Random restarts when gaming, could it NOT be the video card?

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Well I just sent my X1800xt back because when I was gaming my PC would just restart itself. I checked the clock speeds and they were UNDERCLOCKED, so I tried to bring them up to standard speeds, but it couldn't be done without the system restarting.

Just wondering now that the video card is being replaced, could it have possibly been a psu issue, or RAM issue? The system never restarted unless I was gaming and seemingly only when I turned AA on.
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Mobo: Asus P5W DH Deluxe
Processor: E6300
Video: X1800XT 256MB
Memory: 2gb corsair PC-6400
PSU: Seasonic S12 600w PSU

ONLY seems to happen in gaming when AA is on, left it on for over 5 hours downloading and nothing.
 

Slava

Distinguished
Mar 6, 2002
914
0
18,980
0
Your system looks good. It could be the RAM though. Download and run MemTest (it runs from a bootable diskette.) Run it in the evening before bed time and leave it running overnight. When you wake up in the morning there are two possible scenarios:

1. MemTest displays errors indcating faulty memory modules. Replace the bad modules then.

2. MemTest does not display any errors and keeps running tests. In this case just terminate the tests and reboot. If it does not find anything by morning there is nothing wrong with your RAM.

If (2) above then it is likely that your video card was somehow defective. Your PSU is good enough. I doubt that it is a PSU problem.
 

johngoodwin

Distinguished
Dec 15, 2005
94
0
18,630
0
Gaming puts a lot of stress on a machine. Most games take 100% cpu and render frames as fast as possible.

Possible causes of random restarting when gaming (Not necessarily in order of likeliness):
1) Overheating of CPU
2) Overheating of GPU
3) Overheating of Memory
4) Overheating of Northbridge
5) Bad system memory
6) Something out of spec (ie overclocking)
7) Heat expansion of some item causing grounding if two conductive items are too close.
8) Poorly written game (needs patches)
9) Bad drivers (maybe need driver upgrade)
10) Bad system DLL (maybe a program overwrote/corrupted an important DLL)
11) IRQ conflict (many items these days handle IRQ sharing well, but older video cards needed their own IRQ for stable operations)
12) Virus/Malware/Worm/Trojan/etc
13) Bad PSU/Under-spec'd PSU
14) Bad power from utility company
15) Driver conflict (two hardware items have conflicting drivers). You would be suprised how poorly some drivers are written
16) Memory leak in drivers.
17) Bad moon (maybe your lunar cycle is out of order - LCIOO problem)

Note: Problems are not limited to only 1 problem, but can be compound problems (more than one at the same time).

So to answer your question, yes, a lot of things could cause that problem.

John
 

nottheking

Distinguished
Jan 5, 2006
1,456
0
19,310
16
Well I just sent my X1800xt back because when I was gaming my PC would just restart itself. I checked the clock speeds and they were UNDERCLOCKED, so I tried to bring them up to standard speeds, but it couldn't be done without the system restarting.

Just wondering now that the video card is being replaced, could it have possibly been a psu issue, or RAM issue? The system never restarted unless I was gaming and seemingly only when I turned AA on.
From some of my experiences, the second-most likely source, after the video card, would've been the PSU. It does not matter if it "should" be good enough; after all, even good manufacturers occasionally let out a dud.

I personally don't buy all that crap about PSUs catching fire if they fail, or instantly frying all of your other components... What they DO do, though, is produce less and less power... And when your components aren't getting the power they need, they perform worse, and also heat up more... And this can frequently result in rebooting.

The third most-likely cause would be bad RAM; I've gone through a number of bad modules, and their errors can result in crashing; usually, it's a simple crash, not a reboot, though.
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Well Im sure a 16M super pi run equals about 20 minutes of gaming, as far as the processor, RAM, motherboard and northbridge are concerned?

Temperatures are fine, monitored the card up until the system restarted and temps never rose above 65C for the video card. The processor runs a 25C under full load and won't budge any higher, havn't checked the northbridge temps, but I'm waiting for the new card to arrive tommorow to do that.

I will run memtest and if theres no errors and I'm STILL having the exact same problems in games I'm going to have to assume it's the PSU. I doubt the processor is the problem and I've updated my motherboard to the latest BIOS and all seems healthy on it.

I'll get back to you tommorow with the new 7900GT, any more suggestions? Is a floppy disk large enough to make a bootable for memtest? And can I make the bootable floppy on one pc and use it on a different one?

For PSU issues, are there any giveaways? Like checking the rails (I need to know which software etc) The PSU supports 2x pcie-e, I know it shouldnt matter, but out of good practise should I use the first connector in the chain or the second? I was using the second last time.

For motherboard issues, is there a program I can log the northbridges temperatures to a .txt file and constantly logs and saves? This is what I used to monitor the video card temps.


Thanks for the help so far.
 

fainis

Distinguished
Feb 10, 2006
763
0
18,980
0
Well I just sent my X1800xt back because when I was gaming my PC would just restart itself. I checked the clock speeds and they were UNDERCLOCKED, so I tried to bring them up to standard speeds, but it couldn't be done without the system restarting.

Just wondering now that the video card is being replaced, could it have possibly been a psu issue, or RAM issue? The system never restarted unless I was gaming and seemingly only when I turned AA on.

yo ... it`s 10000000% your psu ....
you can leave it 23821908 days in idle .... or whatever . and still perfectly working....but if your psu is underrated the sistem will crash in games as your power consumption will increase dramatically......
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
ok well I'm now buying a 7900gt which has LOWER power consumption so theres a chance it might work. Im still guessing the PSU should be replaced though?

Are there any other tests I could carry out to confirm it's the PSU?
 

allred

Distinguished
Apr 28, 2006
41
0
18,530
0
And when your components aren't getting the power they need, they perform worse, and also heat up more...
Less power = More heat ..?

Damn computers breaking the laws of physics again...

There is no way that can be true.
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Unless the main share of the power is going to the GPU and the fan is recieving insufficient ammounts of power, but this isn't happening anyway cause I already stated the card is NOT overheating.


If anyone could answer the above questions it would be great, thanks :)

Starting to seriosly think it's a PSU issue... I'm going to see if the 7900GT works flawlessly, but even if it does I still might consider replacing the PSU, it cost a lot and if it isn't functioning properly then I want a refund/replacement.

Any other thoughts?

Thanks
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Gaming puts a lot of stress on a machine. Most games take 100% cpu and render frames as fast as possible.

Possible causes of random restarting when gaming (Not necessarily in order of likeliness):
1) Overheating of CPU
2) Overheating of GPU
3) Overheating of Memory
4) Overheating of Northbridge
5) Bad system memory
6) Something out of spec (ie overclocking)
7) Heat expansion of some item causing grounding if two conductive items are too close.
8) Poorly written game (needs patches)
9) Bad drivers (maybe need driver upgrade)
10) Bad system DLL (maybe a program overwrote/corrupted an important DLL)
11) IRQ conflict (many items these days handle IRQ sharing well, but older video cards needed their own IRQ for stable operations)
12) Virus/Malware/Worm/Trojan/etc
13) Bad PSU/Under-spec'd PSU
14) Bad power from utility company
15) Driver conflict (two hardware items have conflicting drivers). You would be suprised how poorly some drivers are written
16) Memory leak in drivers. A game once crashed with a memory error, no restart though.
17) Bad moon (maybe your lunar cycle is out of order - LCIOO problem)

Note: Problems are not limited to only 1 problem, but can be compound problems (more than one at the same time).

So to answer your question, yes, a lot of things could cause that problem.

John
Red=highly doubt it is the problem
Green= Unsure and could be the problem, could anyone elaborate/verify?
 

MrPops

Distinguished
Aug 8, 2006
7
0
18,510
0
I'm currently having the same problem with my new build.

I have the following:

Gigabyte GA-965P-DS3

SAPPHIRE 100149 Radeon X1900XT 512MB GDDR3 PCI Express x16

Intel Core 2 Duo E6300 Conroe 1066MHz FSB LGA 775 Processor

G.SKILL 1GB 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400)

ARCTIC COOLING Freezer 7 Pro CPU Cooler

COOLER MASTER eXtreme Power 713001180 ATX12V 600W Power Supply +3.3@28A,+5V@30A,+12V1@18A,+12V2@18A,-12V@0.8A,-5@0.8A,+5VSB@2A

My computer reboots when I'm gaming as well. It's very random. I'm also starting to think it's the PSU.
 

4Aces

Distinguished
Jul 18, 2006
102
0
18,680
0
Download 3DMark06 and run it, if your test locks up, shows artifacts, tearing or a combination of those during "Deep Freeze" test then it's a video memory issue of some sort.
 

oenomel

Distinguished
Jan 23, 2006
155
0
18,680
0
It can be a PSU issue but that doesn't mean it's the 12v rails.

I had an instance where it was either the 3.3 or the 5 (was a couple years ago) that went bad and it was causing all kinds of random stuff including reboots.

Log those voltages!
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
Ok when I recieve my 7900gt I'm going to do a few things.

Get 3Dmark 06 and run all the bechmarks (does it include AA with benchmarking?)

Run Memtest overnight to check for memory errors.

Check the rails of my PSU (Can anyone reccomend a program?)

Instead of using the PSU PCI-e connector, use the included molex to PCI-e and try it out.

I'm expecting a re-repeat of what I had with the x1800XT. We will see...
 

pauldh

Illustrious
I doubt it was the the video card. BTW, not sure if anyone brought this up, but you were looking at the 2D clock speeds not the 3D clock speeds. The card was not underclocked.
 

Slava

Distinguished
Mar 6, 2002
914
0
18,980
0
I'm going to have to assume it's the PSU
Seasonic makes some of the finest PSUs in the world. Chances of this PSU being bad are so very slim I would suggest you look elsewhere.

Is a floppy disk large enough to make a bootable for memtest? And can I make the bootable floppy on one pc and use it on a different one?
The answer to both questions is "yes".

Im still guessing the PSU should be replaced though?
Wait till you try the new video card. I will say again: Your PSU is of the finest brand and it is plenty powerful. Do not jump to conclusions about the PSU.
 

Slava

Distinguished
Mar 6, 2002
914
0
18,980
0
not sure if anyone brought this up, but you were looking at the 2D clock speeds not the 3D clock speeds. The card was not underclocked.
Five stars indeed. Actually, in the era of nVidia FX series many cards would not work right when 2D and 3D clocks were different. Starting a 3D game would have the card switch to the higher 3D clock and often the game would crash or lock up. The solution was to use a tool such as CoolBits/CoolBits2 and set the 2D clock to the same value as the 3D clock. I still do this as a matter of habit.

I wonder, are there any such problems with any makes/models of current video cards?
 

shamoo

Distinguished
Jan 29, 2005
154
0
18,680
0
reseat all the cables. listen for noises when it crashes
i had an issue with my computer listen for the hard drive heads to click to home
it will be like a small thunk as the computer resets that means you might need a psu because power was unexpectedly cut and just because you have ran the hardware for a while doesnt mean its not a failing componet
 

ches111

Distinguished
Feb 4, 2006
1,958
0
19,780
0
It is the Conroe!! It has to be... No AMD would act that way!

Conroe was a paper launch and you got an engineering sample you poor sucker...

AMD forever!!

(Do I get an entry into the Horde ;))

Sorry bad day at work !!

I would not count on it being your PSU.... You might find that it was indeed your video card. I had similar issues with an ATI 9600 PRO that needed to be overvolted by a mere .1v (stock was 1.4 needed to be 1.5 on the mobob) when used with an MSI NEO FISR2 board. It exhibited exactly the same as what you are seeing now.


I found a very discreet post on the MSI boards that clued me into the fix (2 days searching/reading).. Did your card leave a crash dump? The older ATI cards like mine havd a VPU recovery issue that needed that feature disabled and the 8x AGP needed to run at 1X AGP Write Back disabled.

All of this to say that some ATI cards can be difficult to work with. The VPU recovery thing would write a crash file and try to recover the system often resulting in a reboot style recovery (what a pain). My overvolt and cofig changes had the system completely stable again with suprisingly little variance in perfomance.

I would test the 7900 first.
 

holcar

Distinguished
Jun 21, 2006
158
0
18,680
0
7900gt arrived, its an xfx standard one.

I got the overclocking utuility and when I selected auto detect settings it did absolutely nothing?

3d mark 06 is on download, so we will see. Anyone got any suggestions about the auto detect not working?
 

edelkin

Distinguished
Aug 23, 2006
21
0
18,510
0
I can strongly suggest MEM test for such problems. I had similar issues for well over a year while i tried everything in the book to diagnose random crashes to desktop and reboots during gaming. Finally someone talked me into testing my RAM (as i thought it HAD to be a video card error or driver or something). Sure enough, 1 of my 4 sticks of RAM was bad. Took out the bad stick and VIOLA! No more random crashes.
 

ASK THE COMMUNITY

TRENDING THREADS

Latest posts