computer "randomly" freezing..

C

ChingChang

Supreme [H]ardness
Joined
Mar 27, 2003
Messages
6,504
I built some kid a computer, and shipped it out last week. I shipped it in the regular case box. I figured it would be fine, but when he got it, he said his computer freezes sometimes. Once when he was playing World of Warcraft with WMP and AIM open, and another time when he was playing Battlefield 2.

He said it just randomly freezes, just everything on the screen stops and the sound shuts off. I did stability tests for two days, both video and CPU were stable. I really don't know what this could be.

The only thing I can think of right now is overheating video card, but usually you see artifacts and sometimes the sound gets fucked up when that happens.

Hardware is:
MSI K8N Neo4
2GB G Skill RAM (2x1gb)
AMD Ahtlon 64 3700+
WD Raptor 74
EVGA 7800GT
Antec TruePower II 380w PSU
Windows XP Professional
onboard sound

I've built 3 exact setups like this, actually the only thing different about this one is I used G Skill RAM instead of Corsair Value, but I don't think it is the RAM.

I'm going to have him run an nvidia demo while running 3dmark06, that should get his GPU quite toasty. If it happens again I guess it's overheating, but I did a lot of stress testing on this system already, so I don't know what could have happened.
 
PSU rails were within specs when I shipped it out, and it's a $60 quality PSU so I don't think that's the problem.

He went out to dinner so when he gets back he'll run 3dmark and that nvidia demo.
 
Yea, it does sound like the PS. If it just does it during major 3d calcs like games then the video card probably isn't getting enough power. The new line of video cards need more then 400W unfortunatly.
 
wattage isn't what matters most. I don't think it would even go over 300w under full load.

maybe it's a bad video driver or something...
 
ok, problem seems to be getting worse.

He says it froze like twice in 15 minutes, then another time while I was talking to him about it, and once last night.
Problem occurs when he is playing WoW, and he said once during BF2.

Voltage rails all within 5%, aside from +12v which is 12.65 according to software monitor.
Temps are fine.. he didn't get a chance to run 3dmark06 yet though.


But I had him send me the event logs, so I'm just reading through the System log, and I see a ton of disk warnings, which were event ID 51
An error was detected on device \Device\Harddisk0\D during a paging operation.
http://support.microsoft.com/kb/244780/en-us

I don't have time right now to find out exactly what this means, what the error is. I'll look into it more when I'm not busy

Anyone ever get this? think it might be a bad hard drive or RAM? any ideas would help lots.
 
it is ram. i am 100% positive. i have been having this issue for some time now. after months of wondering and numerous diags one of my 512 ram sticks are only showing as 256.

when it locks, ill eb playing a game then all a sudden the video stops and the sound skips like a cd then stops completely. after about 5 minutes of waiting it all goes back to normal.

i finally had the extra money to get some new ram and i have some ocz 3200 platinum coming in the mail tomorrow.

put a stick of known good ram in there and let him try it out for awhile. if it still locks up then look into other options.

but ram is what was giving me the same problem.
 
Do you have the same Warning in the Event Viewer? if not it might be different.

I'll ask him to leave the comp on after it freezes, to see if it will ever unfreeze.


I'm hoping RAM is the problem, cause that is very easy to replace...
 
truthfully man. i have been working in teh computer repair field for years and i have never looked at the event viewer. lol. just another useless thing microsoft decided to put in their crappy os.

i was just playing around with stuff and i may have fixed my locking up problem. i jsut downclocked my videocard back to stock and played ut2004 for a while and all seemed ok. but my ram is still dying. lol.

here is a neat idea to see if his ram is bad. run prime95 on teh blend test and play a memory intensive game. if it locks then chances are its the memory.

but the more i think about it, i have seen alot of dead hard drives and motherboards cause lockups. but in his case it sounds like ram to me. dont hold me to it since its hard to diagnose a computer when its not infront of me.
 
I'll have him run MemTest86 overnight, that should let us know if RAM is the problem.

Before I shipped it though, I ran MemTest86 for around 5+ hours, and no errors.
 
run memtest, I bet is it that cheap ram, did you set to SPD in the BIOS or tweek it yourself ? the 380 should be okay but that is a lot of amps you are drawing. If his case and room temp are good then it is okay, if he runs his room at 80 plus F then the power drops quickly. but my first hunch is the DDR just like the other post said. good luck. Corsair is good stuff, stick with it.
 
It's not a heat issue, the RAM might be the problem but G Skill isn't bad. Maybe I should have stuck with Corsair though. The RAM is set to SPD.


He's running memtest86 right now, no errors with one pass. He'll keep it on overnight and during the day tomorrow.

He could not download 3dmark06 because the computer would either freeze or BSOD before it finished.

oh yeah, he got two BSOD's the past two days. He said there was a memory dump but he didn't write down the error. I'm checking the event log again, saw the two System Errors which were the BSODs, so I'm just looking through the 5 microsoft pages it linked to that sound like it could have been the error he got. Not sure if I'll get anything from that but it's worth a shot.

could this be something with a virus or some driver issue if it's not the RAM? I don't think I can get him to test it out on a new hard drive, or new video card or anything, since he does not have extra hardware on hand and he does not have the experience.
 
A.D.FIFTY said:
truthfully man. i have been working in teh computer repair field for years and i have never looked at the event viewer. lol. just another useless thing microsoft decided to put in their crappy os.
Click to expand...
Sorry, I have to reply to this one. The event viewer isn't useless, it can help you pin down an error or a Blue Screen if you didn't see it happen. Not all erorrs are reported to the user, such as the Disk Read errors here. If they are continuous on the logs, then you can make a pretty informed decision about the problem, instead of taking an informed guess.

@OP: The disk read errors come from a few things, normally hard drive, IDE/SATA controllers, but also RAM. What drive set up is in the machine? Like how many drives, SATA or IDE, what optical drives are attached, are they all on the same channel? Its indicating drive D:, so whether thats a second hard drive or partition, or maybe the optical drive struggling?

See if he can get some tests going on the hard drive. Scan disks, defrags, benchmarks (Sandra 2004 has a couple) will test it out. Get him to do a memtest as well to check the RAM out, if he can.

Are there any BSODs appearing at any stage (like in the desktop, surfing the net, in the games)? If there are, check out what error code they are (if he can do that, they are in the managment console -> Event Viewer -> System). As a rough guide, if there are different error codes, you have a RAM or mobo issue. If they are the same errors, it may still be a RAM or mobo issue but it could also be related to the hard drive, depending on the error.

Hope this helps out a bit :)
 
Ok, over 10 hours with memtest86 and 0 errors, so I guess the RAM is not the problem (even if 10 hours isn't enough to fully test it, I'd think at least one error would show with a problem that occurs as often as this systems problem)


4keatimj:
Thanks for the helpful response :) I'll have him run scandisk later today... I didn't even think of that.

His setup is this:
Drive A - awesome floppy
Drive C - WD 74gb Raptor (SATA)
Drive D - liteon DVD Burner (IDE, master primary channel)

The hard drive only has one partition.. I just realized that error code refers to drive D as you pointed out right? That seems strange.. I wonder what could be wrong with the DVD burner that is causing lockups??
Unless for some reason his hard drive was assigned the letter D. I don't know...
I'll ask him to try unplugging the DVD burner if that is drive D.


There have been two BSOD's so far, and they are shown in the Event Log as a System Error (category 102, and event ID 1003). Not much info on what it is, aside from "Error code 1000008e", going to look into this more today. He didn't write down the error, all he said was there was a memory dump with a blue screen. Both of these BSOD's occurred while he was doing something in Windows. I'm pretty sure he was not running a game when it happened.

The disk errors are still occurring when he plays games, they might have happened in Windows but I'm pretty sure its only been happening while gaming. When it happens it just spams the System Log with these errors. Here's a screen shot of what it looks like:
http://www.systekonline.com/users/eric/randomcrap/diskerrors.JPG
also, the system did not unfreeze after he waited 10-15 minutes after the error occurred
 
The 8e BSOD is a KERNAL_MODE_EXCEPTION error. Basically something happened that the kernal didn't like, or could not handle. Funny thing is, this can be caused by drive problems. Only yeaterday, I fixed a guys machine that was really slow, and was coming up with those errors. I swapped out the hard drive and put Windows on it, and it was sweet.

Get him to unplug the DVD burner. He is probably running those games (assuming they are legal copies) with a CD-ROM or DVD, and if the drive can't read them, Windows is going to stop until it can read the data. Windows' thought processes go something like:

START
READ data
IF CAN'T READ data
GOTO START

You can see the predicament here. Actually, before he unplugs the drive, get him to put another CD, say and Audio CD in, and play it. If it freezes, sounds bad, or the error logs keep coming up saying it can't read device (or bad block errors), then the DVD would be the culprit. Get him to unplug the DVD, as you suggested, and then test it out.

As a final test, make sure that all the latest drivers for everything are installed. Graphics, sound, motherboard inf files, everything. It seems unlikely, but it is possible that a driver issue may cause this to happen. Give those things a try, hopefully you can get a solution out of it :)
 
hmm

yeah the games he was playing are legal, but WOW does not require a CD to run, and I think BF2 only reads off the CD when the game is started

maybe he had something else in there that was randomly spinning up while he was playing the games

I haven't been able to get in touch with him today, but when I do I'll have him check out the DVD burner and then run scandisk.
 
Did a little more research, seems to indicate that its a paging problem, which indicates that the hard drive if drive D:? This doesn't seem right, I'm still guessing that its a generic error about a drive issue. I forgot that WOW doesn't need a CD to play. If the drive is faulty, it may be sending signals that let Windows see that its having problems. I'd have to say that its still the DVD causing the dramas.
 
have u ran any anti-spyware and anti-virus tests? or try if u have xp have em' push the winlogo key on the keyboard and pause and go in to delete the pagefile and recreate one..
 
I really don't think there are viruses because I installed AVG right before I shipped it, and the problems started occurring the day or day after he got it. Virus infection is possible though. I actually read about a HaxDoor virus that would give the error code in the BSOD he got.

I talked to him earlier today... he got 3dmark06 to run without problems. He also said his DVD Burner has been working fine, because he had been ripping and burning CD's with no problems. And there were no CD's in the drive when the crashes happened. I told him to disconnect the drive and see what happens...

I told him to run scandisk and see how that goes, but I haven't been able to get in touch with him since. Tomorrow I'll see what the deal is.. and if he can't fix it at all he will be shipping it back to me, and I will test parts individually.
 
I am having many similar problems to what you have and I have the same motherboard. I ran memtest for 4 and half hours today on my ocz ram and it found nothing. I am runing prime95 now so I'll see how that does. I am getting the complete lock ups like you said.

I might end up just rmaing it. I already have a rma setup so I have 14 days until it expires.
http://hardforum.com/showthread.php?t=1010532
 
little update...

he shipped me the computer. It arrived fine, no damage at all to the box. I don't think it got dropped or anything.

so last night and today I've just been running some tests trying to get it to crash/freeze again, but haven't had much luck. It came close to freezing... I was running a nvidia demo (luna) while copying a 5gb file 3 times. Luna froze as well as the copying files animation, but I was able to move the mouse around. After like 10 seconds it unfroze and has not frozen since. In the Event Viewer there was a W32Time error, saying something about none of the time sources being available? I can post the exact error later if needed. It's not on this comp though.

anyway, last night I ran 3dMark06 and got a score of 3500 or something. I ran it again, but this time I had Luna running in the background, and I was also making 3 or 4 copies of that 5gb file, so the GPU was being heavily stressed..

No problems with that. Then I ran scandisk- no problems. I then downloaded DLG Diagnostics for WD drives, and ran the quick test. No problems. I then ran the thorough test and it went through with no problems either.

Overnight I ran the Sandra stress test thing, and it went through like 18 times or something with no problems. All of the tests were selected, aside from floppy and CD/DVD Rom because I did not have the specific CD and do not know how to make one.

So this morning I think it was around 7 or 8am, I downloaded IOMeter to further test the hard drive.. I don't know if I really set it up correctly but this is what I did. In the "Test Setup" tab I have the run time set to 24 hours.. not sure exactly what that will do though. Will it just stop after 24 hours?
And.. in the Access Specifications tab (I guess this is where the different tests are selected) I selected the "Default" test, and also added the "All In One" test. It has been running since around 8am I think, and it is still on the Default test. There have been 0 errors so far.

Also, I've been running Prime95 while I'm running IOMeter, because I saw IOMeter was not utilizing much CPU at all. Figured I would double check the CPU while checking the hard drive at the same time... should I leave Prime95 running?


I'm just wondering what other tests I should run.. I'm running out of ideas :(
 
ChingChang said:
...
So this morning I think it was around 7 or 8am, I downloaded IOMeter to further test the hard drive.. I don't know if I really set it up correctly but this is what I did. In the "Test Setup" tab I have the run time set to 24 hours.. not sure exactly what that will do though. Will it just stop after 24 hours?
And.. in the Access Specifications tab (I guess this is where the different tests are selected) I selected the "Default" test, and also added the "All In One" test. It has been running since around 8am I think, and it is still on the Default test. There have been 0 errors so far.

Also, I've been running Prime95 while I'm running IOMeter, because I saw IOMeter was not utilizing much CPU at all. Figured I would double check the CPU while checking the hard drive at the same time... should I leave Prime95 running?
Click to expand...
So these programs have both been running since yesterday morning. Last night before I went to sleep I stopped the "Default" test in IOMeter, so the "All In One" test could get a chance to run. Prime95 has been running fine with 0 errors, and IOMeter has had 0 errors. When I woke up today I just opened Luna and left the other programs running, so the GPU would be stressed a bit.

The only errors in the Event Viewer that I saw are from W32Time, which occurs when it tries to automatically update the time. I don't know why it doesn't work. But I changed the date to when it should update the time, and when it tried to, the thing froze for a couple seconds then went back to normal. I tried both time servers, they didn't work. And when I tried it a few other times, I got the W32Time error, but Luna didn't freeze or anything.


Does anyone know what it could be? My only guess is the way time is updated, but that doesn't really explain the Disk Errors that have happened before. I can just disable automatic time updating, but I don't even know if that's much of a problem.

Any ideas on other ways to stress the system?
 
Well have you tried running a game you know like he said it froze while he was playing a game and see if you see anything he missed or looked over. Maybe you can diagnose the problem from there. you said it hasent frozen since youv'e gotten it well maybe its the games he plays. What are the system specs, what os is it running on.
 
Yeah I'm going to try playing WoW for a bit later tonight or tomorrow. He just gave me his password so I'll just mess around on his account when he's not playing.

I don't have BF2 with me, but I'll give him my copy of Quake 4 and see if that crashes or anything.

Here's the system specs:
ChingChang said:
Hardware is:
MSI K8N Neo4
2GB G Skill RAM (2x1gb)
AMD Ahtlon 64 3700+
WD Raptor 74
EVGA 7800GT
Antec TruePower II 380w PSU
Windows XP Professional
onboard sound
Click to expand...
 
How much ram is self embeded. Have you tried removing the ram one stick at a time.
 
Alright then that rules out a problem but lets try those games. The games he plays there like store bought and not like Ex: friend burned a game for him right. Now what are those games names again.How big is the video card.
 
Yeah he bought the games, since they require valid CD keys to play online. Games he said froze are World of Warcraft, and Battlefield 2. Video card is a 7800GT, 256mb RAM.
 
Wait Isn't World Of Warcraft an online game. It could be that the game itself froze not the computer, i dont know about the other game. I still think its the ram but you tested it and no errors. I dont know what else to do iv'e never come across a problem like this, except if its the Ram or Video Card. I'll try to look up some information about it.
 
Make and model? I built it and posted the hardware in the first post :)

I ran WoW for like an hour with Luna running, IOMeter, and Prime95 (testing RAM), and no crashes/freezes. I just ran around the world basically. I'm going to try playing quake4 sometime today or maybe I'll install Counter Strike Source or something.
 
Well i dont know, is this kid sure that they crashed and not just froze i mean all computers freeze. I dont know iv'e had a similiar problem like this but after testing the ram i got the awnser. The ram stick partialy came loose during shipping. You've tested everything and nothing came up. Id say that the computer is in fine working order and maybe this kid is jerking your stick, do you know this kid personally or just met over internet. Like i said I Dont know what else to do.
 
Yeah I know him personally. I don't think he'd make this up. Especially when some of the Disk Errors match up with when he said it froze.

CS:S just finished downloading, so I'll just try that out for awhile. If it doesn't freeze or crash I guess it's a problem with WoW or something. Maybe there are other WoW players who have this problem on a certain hardware configuration...
 
ok, I played some CS:S for a few hours. First thing I noticed was that it was getting very low framerates, even with the graphics settings turned down. But I think that is because CS wasn't 100% downloaded.

After about an hour playing an HDR map, everything was fine, but it switched to another map and then I got a BSOD (IRQL_NOT_LESS_OR_EQUAL). Restarted and found nothing in the Event Viewer. Then played more CS:S for like 1-2 more hours and no problems.

that BSOD is usually driver issues correct? I could try reformatting, but I still wouldn't know what the problem was or how I can reproduce it.

oh, and I think the voltage rails are fine... as I said before it says 12.65v in SpeedFan, but in the BIOS it is around 12v. But then again, maybe the BIOS is wrong or there just isn't enough of a load on the 12v rail in the BIOS. One of my friends said he'd bring me a digitial multimeter from home this weekend, so I'll see if he has that on monday.
 
Well The error IRQL_NOT_LESS_OR_EQUAL is generated by both software and hardware issues. The message indicates that a kernel mode process or a driver attempted to access a memory address that it did not have access to. And yes BSOD usually means a driver issue.
 
You must log in or register to reply here.
Back
Top