3d Games Crashes.....please Help

Mine is a long, sad, and from looking round the web an all too familiar story:

This is the best forum I’ve found so far, you lot actually seem to know what you’re talking about so here is my tale:

I’ve built my own computer, my specs are as follows:

OS Name Microsoft Windows XP Home Edition
Version 5.1.2600 Service Pack 2 Build 2600
OS Manufacturer Microsoft Corporation
System Name HOME
System Manufacturer 75V80
System Model 775VM800
System Type X86-based PC
Processor x86 Family 15 Model 4 Stepping 3 GenuineIntel ~2999 Mhz
Processor x86 Family 15 Model 4 Stepping 3 GenuineIntel ~2999 Mhz
BIOS Version/Date American Megatrends Inc. P1.30, 14/10/2005
SMBIOS Version 2.3
Windows Directory C:\WINDOWS
System Directory C:\WINDOWS\system32
Boot Device \Device\HarddiskVolume1
Locale United States
Hardware Abstraction Layer Version = "5.1.2600.2180 (xpsp_sp2_rtm.040803-2158)"
User Name HOME\Gareth
Time Zone GMT Standard Time
Total Physical Memory 1,024.00 MB
Available Physical Memory 697.02 MB
Total Virtual Memory 2.00 GB
Available Virtual Memory 1.96 GB
Page File Space 2.31 GB
Page File C:\pagefile.sys


My problem is that every time I play a 3d game (call of duty 1 & 2 , farcry, SOF2 etc.) is that about 10 minutes into playing my computer crashes.

Normally this is just to desktop, with an error report that doesn’t offer a solution. Occasionally, but not for a while, I get the old BSOD, followed by a reboot and ‘recovered from serious error’ message, which leads to a more than useless Microsoft site telling me it may or may not be a driver conflict (I’ll tell you more about that in a minute).

My computer is totally stable otherwise: I’ll list all the things I’ve tried so far:

1. I’ve ran 2 different types of mem test for over 12 hours: no problems
2. I’ve tried 2 different motherboards and 2 sticks of ram and 2 processors: (the shop I bought them from was very helpful)
3. I’ve tried 2 different graphics cards (a radeon 980xt and now my geforce 6600)
4. I have reseated all the components and checked the connections
5. I have replaced the hard drive
6. I have tried repairing windows, I have reinstalled a raw copy of windows.
7. I have used windows update.
8. All my drivers are up to date: I have used a driver uninstall utility to remove old drivers (in its latest incarnation my I did a raw install and only installed the latest drivers (no conflicts))
9. I did have a creative sound card, but I have since removed that and uninstalled its drivers
10. I have installed motherboard monitor and watched the temperatures before, during and after a crash: my cpu and gpu never get above 50. Just to make sure I have both sides of the case removed and a large fan blowing into it. (Believe me im freezing)
11. My bios is flashed and up to date
12. I have followed the bios set up thingy on this site and have disabled all the bollocks like game ports etc.
13. I have no IRQ sharing or conflicts
14. I am running Norton internet security, but any time I install anything I have used the ‘msconfig’ thing to boot without startup items.
15. I have tried playing games in this mode as well; the results are the same
16. I am about ready to throw it form the window.
17. I have also used gamexp to tweak my ram settings as well, but have since returned everything to it windows default.


I’ve probably done other stuff as well but I cant remember.

I really am at a loose end, the reason im posting here is that I have read somewhere that it could be a faulty PSU. I wanted to see what you guys think before I go splashing anymore money on this god forsaken typewriter:

My PSU is a Jeantech (bought from PC World here in the UK for £45 ($80))

It is rated at 425.5 Watts; here are its particulars:

+3.3v = 30amps
+5v = 28amps
+12v1 = 14amps
+12v2 = 15amps
-12v = 1amp
+5v = 2.5amps

I might be answering my own question, but the 14/15amps on the 12vlt rail seems a bit low to me??? Especially as both graphics cards I tried have had separate power inputs. Is this the cause of my problems???

Also the motherboard monitor shows the following results for my voltages:

+3.3 is averaging at 3.3
+5 is averaging at 5.08
+12 is averaging at 11.46 (with a low of 11.37)
The graphics card is on the 12v isn’t it??? Is this my problem?????

If you need any more info or copies of my minidump files (from Dr Watson) etc. please ask (they don’t mean anything to me.

I think it’s the power supply, but I would really value your opinions before I go and splash the cash. This is the single most infuriating PC problem I’ve ever ever had.

What worries me is that there are loads of people on various forums with similar problems and I cat find a comprehensive solution: some people claim to have tried new PSU’s without any joy.

HELP before I turn to the xbox for ever more……

Playing for 10 mins then a crash usually indicates overheating, but in your case your temps are ok. But this does not mean the PSU temps are ok and you are correct that Jeantech 428W might be on it's upper limit. 12V 29A is not too bad, but depends on how many case fans, HDD (looks like just 1) and DVD drives you have. If it is running nears it's limit it will heat up, add some heat and it's efficiency goes down as does it's current output and you have an unstable PC.

The other thing to try first is have DirectX9.0c and run DxDiag, make sure no errors/conflicts. Is this an AGP card? Do you have Fast Writes in BIOS turned off? Your virtual memory seems a little high for 1GB of RAM, I use 1.5X my RAM size.

For those games go to TweakGuides.com and read about optimizing your PC settings. One quick help is to up your HunkMegs or Heapsize (term is game dependant). This is amount of RAM allocated for game use. R click on the game launch icon/properties. In the target line add this: -noipx -console -heapsize 512000 (could be HunkMegs for COD2)

The PC crashing prob is the hardest thing to troubleshoot unfortunately, there are so many variables.

Try those for now and see what happens and keep us posted.

cheers.

Thanks for all your help and interest.

In reply to your questions:

I have run dxdiag with no problems at all.

It is an agp 8x card, I have disabled agp fast write and have tried all possible combinations of aperture size and agp 4X 8X etc.

Although there is two rails of 12v adding to a total of 29 amps, one rail is maximum 14amp the other 15amp, I seem to remember reading that some GPU’s need at least 17-18amps. I have tried mixing the connected in different combinations and even had the GPU on a rail all of its own, all to same result.

I have looked at tweak guides before, but as a general point you shouldn’t have to mess with stuff like that just to get a game stable, they all ran fine on my previous system with no messing about.

I will try lowering my virtual memory to 1.5 times my ram and see if that helps.

I have not overclocked anything; i have infact tried underclocking components in an effort to improve stability; all to no avail.

The ram is a single stick of unbranded 1gig ddr400, i have tried running at different speeds (slower) to see if that helps: no joy. As i said before i have tried 2 different sticks of ram, both of which tested fine on two different emtests over a 12 hour period.

The motherboard is an asrock 775vm800; it is more than capable of handling the components im using. The processor and ram came shipped with the motherboard, and as i said i have tried 2 different lots now (in various combinations).

The processor is a 3gig p4 with hyperthreading and all that stuff, my moptherboard supports this chip and its features entirely.

As a side note i forgot to mention i have antispyware as well: the microsoft one and the one from lavasoft, so i dont think its a spyware issue. I also dont have any viruses or trojans.

I would be surprized if its the monitor: I have kept the same monitor fomr my old system (amd2600xp, 2X512ram333, geforce 4200ti) which used to run all these games fine (all be it at a lower resolution).

I have tried running all the games at low resolutions and even in safe mode: wih the same result everytime (about 10-15minutes into the game the crash).

I am about 99% sure it is not a heat issue as i have monitored temperatures of case, mb, processor and gpu (although if anyone has knows of any alterative gpu tempertature monitors i could give them a try.)

The LED's do not dim at all, but they are only small anyway; its been a long time since ive had the BSOD; almost every crash now just goes to the desktop, i can then restart the game straight away and play for another 10 minutes (ish).

Oh and BTW I have tried increasing the voltage to the graphics card and ram both seperately and together (in bios), all to no avail.

Heat wise I have have the heatsink and fan over the processor, heat sink and fan on the Graphics cards, and a case fan. As i said i have also removed move sides of the case, turned off the heating in the room, opened the windows, and had a fan blowing into the case. I have also monitored the temperatures and the processor was the hottest thing reaching a whopping 48celcius.

If yu can recommend a decent gpu temperature monitor i would appreciate it as im not convinced of the accuracy of the one im currently using.

The bsod has not been a problem for a while; it just crashes to desktop. When it did do that the errors were always different but googling them refered to a memory problem.

from what i've read on various sites altghough the error message may make it loook like a ram problem; these ram problems can infact be caused by a dodgy power suppply; is this true??