellaa]

Still having this problem.. Have tried a lot things to make it work properly, but with no luck..

- Temperatures before and while gaming (though the restarts also occur while surfing the internet, listening to music etc)
- Run chkdsk on both drives
- reinstalled and updated drivers for gpu, audio etc
- installed all windows updates (except for one of the .net framework updates which is listed as failed)

I really need some help with this =)

 

[koellaa]

Anyone have a clue if it's a hardware or software problem? Will WinDBG help me out, and if it can, how does it work? I'm really desperate.. Can't use the computer for more than 20-30 minutes without freeze or restart...

 

[obsidian86]

what power supply ?

 

[koellaa]

MIST 650W with removable cables

 

[cory1234]

MIST 650W with removable cables

Sounds like it could be a couple of things.
First it definitely sounds like a hardware problem.
It is more than likely that your power supply is dying. You can test this by unplugging the ps from the motherboard, connect a fan, and use the paper clip method to jump start the ps.

http://www.overclock.net/faqs/96712-how-jump-start-power-supply-psu.html

Download prime 95 and run the blend test for two hours. Monitor temps using realtemp or coretemp while running the test, speedfan is usually low ~10C. Speedfans 20C is likely 30C. If it fails trying running just the small ffts test. If the second test passes you probably have bad memory. You can run memtest86+ to also test this. (Use one stick at a time in the same slot on the motherboard)

Are you overclocking?

Your motherboard could also be faulty, but I doubt it since your able to use your computer for some time before it fails.

 

[koellaa]

PSU is dying though it's only 1 month old?

I will try this, and will also run the tests.. I'm not overclocking, btw.. Thanks alot so far =)

 

[koellaa]

Didn't have time throughout the weekend, but now I'm back.. Had a look at the PSU jump start article, and to me it looks like it's just for testing if it will even power on.. In my case it works fine for some time before the system crashes.. So no need for that? Or will it show other problems?

Will download prime95 now and run those tests (have also downloaded memtest86+ for backup) =)

 

[koellaa]

10 seconds of the stresstest was enough:
[Mon Nov 23 11:38:05 2009]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

Is this memory voltage or something? I run and AMD-based system, though my memorysticks are Intel-optimized (i5).. Didn't think this would be a problem, tho.. Coretemp quickly went up to 42-43C when I started the test, but I don't think CPU overheating is a problem.. It has never shown signs of such earlier (speedfan while gaming)..

Will try the other test now..

 

[koellaa]

5 tests on the small ffts test without a problem.. CPU @ 44-45C.. Running the large ffts test now - at test 2, no problems, CPU @ 44-45C..

Not CPU overheating, then.. Bad memory, or can it still be the PSU causing the problems? Should I try memtest86+ first?

 

[koellaa]

Checked around in BIOS, and found DRAM Voltage at Auto.. Thought I'd try to change it, set it to 1.68V (ocz.com says 1.65V) - now the blend test seems to work fine.. Thanks for the help, cory =) Hopefully it will work perfect!