Tech Support banner

Status
Not open for further replies.
1 - 20 of 21 Posts

·
Registered
Joined
·
11 Posts
Discussion Starter #1
For the past month, my computer has been experiencing random crashes (illegal operation, etc), BSODs and the occasional freeze requiring a hard reboot while playing games. I have tried just about everything I can think of to remedy this problem and nothing seems to help. I've updated drivers for everything in my system that can be updated, run anti-virus scans, defragged, scandisked, scanned for spyware etc, all to no avail. It's starting to get a little ridiculous.

The most unusual aspect of this problem is when the crashes/freezes/bsods occur. Sometimes they don't happen for an entire day. Other times they occur every few minutes. Most often once a particular game has crashed once or twice, it runs stable for the rest of the evening with no hitches. Switching to a new game may or may not result in a fresh round of crashes. This seems to also be linked with how long the computer has been powered up. After a BSOD or freeze, when the computer is rebooted it's much more likely that another issue will appear within a few minutes of launching a game. If I'm lucky and all that happens is a crash, I can usually restart that game and play for hours without a problem, only to have the same thing occur the next time I reboot or switch to something else.

I can't recall installing or changing any software or drivers before this began happening -- it just seemed to come completely out of nowhere. I've since tried mucking around with drivers and settings since then, however, but nothing helps.

System specs:

Windows XP Home SP2 w/ all crit updates
3.4GHz P4
MSI 865PE Neo-2 Mainboard
Onboard sound (CMedia AC97)
Onboard LAN
NVidia GeForce 6800 GT/256MB w/ ForceWare v78.01 (latest)
1 GB Corsair RAM (Value Select IIRC)
120GB Maxtor HDD
Sony DVD-R (Don't know the model number offhand, XP reports it as DW-D22A)

Software (ok, games =P) in which the problem has been seen:

World of Warcraft (Usually crashes once and then is perfectly stable)
Quake 3: Arena (Has only crashed, never BSOD'd or froze)
VisualBoy Advance CE (Has only crashed)
The Sims 2 (USUALLY causes system freeze, tends to work (perfectly) after computer has been running for several hours)

Any help or insight would be greatly appreciated.
 

·
Registered
Joined
·
25,111 Posts
Hi,

Before tackling this problem, I would like to see you talk about the brand name, watts, and listed amps on each rail for your power supply.

Also, would love to see both at-rest and under-stress temperatures for your unit.

The last thing before we get started would be to know what the voltage outputs are on each rail when the computer is running.

That information should give us a better foundation to find what the problem might be.
 

·
Registered
Joined
·
11 Posts
Discussion Starter #4
Tumbleweed36 said:
Hi,
Before tackling this problem, I would like to see you talk about the brand name, watts, and listed amps on each rail for your power supply.

Also, would love to see both at-rest and under-stress temperatures for your unit.

The last thing before we get started would be to know what the voltage outputs are on each rail when the computer is running.

That information should give us a better foundation to find what the problem might be.
It's interesting that you are both inclined to suspect the power supply since it's technically below spec for my video card. When I upgraded to my 6800 I had no idea the requirements included a 350W PSU. (I hadn't previously seen ANY computer part that required a particular wattage in fact) I ordered one to replace the 300W one I have and while waiting decided to try it out anyway and was surprised to find there were no problems whatsoever, which turned out for the best since the one I ordered didn't work, and the replacement they ended up sending me for the broken one they sent was itself broken. :upset: I've been running the system with the lower-watt PSU ever since and have never had a hiccup until about a month ago.

Anyhow, I'm not sure what you mean by the term "rails", but I'll list off what my PSU says on the label and hopefully that will give you what you're looking for:

Enermax EG301P-VE

AC Input:

115V/230V-, 50Hz-60Hz, 8A/5A

DC Output:

+3.3V: 28A
+5V: 30A
+12V: 22A
-5V: 1A
-12V: 1A
+5VSb: 2.2A
+3.3V & 5V: 170W
Total Power: 300W

For the temp and voltages I'm using MBM5. Again I don't know if these voltages are exactly what you're looking for, let me know if you need more and what I can use to find it.

Idle: CPU 63C/Case 36C
My GPU reads temps as 73C core/39C ambient

Load: CPU 83C/Case 40C
GPU: 94C/48C
(Load in this instace is World of Warcraft running windowed 1024x768 while standing in a highly populated area with lots of activity.)

Voltages:

Core 0: 1.31V
Core 1: 1.34V
+3.3: 3.38V
+5.00: 4.97V
+12.00: 11.86V
-12.00: -12.52V
-5.00: -5.34V

Regarding system temps: This is something I've been battling in my systems for years, I assure you I don't overclock and I've tried about a dozen different case variations, including custom heatsinks and fans to get the temps down. I know the temps above are high, I just don't know how high because despite hours of searching I can't seem to find a website that will give me exact numbers on what my temps should look like for my CPU. Keep in mind I've been running with this configuration for over a year, and it would surprise me if the heat has only now become an issue.

Anyway, if there's anything else you need to know, just ask. If there's anything you think I could do that might help, name it. Good luck and thanks again for the assistance. :sayyes:
 

·
Registered
Joined
·
11 Posts
Discussion Starter #5
keymaker said:
hey hotwax,

i second the power supply just like tumbleweed said. what brand and how much power are you running?

do you have a spare power supply to test on your computer?
Regarding the spare power supply: In a pinch I could take the one out of my GF's computer, which is also 300W and I believe also an Enermax. If you think the problem is just low wattage then obviously that won't help, but if you're reasonably sure that's the culprit I'd have no objections to buying one (assuming I can actually get one that works this time around...)
 

·
Registered
Joined
·
11 Posts
Discussion Starter #7 (Edited)
dai said:
a fan low on the front of the comp drawing air in and one higher at the rear exhausting the hot air,will give you a cool flow across the comp.
you need to go for a 500w+ supply to allow for futher upgrades
My concern with this setup in my PC is that a vacuum would be created above the CPU and it actually seems to increase heat when I mount the rear fan to blow air outward (I understand this is the normal way it's done). In my case I have a single front fan at the base of the case sucking air in, a second fan on the left side (immediately next to my vid card) also blowing inward, and the rear fan (originally mounted to suck the air out) pulling yet more air in. (The only fan exhausting air is the exterior PSU fan) The location of my CPU heatsink and the interior fan on the underside of my PSU, both pulling air away from that rear fan, worries me that a vacuum would be created if all three fans pulled air out, overworking the fans and getting less cooling accomplished. To confirm this, I tested the rear fan in both directions and inward actually seems to produce slightly lower temps.

I've considered a suggestion from a friend to get a new case that has only two 120mm fans -- one low in the front pulling air in and one high in the rear sucking air out, but I'm worried the same situation would just occur. What would you guys suggest in that dept?

The Arctic Silver is a good idea and one I've tried before with other CPUs. Right now I have the stock heatsink on the CPU which I'm sure could use improvement. The linked page will actually come in really handy because I'm sure the last time I applied Arctic Silver I didn't do it that well. =P

As for the PSU, I'm starting to agree that it's a likely source of the problem and wonder what you guys would recommend brand-wise. I don't have an infinite amount of money to throw around but power supplies aren't typically that expensive so there's some leeway there.

Edit: Suggestions of good heatsinks/fans for the CPU are also welcome if you feel this is needed. Frankly I'd rather have a brand new heatsink to apply the
Arctic Silver to rather than trying to clean the gunk off my current one.
 

·
Registered
Joined
·
25,111 Posts
Hi,

All good information coming your way for the previous posters. I wanted to emphasize that the rear fan you have in your case MUST blow the air out the BACK of the case and not pull air in. What you have is turbulance and not a smooth air flow. That causes a great deal of heat buildup.

Air pulled in the front and sides and exited out the back is the best solution for cooling and you must get those temps under control. Also, make sure that your heatsink fan is blowing DOWN over the heatsink and not pulling the air up.
 

·
Registered
Joined
·
878 Posts
Commenting on the heat issue...

I just built up a new computer with an ASUS P5WD2 Premium board, and an Intel 3.2ghz processor. I noticed the high heat right off the bat playing with the AI booster and the hardware monitor in the BIOS. I searched around and found that around 50c idle and up to 80 some degrees celcius was "normal" for these processors, and they dont' shut down until more like 100c. I was in no way comfortable with those temps, even though people didn't seem to think they were a problem. My processor was around 53-55c idle and high 70's under stress. The stock heatsink and fan were very light, and didn't fit tight enough on the processor to form that nice mechanical bond for good heat transfer. I was also using the stock heat compound that came with it.

I upgraded to a copper Coolermaster brand heatsink/fan combo and arctic silver thermal compound i found at Compusa one day when i was out. It came with several backplates and adapters for different processors.

I like the fact that it had a backplate so that you can tighten down the heatsink on the processor for better heat transfer.

That coupled with the better thermal compound and the copper sink reduced my idle temps by over 20c and over 30c under stress!!

Good insurance for stability and better overal performance and longevity for 50 bucks. You invest so much in your computer, you have to protect it as well :)
 

·
Registered
Joined
·
11 Posts
Discussion Starter #11
Okay, it seems that we're focusing on the PSU and heat here, so I started looking into a new power supply. Going with dai's suggestion of a 500W power supply, I'm looking at this one right now:

http://www.extremeoverclocking.com/reviews/cases/Antec_TX1050B_6.html

What do you guys think? Also, before I go shelling out the cash for it only to find out it's not causing the current problems, is there a way I can test to see if it is indeed my PSU first? Maybe unplug some unnecessary perhipherals and try running the PC for awhile, or will that not make a difference?

Finally, this is probably a stupid question, but will the increase from a 300 to 500W PSU dramatically increase the power demands of my computer, because right now my combination of PC, air conditioner and entertainment center on the same circuit (bad planning >_<) occassionally trips the circuit breaker. If it's going to be a problem I can always relocate my computer to another room I guess...

As for the heating issue, one nice thing about the above PSU is the location of its intake fan -- the "vacuum" problem I talked about before shouldn't be an issue so hopefully that will improve airflow in my case somewhat. If it's still a problem after getting the new PSU I'll look into a new heatsink for my CPU then.
 

·
Premium Member
Joined
·
78,003 Posts
you should not have any problems with that supply,it is not going to increase the load much on the power circuit
it is usually a good idea to see if you can borrow a supply first,or get one from a supplier with a restocking policy
 

·
Registered
Joined
·
11 Posts
Discussion Starter #13
Hrm... This can't be good. I've been playing around with unplugging other devices (not that it's helping mind), and decided to disable the side fan temporarily (it's extremely high speed and sounds like a friggin' vacuum cleaner when it's on, so I figured it was a large power drain), and my video card is running 89-92C AT IDLE with the side of the case off. I realize that's not an optimum cooling setup and wouldn't ever run the computer at load that way, but that hot at idle doesn't make any sense to me. I'm wondering if this issue might have more to do with my video card than my PSU now. If so, this will be the second NVidia card I've had where the stock heatsink/fan failed on me after about a year, and I won't be very happy. :mad:
 

·
Registered
Joined
·
11 Posts
Discussion Starter #14 (Edited)
Update: My neighbor was gracious enough to allow me to try out his 400W PSU on my PC, and absolutely nothing changed. Clearly the PSU isn't causing this issue.

I've since tried disabling the onboard sound, running with each RAM chip alone, scanning for problems with SiSoft Sandra, tweaking various BIOS settings, all to no avail. This is just plain frusterating now, and the next step is a reformat I'd really like to avoid, particularly because I highly suspect it will do no good whatsoever.

I've added a new game to my list of symptoms. Need for Speed: Underground 2 freezes 100% of the time within mere moments of getting into the game. The method of freezing is what's interesting; after a few moments the screen freezes solid, sometimes with horizontal pink lines appearing in a few places on the screen. After a couple of seconds, a very short snippet of sound is played as if the game is trying to continue, and the frame sometimes updates. Basically the game is running at about 1 frame every 10 seconds or so. After about 30-50 seconds, the game finally freezes solid and nothing happens from that point on until I manually shutdown the system.

I'm getting to the point where I'm willing to try just about anything to diagnose and fix this problem. I've started backing up my files to DVD so I can reformat, but as I said I don't have much hope for that doing anything. Can you guys recommend anything else to try to work this out?

Thanks in advance.
 

·
Registered
Joined
·
25,111 Posts
Hi,

I don't mean to sound like I am beating a dead horse, but what kind of power supply does your neighbor have that you tried? Nice of him to offer that. but my curiousity got the best of me. You need to just cover every base with all these issues and that is the reason that I asked for that information.

Other than that, HEAT, HEAT, HEAT. I reread your posts....wow, you do have a real problem in that area.

Did you change out the factory pad for the Artic Silver yet on the CPU? If so, did you clean off the old stuff first? Did you follow the directions exactly for the Artic Silver? Did you change the direction of the fans to you have one blowing in the front and one blowing out the back, and the heatsink fan blowing down and not up? Oh, not to confuse the issue, you can have others blowing in also with no problem as long as you don't create too much turbulance.
If you take off the side of the case and point a regular fan toward the CPU, does that bring the temps down and stop some of the problems you are having?

I noticed one poster listed that he saw information that the heat was normal at: I searched around and found that around 50c idle and up to 80 some degrees celcius was "normal" for these processors, and they dont' shut down until more like 100c. I was in no way comfortable with those temps, even though people didn't seem to think they were a problem. My processor was around 53-55c idle and high 70's under stress.

I strongly disagree with this information. In my opinion, if you have temps for this CPU of 35c to 45c at-rest, then you have normal temps. Consquently, if you have temps for under-stress of 50c to 63c, then you can consider that normal. Anything above 63c, I would never settle for with this CPU. I did like that he posted that he was not comfortable with these temps, so glad to hear that. This is simply too hot for any CPU in my personal opinion.

Post back before you format and reinstall, because I also don't think that will fix your problem.
 

·
Registered
Joined
·
11 Posts
Discussion Starter #16
Tumbleweed36 said:
Hi,

I don't mean to sound like I am beating a dead horse, but what kind of power supply does your neighbor have that you tried? Nice of him to offer that. but my curiousity got the best of me. You need to just cover every base with all these issues and that is the reason that I asked for that information.
I saw the brand when I tried out his PSU but didn't think to jot it down. I do know it was 400W but that's about it. =P It wasn't a brand I'd ever heard of before or I'd probably have remembered the name. Since it came with the PC he's got it's very possible that it's a noname brand.

Even so, the likelihood of not one but two PSUs causing the exact same problem in my PC seems small. Although I agree that future upgrades will certainly require some more beef from the power supply, I made a point of eliminating non-essential components when I upgraded to the 6800 and as I said have never had a problem previously.

Other than that, HEAT, HEAT, HEAT. I reread your posts....wow, you do have a real problem in that area.

Did you change out the factory pad for the Artic Silver yet on the CPU? If so, did you clean off the old stuff first? Did you follow the directions exactly for the Artic Silver?
I haven't had the chance to do this yet, and would like to exhaust every possible resource I have to pinpoint the definite cause of the problem before I start spending money. (On the other hand just getting a tube of Arctic Silver would be pretty cheap and I could probably do that easily enough)

Did you change the direction of the fans to you have one blowing in the front and one blowing out the back, and the heatsink fan blowing down and not up? Oh, not to confuse the issue, you can have others blowing in also with no problem as long as you don't create too much turbulance.
I'm actually glad you asked because I was planning to repost my temps after reversing the fan and forgot. (FYI the heatsink fan was always blowing downward as it should be) Here are the temps:

At idle:

CPU: 60 Case: 35
GPU: 70 (37 ambient)

Load:

CPU: 83 Case: 36
GPU: 86 (42 ambient)

As you can see, the temps are a little bit better but not phenomenally.

If you take off the side of the case and point a regular fan toward the CPU, does that bring the temps down and stop some of the problems you are having?
It actually seemed to make it worse when I did that. =P In any case the problems definatelly didn't go away.

I noticed one poster listed that he saw information that the heat was normal at: I searched around and found that around 50c idle and up to 80 some degrees celcius was "normal" for these processors, and they dont' shut down until more like 100c. I was in no way comfortable with those temps, even though people didn't seem to think they were a problem. My processor was around 53-55c idle and high 70's under stress.

I strongly disagree with this information. In my opinion, if you have temps for this CPU of 35c to 45c at-rest, then you have normal temps. Consquently, if you have temps for under-stress of 50c to 63c, then you can consider that normal. Anything above 63c, I would never settle for with this CPU. I did like that he posted that he was not comfortable with these temps, so glad to hear that. This is simply too hot for any CPU in my personal opinion.

Post back before you format and reinstall, because I also don't think that will fix your problem.
I greatly appreciate your input, but have to point out again that the temps I'm seeing now are what I've been running at for a year. I can certainly get some Arctic Silver and might even go so far as to replace the CPU heatsink, but that doesn't even affect my video card at all, so if the problem lies there it doesn't really help. I'd consider replacing the vid card's heatsink if I hadn't done it once already. Not a fun experience. =P While I am in full agreement with you that my temps are a problem, I honestly don't believe them to be the root cause of this particular issue.

Just look at the symptoms. If heat were the issue, why would the system crash after 5 minutes and then run solid for hours and hours straight? Having had major heat problems in the past, I can assure you that's not how it works. It just doesn't make sense that the system would overheat almost immediately and then suddenly cool down for no good reason and provide perfect performance the rest of the day.

Is there no other avenue I could look into before I start spending money on stuff that in all probability won't help fix the problem here?
 

·
Premium Member
Joined
·
78,003 Posts
check your power requirements here
http://www.jscustompcs.com/power_supply/
based on a quality supply and add 30% to the result
lack of power could be causing some of the heat problem as the airflow through the case must be reasonable as your case temps are not bad
have you checked for dust buildup and checked your video and cpu fans
 

·
Registered
Joined
·
878 Posts
Well, apparently i got some bad info on the temps, but cooler is better and i didn't like the flimsy heatsink, so i had nothing to lose by upgrading it.

Now that you mention the Video card running hot, i had instability issues with my last computer. It would run alright during normal operations, but try and play a game for more than a few minutes or a video, and wham, it'd shut down. Furhter investigation revealed that the stock fan on my ATI Radeon 8500 128mb had slowed down to a mere crawl. After a good cleaning and oiling it was back up to snuff and i haven't had a random shutdown since.

:4-dontkno
 

·
Registered
Joined
·
11 Posts
Discussion Starter #19
dai said:
check your power requirements here
http://www.jscustompcs.com/power_supply/
based on a quality supply and add 30% to the result
lack of power could be causing some of the heat problem as the airflow through the case must be reasonable as your case temps are not bad
have you checked for dust buildup and checked your video and cpu fans
According to the page, and assuming that the onboard lan and sound chip adds an extra ~11W (and the USB mouse and keyboard draw 5W each), the PSU recommended is 324W at peak load. This only really confirms what I already knew, which is that the PSU is under spec in my system. Now if I remove one of the fans and the DVD drive, the recommendation drops to 292W. Yet the computer still crashes under those conditions.

I plan to check out my neighbor's PSU more thoroughly and give you guys the info I get from it, but I'd be surprised if it wasn't the advertised 400W, in which case it should definitely have solved all my problems if the issue was only the PSU.

The dust buildup idea make sense. I'll probably grab a can of air next time I'm at the store and see what that gets me. Thanks again for the suggestions.

Oh, and to muddy the water, World of Warcraft *seems* to have suddenly and inexplicably started working perfectly. Go figure. NFS:U2 still crashes within seconds, and I've run it on this computer before with no problems so I'm assuming it's related to this problem. I need to test more with the other software but it was always hit and miss trying to diagnose the issue so that might take some time.
 

·
Registered
Joined
·
11 Posts
Discussion Starter #20
Rashiki said:
Now that you mention the Video card running hot, i had instability issues with my last computer. It would run alright during normal operations, but try and play a game for more than a few minutes or a video, and wham, it'd shut down.
Those sound like exactly the same symptoms I had when my GF4's fan gave up the ghost, and it's what I'd expect if that were the problem here. It just doesn't make sense to me that the card would overheat almost immediately and then work fine for hours.
 
1 - 20 of 21 Posts
Status
Not open for further replies.
Top