Sign in with
Sign up | Sign in
Your question

"Bad" memory cause computer freeze/lockup?

Tags:
  • Memory
Last response: in Memory
Share
April 30, 2007 5:47:35 AM

Ok guys, about 3 months ago i put together a C2d system.

Asus p5ne-sli
Ultra 600w PSU
E6600
8800gts
Cooler Master case/etc etc.

I had really bad luck with my initial batch of memory, which was corsair 4-4-4-12 DDR2-800 mem that was DOA. So rather than wait 3-4 weeks for a rep, i sent it back for a refund, and bought some Supertalent 5-5-5-15 DDR2-800 mem from a local shop. I'm not entirely sure if the mem is the problem. But ive had nothing but constant problems, i even had the comp freeze during the win install. Since then, i've done a fresh install on a new HDD (A roomie was welding in the garage and hit the breaker on my area of the house while i was in the middle of a defrag, pretty much hosed that hdd/copy of windows).

Basically its completely random, the comp will freeze, and if i have sound playing, it will loop back on itself. Sometimes it will freeze, ill restart, and then it will freeze 5 min later, ill restart again, do the same thing, and it will go for 3 hours. Its bizarre.

I've tried it with an Audigy 2zs, and audigy x-fi, and onboard sound. Problem still exists, i've got a fresh as of 2 weeks ago install of windows XP, problem still exists. I've tried underclocking the shit out of the 8800gts, still locks.

The only thing i can come up with is maybe the mobo is being stupid, the mem, or possibly the CPU is not seated properly, which i doubt.

My CPU temp generally hovers around 39-43C, mobo temp from 48-53c, and the vid card from 75-80c (although with declocks ive gotten it down to low 60's under intense load).

I believe the mobo temp is high because the 8800gts is like 1 inch from the chipset "chip"

Either way, if anyone might be able to shed some light it would be awesome. I'm ready to kick this thing out of the window its driving me nuts. I've been building my own comps since my first 486sx25 when i was a kid, and i've NEVER had this many issues with a comp, not even close.

edit: i should probably also clarify it happens predominantly in games, though it will also freeze when sitting utterly idle on the desktop, though that seems to be a hell of a lot less frequent.

Also, the only game ive been able to somewhat "replicate" the problem is in WOW. There was a certain AH chick in Stormwind that if i went to her, i was guarunteed a lock within 5 minutes, but thats the only situations ive found that i could "force" the lock so to speak.

More about : bad memory computer freeze lockup

April 30, 2007 5:52:22 AM

You definitely need to boost your case ventilation when your sys temp is 10C hotter than your CPU temp!
April 30, 2007 5:30:34 PM

Nothing else i can do, i already have a 120mm intake and 2x 80mm exhaust fans in the case. The only other thing i can think of is the top of the case has some firewire/USB ports, that part is removable and i can place an 80mm intake or exhaust fan. Though im not sure how much that would help.

Seriously i think the only reason its so hot is because that 8800gts is sitting less than an inch from the heatsinks for the mobo.
Related resources
April 30, 2007 6:35:51 PM

You should try and run memtest86+ - this will check if your memory has problems. At least this way you can rule it in or out.
Not sure how periodic your problem is, but you could let memtest run for at least a couple of passes...
April 30, 2007 7:38:04 PM

Quote:
..., i already have a 120mm intake and 2x 80mm exhaust fans in the case.

You may have a mismatch there. Try unplugging the intake fan. You might also consider getting more powerful 80mm exhaust fans. Any chance of posting a photo or two to show us how things are set up inside your case?

Quote:
...y i think the only reason its so hot is because that 8800gts is sitting less than an inch from the heatsinks for the mobo.

If you had proper air circulation inside the case, that wouldn't be a problem.
You might also consider going to active cooling (fans) for your northbridge and southbridge.

Also, are you by any chance using an aftermarket CPU cooler? If so, it probably isn't blowing air down onto the MB and out onto surrounding components (like the Northbridge heat sink) as the OEM Intel cooler would. You might try using the Intel cooler.
April 30, 2007 8:14:34 PM

I second running Memtest. It's a good place to start. Although just because you fail the test doesn't necessarily make it your RAM at fault. It could be the socket the DIMM is put in, or the motherboard itself. First IF it fails, then you need to isolate which DIMM is failing (assuming you have more than one), by removing the DIMM's one at a time tell it passes. Once your this far, then you have to put either the apparent good DIMM in to the socket that the apparent bad DIMM came out of or vica versa. This will isolate whether it's the socket or the DIMM. If these don't narrow it down (if neither DIMM appears to work) then you can start to suspect motherboard.

Also you may want to check whether there is a newer version of BIOS than the one you have. Newer BIOS rev's often fix memory incompatibilities. You could also try manual adjusting your memory timings to loosen them up a little and see if that helps.

Let us know how you make out.
April 30, 2007 8:26:01 PM

You can run Memtest to cheack the RAM...but given this:
Quote:
Ultra 600w PSU

I would say it was the PSU at fualt.

That is the worst brand I have ever seen.
I bought a case of them and the first out of the box was DOA and the next died in under 24 hours of burn in.
I sent the whole case load back for a refund and will never use them again.

You have top end parts for the rest of your build and should have spent the extra money for a real PSU befor buying anything else.
When that PSU dies it will likely take alot of things with it.

Falcon Northest Computers use only Silverstone PSU's and they would be my 2nd pick,but I perfer to use PC Power & Cooling as nothing comes even close to them.

If you call PC P&C and talk to a Tech he will ship you a 750 (825 surge) Watt unit rated at full load @ 40C rather than other companys 25C for only $169.
April 30, 2007 8:49:12 PM

While I agree he should upgrade to a better quality PS, his problems don't seem like classic PS problems. It would be nice to know what model Ultra he's got -- their "higher-end" models are just average, not terrible like their V-series.
April 30, 2007 8:55:22 PM

Quote:
Asus p5ne-sli
Ultra 600w PSU
E6600
8800gts


How many banks are you using to fill in the RAM?
I have a machine very similar to that, with an ASUS P5N32-SLI Premium. I filled 2 of the banks with 1 GB DIMMS, and my machine worked great.

I added another 2 DIMMS to fill all of the banks, and the machine started to run like crap. As I would use the computer it would slow down over time. Rebooting would always fix things. Occasionally I would get the BSOD.

As I have learned from reading other peoples accounts on the Internet, these ASUS motherboards have problems when you fill all 4 memory banks. I took out 2 of them, and sure enough my machine works like a dream again.

Good Luck
April 30, 2007 11:16:29 PM

Ok, i'm at work atm, so i cant test anything, but i will run memtest when i get home tonight. But i can answer a few questions:

Quote:
"You can buy a new case..."


Its a coolermaster praetorian 730, its a good case, a little cramped but i dont feel that a new case is an option here.

Quote:
"You may have a mismatch there. Try unplugging the intake fan. You might also consider getting more powerful 80mm exhaust fans. Any chance of posting a photo or two to show us how things are set up inside your case? "


I might hafta do that. Supposedly the fans are flow matched in that the 120mm flows equally to the 2 80's. We'll see. I'll try to get a photo up tonight.

Quote:
"If you had proper air circulation inside the case, that wouldn't be a problem.
You might also consider going to active cooling (fans) for your northbridge and southbridge."


I agree, although im not sure about the proper cooling part, i would definitely like to go active cooling on the north and southbridge though im not entirely sure how to go about that.

Quote:
"Also, are you by any chance using an aftermarket CPU cooler? If so, it probably isn't blowing air down onto the MB and out onto surrounding components (like the Northbridge heat sink) as the OEM Intel cooler would. You might try using the Intel cooler."


Stock C2D cooler, I bought a Tuniq Tower 120, but it was too tall for the case.

Quote:
"Also you may want to check whether there is a newer version of BIOS than the one you have. Newer BIOS rev's often fix memory incompatibilities. You could also try manual adjusting your memory timings to loosen them up a little and see if that helps."


Updated bios to latest revision earlier last week, also got the newest mobo/vid/soundcard drivers.

Quote:
"I would say it was the PSU at fualt.

That is the worst brand I have ever seen.
I bought a case of them and the first out of the box was DOA and the next died in under 24 hours of burn in.
I sent the whole case load back for a refund and will never use them again."


I was wondering the same thing. I actually carried this PSU over from an older 7800gt/athlon 3700+ build i had. Its been "working" fine although the fans are going out. I will be replacing the PSU very soon. I agree that it was not the best, but at the time it was all i could afford in terms of getting a 550w+ PSU that was dual rail. I always used antec PSU's before, but at the time i was building the system, they were having all kinds of wack issues with Asus mobos. I'm a big Asus fan so i wont change there.

Quote:
"While I agree he should upgrade to a better quality PS, his problems don't seem like classic PS problems. It would be nice to know what model Ultra he's got -- their "higher-end" models are just average, not terrible like their V-series."


I'll see if i can find a link:

http://www.zipzoomfly.com/jsp/ProductDetail.jsp?Product...

Quote:
How many banks are you using to fill in the RAM?


2 banks, 1gb module in each slot

One other question i was wondering about, maybe someone will know.

Since i have an SLI motherboard, do i have to use the top slot if i'm doing only a single card? I suspect the answer is Yes, but if i can use the bottom slot it would help the cooling situation.

I was also considering getting a slot cooler to throw under the 8800gts, although im not sure how much it would help.

I would also like to add i have 3sata HDD's active and mounted, which could be causing part of the "flow" problems. Either way, thought id list it all to give as much info as possible.

edit2: opinions on this PSU?

Enermax ELT620AWT
http://www.zipzoomfly.com/jsp/ProductDetail.jsp?Product...
May 1, 2007 11:13:02 PM

Bump...

Nobody? All you guys that posted before, any more ideas after reading my response's to your questions?
May 2, 2007 5:41:17 AM

Did you run memtest like everyone suggested?
May 2, 2007 6:42:07 AM

Not yet, apparently it can only be run from a bootable cd or 3.5". Since i dont have a floppy drive, and i have no CDR's lying around, i need to make a trip to the store to get some. I will do that soon.
May 2, 2007 7:57:04 AM

update: roommate had a CDRW lying around, ran memtest twice, passed both times, no errors.
May 2, 2007 8:44:54 AM

download and run memtest and see what it has to say

in the bios manually set everything to do with ram - eg speed, timings and up the ram vcore to 1.9v
May 2, 2007 9:56:56 AM

Running Memtest like that (twice) doesn't usually show anything. Anytime I have seen interrmittent rebooting locking, the memory doesn't fail immediately. You need to run Memtest overnight or whatever. Just let it run for 8 hours or so and then check. I had a bad DIMM that completed 5 passes before it started to error. Once it did, it gave a consistent error same test same location. Two passes is just not enough to eliminate RAM.
May 2, 2007 12:45:38 PM

It's grounded try puting a anti static or put electric tape on the back and on the stands...remove each memory put it one by one to be sure try to load bios on default....try it
May 2, 2007 4:49:24 PM

R_D - your post is incoherent.
May 2, 2007 5:22:20 PM

Quote:
Running Memtest like that (twice) doesn't usually show anything. Anytime I have seen interrmittent rebooting locking, the memory doesn't fail immediately. You need to run Memtest overnight or whatever. Just let it run for 8 hours or so and then check. I had a bad DIMM that completed 5 passes before it started to error. Once it did, it gave a consistent error same test same location. Two passes is just not enough to eliminate RAM.


Ok, i will have it going all day today at work, should be about 12 hours.

I'm also ordering a new PSU as we speak. If nothing else so i dont have to have my desk vibrating from an oscillating PSU fan.
May 3, 2007 2:46:20 AM

Ok, 9 hours of memtest 86, no errors.
May 3, 2007 5:31:30 AM

Quote:
Ok, 9 hours of memtest 86, no errors.


Well that's pretty much eliminated your RAM as the problem. Best to do the troubleshooting that doesn't cost you anything first.

That's great, now as someone mentioned I would look at your PSU. I have no experience with Ultra PSU's, but I haven't read anything good about them. They seem to be ranked right at the bottom. What would be great is if you knew someone with a spare supply or one that he/she wouldn't mind lending you for a day or so. It sucks to lay down a bunch of money in the attempt to troubleshoot a problem only to find out that's not it.

Your motherboard is also running a little hot, but nVidia chipsets do run hotter than Intel. You could try taking off your side panel and directing a fan into it and see if that makes a difference temperature wise and consequently has an affect on you system stability.

The cooling step maybe the easiest and cheapest step in the process of elimination.

Good luck and let us know how you made out.
May 3, 2007 11:35:06 AM

Quote:
Ok, 9 hours of memtest 86, no errors.


Well that's pretty much eliminated your RAM as the problem. Best to do the troubleshooting that doesn't cost you anything first.

That's great, now as someone mentioned I would look at your PSU. I have no experience with Ultra PSU's, but I haven't read anything good about them. They seem to be ranked right at the bottom. What would be great is if you knew someone with a spare supply or one that he/she wouldn't mind lending you for a day or so. It sucks to lay down a bunch of money in the attempt to troubleshoot a problem only to find out that's not it.

Your motherboard is also running a little hot, but nVidia chipsets do run hotter than Intel. You could try taking off your side panel and directing a fan into it and see if that makes a difference temperature wise and consequently has an affect on you system stability.

The cooling step maybe the easiest and cheapest step in the process of elimination.

Good luck and let us know how you made out.


Aye, i definitely agree, i'm not in any position atm to be throwing money around. But, the PSU has been annoying me for a while anyways. The fans have been going out, and just a couple of days ago have started oscillating so badly that its vibrating my desk. I'm not really comfortable trying to replace the fans, and i never really like the PSU anyways. I bought an Enermax liberty 620w with the cable management system, which should give some minor help to my airflow situation.

As for cooling. I'm not entirely sure its the issue. If heat were the case it should be more consistent in its lockups. It certainly shouldnt run things like 3dmark06 for several passes in a row before locking. Also, to answer your question, i havent tried putting a large fan flowing directly into the side of the case, but i do have a 120mm side case fan that i have blowing almost directly over the mobo chipset heatsinks. Although i'm not sure thats doing anything more than blowing the hot air inside the case around.

Also, approx the time that the fans starting oscillating, the computer started doing random restarts in addition the the lockups. This was also about the time i added a 3rd HDD. The past 3 months the only issue ive had was the freezing, never had it restart itself, it also wasnt nearly as prevalent, i could usually expect 1-3 hours of play before it would freeze. Its to the point now that if i see it go 20 minutes im surprised. This only furthers my suspicions thats its the PSU. I'm just wondering if maybe its getting overloaded and shutting off and/or not being able to provide enough power, which could be causing the vid card to have issues, or any number of things really. I'm not 100% sure really.

If it doesnt end up being the PSU, my only guess is the mobo. I don't think it could be the CPU, especially with the CPU running so cool. I guess it "could" be the vid card, but i dont have another system to test that in, that and i've already done the whole underclocking thing to bring down its power reqs and temperatures, and it didnt seem to help a bit.

Oh vell, we shall see. PSU should be here on friday. I'll keep checking up here to see if anyone else has any more input.

Thanks again for everything guys.
May 3, 2007 11:48:24 AM

Wrong post sorry!!! try cheking the memory on memtest and puting cooling devices, cooler the system the stable it will be Chow!!!
May 3, 2007 3:32:00 PM

Love Enermax supplies. I bought a 465W supply of theirs in February 2002, it's still working great in my boys computer. I replaced it with this Antec in my sig. I can't necessarily say anything bad about the Antec, I've had no problems with it, but I just don't have as much faith in them. My next supply will likely be an Enermax again.

So anyway, you said you bought, as already have or coming? Just interested if you have it and whether or not it's made a difference.
May 4, 2007 11:40:17 PM

Aye, i bought the Liberty 620w i linked in a previous post. Actually about to power down to install it right now. Just wanted to check this thread first. Cya guys in a few hours once i can confirm.
May 5, 2007 1:01:55 AM

Ok, PSU is in, very happy with it. Its ridiculously quiet, so far everything has been fine, but its only been a little while, i'll check back here tommorow to let everyone know what if anything happened.
May 5, 2007 6:31:29 AM

Well, the freezes seem to be gone. Also with the new cable mgmt system, my vid card seems to be substantially cooler. Not entirely sure why. Never the less.

I did get a couple of BSOD's, the winxp equiv anyways, giving a:

IRQL_NOT_LESS_OR_EQUAL

After some research, it looks like it could be A, the north or south bridge being too hot (this would be my guess) or B, the CPU overheating (doubtful as its usually 10-15c lower than northbridge temp).

Can anyone point me to some more "active" cooling solutions for my mobo? Right now there is just a row of heatsinks on the north and southbridge. So i'm not sure.
May 5, 2007 6:47:39 AM

try puting this small fans but before you put this try puting a thermal grease first it will make your chips cooler than before i try it and it seems to work..
May 5, 2007 11:02:01 AM

The rest of the error message may help in determining what the problem is.
May 5, 2007 7:56:53 PM

"A problem has been detected and windows has been shut down to prevent damage to your computer.

IRQL_NOT_LESS_OR_EQUAL

*** STOP: 0x0000000A (Ox0000001D, 0x000000FF, 0x0000001, 0x8054199A)"

The rest of it was text on what might be the problem, etc.
May 5, 2007 7:59:22 PM

Quote:
try puting this small fans but before you put this try puting a thermal grease first it will make your chips cooler than before i try it and it seems to work..


I am going to try that, i just need to figure out how the hell to attach them.

I was also doing some tests last night. I loaded up a game i could safely alt+tab out of. Got both my Vidcard temp and Asusprobe II up to monitor temps.

CPU did great, went from 38 to about 43C after approx an hour. The mobo on the other hand went from about 47 to 58C, which is right about when it crashed. The vid card never broke 78c, which is about 5 degrees cooler than with the old PSU.

So it definitely seems like its the mobo. The default setting for the warning on mobo temp in Asusprobe is 60C, which leads me to believe that may be the problem.
May 5, 2007 7:59:27 PM

The CPU should slow itself down instead of overheating. Northbridge temps higher than the CPU do sound worrisome. I'd probably just try mounting a small fan on top of the NB heatsink; one of the stores with all kinds of stuff like this is xoxide.com.
May 5, 2007 8:00:31 PM

Quote:
The CPU should slow itself down instead of overheating. Northbridge temps higher than the CPU do sound worrisome. I'd probably just try mounting a small fan on top of the NB heatsink; one of the stores with all kinds of stuff like this is xoxide.com.


You rock, heh, you posted this literally at the same second i posted my last message. And thank you so much for the website link.
May 5, 2007 8:21:39 PM

Well, i've found several cooling solutions that say they work for the nforce4 chipset? Ok, found a review of my mobo:

http://www.hothardware.com/articles/Asus_P5NE_SLI_Nvidi...

Asus does leave mounting holes, however, should you want to apply your own third party chipset cooler. The Northbridge cooler also uses standardized mounting holes, so it's possible to remove the large aluminum alloy cooler to apply a different cooler on this chip as well.

So, hopefully i can resolve the problem soon.

Does anyone know of any active cooling solutions for it, i.e. with fans. All of them that i saw on xoxide with the exception of one were fanless.

Oh, also, side note. I turned my 120mm side case fan up to high, and it dropped the mobo and cpu temps by about 4C
May 5, 2007 8:22:39 PM

jesus christ, wtf kind of shit is this:

"Interestingly enough, we found in testing that the board had a tendency to be more stable when using the "black" DDR2 slots on the board rather than the yellow slots, which is what Asus recommends if you're only using 2 x DDR2 modules. In terms of memory support, the board seems a bit quirky at this time, but when modules were placed in the black slots, we had no problem getting them to run at their fastest 4-4-4-12 timings and retaining full stability."

Now i get to pull my mem, everyone hope this solves my problem lol.
May 8, 2007 5:21:51 PM

Well, after a couple days of playing and using the comp, it hasnt frozen once. Had a couple times where the comp just restarted. But i've been reading up and apparently the X-fi has a tendency to do that in the game i play. But, seems to be running ok.

Wanted to say thanks for everyone's help.
May 8, 2007 8:14:38 PM

Glad to hear your system is now stable -- "random" instability can be very frustrating!
February 6, 2009 2:56:31 PM

I CAN SAY THAT AFTER READING ALL THE POSTS. I SEE EVERYONE LEANING TOWARDS RAM-MEMORY AND OR CPU TEMP. WELL I BELIEVE YOU ARE ALL WRONG. 10% OF THE TIME IT MIGHT BE A MEMORY ISSUE. BUT FROM WHAT I KNOW FROM EXPERIENCE. THE PROBLEM FOR SYSTEM FREEZE IS RGISTRY ISSUES. JUST DOWNLOAD A FREE VERSION OG CCLEANER AND ALL PROBLEMS ARE SOLVED. CLICK WEBSITE BELOW. THANKS
http://www.ccleaner.com/
February 12, 2009 9:58:03 PM

"Exhausting" the warm air is always better in the long run than trying to cool using extra "intake" air...just my .02cents :) 
April 10, 2011 8:35:07 AM

Hello everybody! I unfortunately am having the same issue as the original poster and would like to reactivate the thread to solve my problem if anyone can help.

It was surprising to find out that I have the same exact motherboard as the original poster.. a P5N- E SLI ASUS board. Now lol I've heard of the problem with using all 4 memory slots and I believe I had this same problem before I upgraded to two sticks with equivalent capacity (4gbs) I previously ran 4 sticks of 1gb memory. I recently wanted to upgrade to 8gbs because I'm now running windows 7 64-bit so I'm back to 2gbs x 4 slots. Annnndd I'm getting random freezes yet again. I went through all the trouble blaming it on a new video card I recently purchased, but then I remembered my previous problems with the memory on this board.

Now does anyone know if the bios for this board was upgraded to satisfy this problem? Or should I stick to 4gbs and pull out the 2 others.

BTW the yellow slots seem to work fine, its the black ones I'm having trouble with hehehe.
April 10, 2011 8:51:17 AM

BTW I'm running Bios version 1301 and since there hasn't been any details on memory with updates after that I assume I should just go back to using two slots.
April 11, 2011 10:56:22 AM

Mrpan, you should just open a new thread.
May 5, 2012 9:32:58 PM

techgeek said:
Quote:
Ok, 9 hours of memtest 86, no errors.


Well that's pretty much eliminated your RAM as the problem. Best to do the troubleshooting that doesn't cost you anything first.

That's great, now as someone mentioned I would look at your PSU. I have no experience with Ultra PSU's, but I haven't read anything good about them. They seem to be ranked right at the bottom. What would be great is if you knew someone with a spare supply or one that he/she wouldn't mind lending you for a day or so. It sucks to lay down a bunch of money in the attempt to troubleshoot a problem only to find out that's not it.

Your motherboard is also running a little hot, but nVidia chipsets do run hotter than Intel. You could try taking off your side panel and directing a fan into it and see if that makes a difference temperature wise and consequently has an affect on you system stability.

The cooling step maybe the easiest and cheapest step in the process of elimination.

Good luck and let us know how you made out.

I don't agree with the eliminating the ram I would check ram voltage against recommended by vendor, then go into bios check your voltage if it is on auto put it on the recommended voltage, memtest is not accurate tool. also switch around your ram sticks
chech timng should be around 3:10 ratio good luck
!