Please Help, I'm freaking sick of this!

M

Manic_Midwestern

n00b
Joined
Jul 23, 2007
Messages
61
I am plagued with game crashes. They are pretty random but only happen during games. Some times I can go a few days without a crash and other times I can't go 5 minutes without a crash. The games I have that it happens most on are Assassins Creed and Dirt. I will be playing then all the sudden everything will freeze but the sound will continue and some times it will kick back to the game after the VPU Recover thing and other times i have to just reset. I have all the latest drivers for every spec of hardware on my computer. Here are my specs go ahead and ask any questions you want and saying you doing like amd or ati isn't considered helpful. I have a hunch it might be the power supply but its weird because of the randomness. I played Assassins Creed last night for 3 hours without a hitch and today I can't even make it to my objective without the game crashing.

CPU = AMD Phenom 9500 Agena 2.2 GHz
GPU = 2 X HIS Radeon 3870
RAM = 2 X 2GB OCZ Reaper DDR2 800
HDD = 2 X 320GB Samsung Spinpoint T
MOBO = Gigabyte GA-MA790FX-DS5 (flashed to F6)
PSU = XCLIO STABLEPOWER 1000W
 
I'd say try testing with a different power supply. I use to get random crashes on my old rig before I changed the power supply to a better brand. After that it was smooth sailing.
 
i'm very stupid when it comes to pc hardware, does the "(flashes to F6)" mean that you're altered the mobo? is it possible to reset it to factory standards? also are you running a lot of programs in the background? check your task manager and see how many processes you're running.

it might be a good idea to reformat your computer.

i would like to point out, that while i really miss playing pc games (nwn, cs:s) that game crashes were a big reason i moved to consoles. i know that many people will say i'm too stupid to game on a pc, but that's fine.

sorry didn't mean to threadjack. good luck on fixing it
 
There isn't enough info to know what's wrong..

Could be anything from viruses, to bad game coding (not likely) to a bad OC, to just plain faulty hardware.

Can't help you with that little information
 
Have you made a note of your temperatures? Are they high? Things you can try to narrow down the culrpit (in order):

First:
Make sure you have the additional power connectors plugged into your video card, and also make sure there isn't any other devices leaching from the same cable going to the card.

Second:
Run memtest to make sure you dont have a fault stick of memory.

Third (if you can):
If you have an alternate power supply, give it a shot. If you don't have one, try unplugging any devices you do not need to boot up and test (cd/dvd drives, spare hard drives, etc).

Fourth:
Remove any additional add on cards and see if that helps.

Fifth:
Have you tried running a single 3870 to see if its stable? If one is stable, try the other by itself.. if it crashes with that card well its obvious where the issue lies.

Honestly, it could be numerous things. Without some basic trouble shooting to help narrow it down, any advice we give you is a shot in the dark.
 
One of the strange behaviors of Multi GPU setups is response after sleep mode.
I've had crashes on a 4870 Xfire rig and a 9800GX2 after sleep/hibernate resume, but none after a fresh reboot.
I would suggest you check tweakguides for basic PC troubleshooting/setup and take it from there.
If it is a hardware issue the first thing I'd check is every GPU separately ONLY on the games you get crashes to see if they're both OK. Then I'd proceed to RAM and then to PSU...
Good Luck and post back what you find!
 
google "prime95" and download a copy, run one copy per core of your CPU, just select the torture test and run the test that says it tests lots of memory and see how long you can run it without errors, if it's above 8 hours your ram and cpu are fine.

google "furmark" and download a copy and run that for as long as possible again, see if you get any issues, crashes or BSODS

you need to stress test your hardware to find whats wrong, if you have errors then download something to measure temps with for your kit and post us the temps.
 
I would test with a different power supply...also, do you have a seperate sound card(not onboard)? I was plagued by crashes with my Audigy 4 because Creative sucks at anything driver related.
 
Do you overclock? ANYTHING? If so - don't.

Just try it. Run a week without anything overclocked, see if you have any crashes.

If still - then I would suggest looking at your OS (drivers, etc) or power supply. OS is easy to check if you have another hard drive laying around (even an old one). Just pop that in, reinstall a 'clean' Windows install (NO third-party apps, NO IE or media player plugins, NOTHING except Windows, your video and sound drivers, and the games you are having a problem with), and run that for a week. See if that crashes.
 
i would run memtest first before everything else. Crashes are most likely due to memory problems.
 
Does the game just crash and boot you to the desktop or does it blue screen? or... does it restart you computer?

If it restarts its most likely a video problem, I would check the PS aswell


If it blue screens its a memory problem.
 
I was crashing in Assasin's creed a lot at first on my Radeon. I then went to the folder and made sure I was runing dx10 mode. After that I enabled Catalyst AI and I managed to beat the game without any further crashing. I think my crashes were related to my Creative Xfi. Infact it hasnt crashed since taking that shitty peice of hardware out of my pc.
 
I originally thought it was a heat issue as well but I checked and my CPU maxes out at about 50C-55C and my VPU's each max out around 59C. When I say my bios is flashed to F6 it is the Factory Flash. If you know the board it came out at the beginning of the ATI/AMD spider platform and has had lots of updates to tweak things and make stuff work smoother, however I have had the crashes since F1 the factory bios. I have reformatted a few times and even downgraded back to xp but no luck. Also I have tested the memory, thats what I had assumed it was before and I even switched over to the OCZ from some mediocre memory to try and help but to no avail.
 
Also i just bought a new sound card its an Azuentech X-Mystique because my old one wouldn't send 5.1 through the Optical wire to my receiver. Is there a way to test the functionality of my power supply?
 
Power supply testing equipment that would actually be able to load-test your unit would be fairly expensive. I suggest you hop on the internet and look up some reviews for your specific model. See if there are any common issues with it being unable to handle load, crashing under heat, having high power fluctuations, etc. Since you mentioned that you get a VPU Recover screen when this happens, it leads me to believe that you're problem is directly related to the video cards, be it drivers, power, clock speed, heat, or whatever.
 
Do as I recommended, test each video card individually and see if it occurs when only one card is in the system, then swap it out with the other and test again.
 
Everyone pointing to the Stablepower as the cause of this problem... I am fairly sure you're wrong. Xclio isn't a very well known brand, but they make very solid products, and the Stablepower line is their high end stuff. I believe that's a 1kw CWT build -- it can handle the OP's system without pulling anything more than 50, maybe 60% of its rated capacity at full load. Unless it's just flat out defective (or being fed with very poor quality power due to old/faulty wiring, perhaps) it is not the problem here.

Dome said:
i'm very stupid when it comes to pc hardware, does the "(flashes to F6)" mean that you're altered the mobo? is it possible to reset it to factory standards? also are you running a lot of programs in the background? check your task manager and see how many processes you're running.
Click to expand...

All the OP means by "(flashed to F6)" is that he updated his BIOS and is currently running the F6 revision of it. There is no reason to return to the factory default unless the problems described are known to be caused by this BIOS revision... which is very unlikely.
 
Two words. Motherboard Drivers.:D

Its scary how many times i've reformated and forgotten to install the mobo drivers. Causes no end of problems. Depending on what you do the pc can run for hours without a problem, other times it can crash in seconds.
 
I think I may have figured it out, but I did too many things to fully understand what the main culprit was. For one I put the computer on its own power strip. Two I down clocked my memory to 667. Then I down clocked my graphics cards to 750 mhz. Finally I went into the bios and manually assigned voltage levels for everything. So far I have gone a week with just 2 crashes and they were on load screens. Strangely enough after downclocking everything I now have more stable framerates and everything seems to look and play better. Crysis is running a solid 22 fps with very few drops.
 
Did you underclock your machine from stock clocks, or just lowered your overclocked frequencies?
 
Manic_Midwestern said:
I think I may have figured it out, but I did too many things to fully understand what the main culprit was. For one I put the computer on its own power strip. Two I down clocked my memory to 667. Then I down clocked my graphics cards to 750 mhz. Finally I went into the bios and manually assigned voltage levels for everything. So far I have gone a week with just 2 crashes and they were on load screens. Strangely enough after downclocking everything I now have more stable framerates and everything seems to look and play better. Crysis is running a solid 22 fps with very few drops.
Click to expand...

Down-clocking (below stock clocks) resolving the issue points straight at the PSU. OCZ Reper ram, especially, demands very PURE (and above-average) voltages. Variances in what comes out of the PSU will crash that stuff 100% of the time.

I don't have experience with that particular video card, but it doesn't surprise me to hear that downclocking it would improve things. Again, if a flakey PSU, the video card is often the most sensitive component in the system to it.

By lowering clocks, you increase tolerance for voltage variances.
 
My honest guess is that you have your RAM voltage set incorrectly. Triple check
 
You must log in or register to reply here.
Back
Top