Eventhough this is pretty much what I do for a living (tech support), I'm stumped by a problem that I'm having with my home PC.
When I first built the PC around Jan of '04, I had a Soyo P4VGA mobo with a 2.6Ghz Intel Celeron. Jan of this year, I swapped it out for a 2.8Ghz P4 w/ HT and a Elite Computer Systems 661FX-M mobo.
After about two weeks, the ethernet and audio quit working on the board. I figured it was just a bad board so I got an RMA and received a new motherboard in the mail. It was the same model, 661FX-M and seemed to work alright at first.
Then, after a few weeks, the ethernet quit again. Rather than having to replace the board again, I just bought a PCI ethernet card which is now working fine. A few months later, I splurged on some parts. I spent a little over 300.00 at newegg.com for a Radeon 9800Pro video card, an extra stick of 512mb of Ram, a 16x DVD-R drive and a new Hitachi 80Gb SATA hard drive.
The first problem that I noticed was with the RAM. I already had a stick of 256mb PC3200 and added the stick of 512mb PC3200, but the computer would only recognize one at a time. I would boot up and it would see the 256 then I would reboot and it would see the 512.
I ended up taking out the 256 and leaving only the 512 on the board since Elite Computer Systems could not tell me what was causing the problem on the board.
That was when the problems really began.
Say I was playing a game (doom 3, half life 2, UT04, etc), the computer would reboot randomly during the game. Then, it began rebooting by itself at random times.
I thought that since it was rebooting during games, it might be something with the new Radeon card. I contacted Power-Color (the manufacturer) and they didn't have an answer and advised me to try their support forums. I did and they recommended two things - get a new power supply and cool the case down.
Well, I went and bought a new power supply (450w Ultra) and bought an Antec blower that I slid into a PCI slot below my video card. It didn't help at all, and the problem kept getting worse. You could be typing in Word and the machine would restart. I swapped the RAM just to make sure it wasn't causing the issue, and I swapped my Radeon 9800 with a Radeon 9200SE and it still had the same issue.
So, I installed Motherboard Monitor on the computer to see what my proc temp was running at. It normally runs anywhere from 65-75 degrees C at any point and pushes 80 when running a game.
Then the computer started giving me CRC errors whenever I tried to unzip files or when I tried to open exe files. This happens with every single exe/zip that I try to use. I looked up CRC errors and it seems to be related to bad motherboards.
I have tried updating the bios on the motherboard per ECS's recommendation, but nothing helps.
So, here is what I've done.
I bought a Zalman 7000B fan for my processor (http://www.computerhq.com/images/529610_l/Zalman/CNPS7000B-CU-LED.jpg)
And, I bought an ASUS motherboard (http://www.laptops4me.com/images/pict/MB-P4S80DX_LG.gif)
Is this finally going to have my issue fixed? I can't believe I've had so much trouble with my own PC, when ones I've built for other people NEVER have problems.
I'm counting on you other geeky guys to help me out.
Post Edited (07-27-05 07:32)
Skaboi wrote:
>
> After about two weeks, the ethernet and audio quit working on
> the board. I figured it was just a bad board so I got an RMA
> and received a new motherboard in the mail. It was the same
> model, 661FX-M and seemed to work alright at first.
>
> Then, after a few weeks, the ethernet quit again. Rather than
> having to replace the board again, I just bought a PCI ethernet
> card which is now working fine.
I've noticed the on-board NIC's in general are of very low endurance. I've such failures (not specific to your board, just in general); using a dedicated NIC is probably the best bet here.
> A few months later, I splurged
> on some parts. I spent a little over 300.00 at newegg.com for
> a Radeon 9800Pro video card, an extra stick of 512mb of Ram, a
> 16x DVD-R drive and a new Hitachi 80Gb SATA hard drive.
>
> The first problem that I noticed was with the RAM. I already
> had a stick of 256mb PC3200 and added the stick of 512mb
> PC3200, but the computer would only recognize one at a time. I
> would boot up and it would see the 256 then I would reboot and
> it would see the 512.
Lots could be going on there; are the two RAM sticks the same brand? Are the timing specs the same (PC 3200 is only part of the story...what are the RAS and CAS timings, as well as read latency?) Mix-matching RAM is notorius for giving stability problems. The 'best bet' I've observed is to keep the RAM as a matched 'set.' I've also seen some boards having trouble with RAM sticks of two different sizes, even if they otherwise spec the same. SOMETIMES, this is a slot issue; the manual of the board should indicate if the larger stick is required in a certain slot. Or, it could be a bad RAM stick.
>
> That was when the problems really began.
>
> Say I was playing a game (doom 3, half life 2, UT04, etc), the
> computer would reboot randomly during the game. Then, it began
> rebooting by itself at random times.
>
> I thought that since it was rebooting during games, it might be
> something with the new Radeon card. I contacted Power-Color
> (the manufacturer) and they didn't have an answer and advised
> me to try their support forums. I did and they recommended two
> things - get a new power supply and cool the case down.
>
When I hear rebooting under load, I think heat first. Cooling the case down may not help...it's the processor that may be getting hot. (more below).
As for the PS, does the hardware monitor tool you installed (or BIOS) give voltage readouts? 5 and 12 V lines should be +/- 5%; if out of spec, this will cause problems. Also, you might want to check that Vcore is 'right' for your chip, since you changed processors.
>
> So, I installed Motherboard Monitor on the computer to see what
> my proc temp was running at. It normally runs anywhere from
> 65-75 degrees C at any point and pushes 80 when running a game.
>
That's way too hot. 60 C is a good, safe SHUTDOWN temperature (at least for Athlon XP, and I'll admit, I don't know the normal numbers for P4 right off).
Idle temp should be less than 40 C, I'd think. Again, that's an AthlonXP number.
If running 80 C for any length of time, there's a good chance it could damage the CPU itself.
>
> Then the computer started giving me CRC errors whenever I tried
> to unzip files or when I tried to open exe files. This happens
> with every single exe/zip that I try to use. I looked up CRC
> errors and it seems to be related to bad motherboards.
>
Umm, if your system is running that hot, the disk errors may be secondary to cpu heat issues. I'd fix that first. I've got a system that when it hits about 59 C, the disks start getting all screwy; after it cools off, the disks are fine. I've yet to figure out if the cpu itself is bad, as I'm in the middle of a similar troubleshooting problem you are having.
> I have tried updating the bios on the motherboard per ECS's
> recommendation, but nothing helps.
>
> So, here is what I've done.
>
>
>
> I bought a Zalman 7000B fan for my processor (http://www.computerhq.com/images/529610_l/Zalman/CNPS7000B-CU-LED.jpg)
>
>
If it's already cooked, it's done. Having the fan for a new CPU might be a help, but until you KNOW this processor is still good, you may be kinda spitting into the wind.
>
>
>
> And, I bought an ASUS motherboard (http://www.laptops4me.com/images/pict/MB-P4S80DX_LG.gif)
>
>
> Is this finally going to have my issue fixed? I can't believe
> I've had so much trouble with my own PC, when ones I've built
> for other people NEVER have problems.
>
If the CPU has been fried, changing boards or flashing BIOS is not going to help. Can you install your old Celeron and see how it works (or are the new boards not compatible?)
Good luck.
The board would be compatible with the Celeron, but I have already sold that processor to someone.
I'm hoping that the new Zalman fan will cool down the processor quite a bit. It's a fairly large fan - 92mm and all copper. The stock Intel fan is terrible and has never really cooled down the processor correctly.
As for the normal running temp of a P4, it runs in the range of 50-55 C, so at the peak I'm close to 20-25 over the normal temp.
I hope the proc isn't fried. My experience has been that in the past, if the proc is fried then it will crack. I've checked and there doesn't seem to be any (visible) damage.
Skaboi wrote:
>
> I hope the proc isn't fried. My experience has been that in
> the past, if the proc is fried then it will crack. I've
> checked and there doesn't seem to be any (visible) damage.
>
I have seen them go flakey without visible damage. If you have another board you can test the cpu in, that would be a good test, too. I'm not sure how resilient the P4's are to getting hot, but 80 C definitely seems in the red zone to me.
You computer nerds are giving me a headache with trying to follow what you're saying. Man, I was proud of myself years ago when I added memory and replaced the graphics card in my old system...I could never imagine being able to do what you're talking about.
I suspect this is why you guys scored so high on that test.
Skaboi wrote:
> Say I was playing a game (doom 3, half life 2, UT04, etc), the
> computer would reboot randomly during the game. Then, it began
> rebooting by itself at random times.
>
> I thought that since it was rebooting during games, it might be
> something with the new Radeon card. I contacted Power-Color
> (the manufacturer) and they didn't have an answer and advised
> me to try their support forums. I did and they recommended two
> things - get a new power supply and cool the case down.
>
> Well, I went and bought a new power supply (450w Ultra) and
> bought an Antec blower that I slid into a PCI slot below my
> video card. It didn't help at all, and the problem kept
> getting worse. You could be typing in Word and the machine
> would restart. I swapped the RAM just to make sure it wasn't
> causing the issue, and I swapped my Radeon 9800 with a Radeon
> 9200SE and it still had the same issue.
I've been having this same issue myself, but with a Shuttle mobo and Athlon 3200+ proc. I've been assuming it was the video card (GeForce3Ti200) overheating, seeing as how I live in South Texas in a house without air conditioning. I put an extra fan on the video card (my proc fan is an 80mm with a huge heat sink, so for now I'm guessing the video card on my system. As you mentioned, the extra video card fan is below the card, which I suspect does little real good since heat rises. I even added a 120mm fan that blows directly onto the video card, and that has helped a little. At least, the computer reboots less often when it is a bit cooler. How warm is the room the computer is in? This can affect the overall cooling ability of the fans (duh). Then again, it could be the power supply, as Ulthar said. It sounds like you are taking the right steps, eliminating things one by one. My problems also started after an apparent power surge knocked out the onboard NICs on both my computer and the one my wife uses, despite both being plugged into battery backup/surge protectors.
I may end up designing my own case that can be properly cooled. After all, I have an English degree, so I must be able to do a bit of airflow engineering, right?
I know that the heat in the room certainly isn't helping any. We have no trees around our house and we live right on the coast of North Carolina....our house might as be an oven.
Today the temperature outside of the house was 120F (considering the humidity). Inside the house, it's probably around 85-90 degrees at any point. That's when the computer has the most problems. I know that blasting hot air in the case like that isn't helping any.
We'll be moving soon to a new house with proper AC and such so that should help with the temp problem in the house. As I type this, motherboard monitor is reporting 67 degrees C for my processor.
The venelation in my case is pretty good. I've blown smoke into the back via an open PCI slot and the smoke disappears almost instantly so I know the airflow is good.
I do have a feeling that the new proc fan and mobo will fix my problems. If not I give up.
As for your computer Derf, you said it's a Shuttle mobo. Are you using a Shuttle type XPC? If so, that can cause a lot of heat issues. We use a lot of Shuttles at my workplace - mainly for PBX systems (runnin linux 'natch) and we've had heat problems with them in the past.
Another thing to consider with those small cases is electromagnetic interference. I know that sounds like a bunch of crap, but we recently bought a small XPC Antec case to sell as a VOIP PBX. The nic in the mobo was loosing like 50% of the packets.
We replaced the mobo (biostar) with a new Asus micro-atx with the same problem. It ended up being electromagnetic interfence from the power supply. We had to wrap the PSU with tinfoil to finally fix the issue. Those small cases are nice but can have serious issues.
My mobo is a Shuttle AN35(N) Ultra. If I gave the impression my case was small, sorry; it's a full tower case I got from work when we replaced our server. It's got plenty of room for ventilation, but it still seems crowded since I've got a floppy, a LS120 floppy, a DVD-Rom, a CD burner and a slot fan taking up all but one bay in the front, and fans galore taking up the side and rear. I actually have started planning a case to better suit my equipment and style, complete with plenty of airflow in and out. As I said earlier, in my case, I think my video card may be overheating, or else the power supply I bought at the time I upgraded to this Shuttle mobo may not be as reliable as I'd like it to be. As for electromagnetic interference, I don't think that's my problem; I've got nothing near my computer except the monitor.
And at any rate, you started this topic to get help for your problem. Didn't mean to commandeer it. Thanks for the tips, though.
Not a problem Derf. When it comes to computers, I just geek out...a lot... lol
When it comes to computers, heat has became the enemy number 1. I've had issues with evey computer I've owned in the last years, and with my latest system, an AMD 64 3400, I asked in advance for an extra fan in the back (it only came with a 12 cm one in the front). Still, unless the room is ventilated (it is a very small room, and in summers temperature often reaches 30ºC), I get crashes now and then. I've been thinking of adding liquid refigeration, but soo far all the kits I've seen are too expensive, about 200 USD.
I read an article about a water cooler, the Cooler Master Aquagate Mini, that sells for about USD $100. Details are at www.coolermaster.com. The article said it was an excellent cooler, though I don't have any experience with it.
Also, you can try peltier cooling:
http://www.pcmech.com/show/processors/140/
http://www.overclockers.com/tips45/
http://www.dansdata.com/pelt.htm
Thanks people. I may consider some of these cooling solutions, since they seem far more reasonable than my first choice (a Revoltek Athlon64 specific kit). Still, I don't think I'm doing any of these soon. The Athlon64 bill is still too recent in my memory right now...
I wanted to follow-up on this. In an earlier post, I mentioned having a computer that was going all squirrelly that I thought was due to heat and/or a power supply problem.
Before, things would get all weird at about 59.5 C (under CPU load), including disks losing their minds. The 5 V line would read 4.40 V, which is out of the 5% tolerance area.
I've replaced the power supply and have now been burning the CPU for well over an hour. It is staying at 60 C, but is stable. The 5 V line is 4.70 V, which is just a skosh below 5%.
A few tenths of a volt can make a LOT of difference.
I've noticed the CPU core voltage is more stable, also.
In any case, I'll let it burn overnight and see if it is still stable tomorrow. Since it is not uncommon for this computer to run high CPU load calculations for 10-24 hours (or more) at a time, I need to cook it for a while to be sure. But, for now anyway, it seems replacing the Power Supply helped.
Skaboi (and others), I dunno how relevant this is to your issue, but hopefully this will give you something to check if other fixes don't help.
Well, the worst thing about my issue is how sporadic it is. It will run for days sometimes without interruptions and then sometimes it won't run but for about 5 minutes before restarting
I got the motherboard in yesterday and am waiting for the proc fan to arrive. I'll probably have the machine put back together early this coming week, and I'll post an update on how everything is doing
If you don't have it (or an equivalent), I can recommend cpuburn:
http://pages.sbcglobal.net/redelm/
It is specifically designed to maximize heat/power consumption. With the cpuburn for K7 running on my AthlonXP, it ran for over ten hours at 60 C. Running a "real" cpu intensive app, it ran about 4 degrees cooler than with cpuburn, so that utility is really stressing the system.
cpuburn also includes memory burn utility, but for memory testing, I've been using
http://www.memtest86.com/
Both of these have Windows and Linux downloads.
Well, my wife and daughter were doing some shopping yesterday so I took the opportunity to throw in the new mobo and proc fan.
And you know what?
IT WORKS!
The Asus mobo feels so much much less cheap than the ECS board that I was using. I threw it in the case, threw in the proc, proc fan, memory, vid card, etc and had everything wired in just about 30 minutes.
The Zalman fan was especially easy to install. I hate the standard Intel fans that come with the procs as I always feel that I'm going to either break the clips on the mobo or crack my processor. Zalman's fan has two bars that go under the clips on the mobo. Then you sit the fan on the proc and screw it into the bars. It only took about 5 minutes to install, and I couldn't be happier. Now my PC idles at 50 degrees C and about 65 or so when gaming.
There was a huge problem with the mobo though. I'm using a Hitachi Deskstar 80Gb SATA drive. Well, I put in the motherboard and it booted to the XP Pro load screen. After about thirty seconds, it would restart.
I figured I just needed to repair the Windows install, which I've had to do in the past after a mobo swap. I booted to the Windows install disc, yet it said that no drives were found. I called Asus's support and they said that I needed to install the SATA driver as a third party RAID driver when using the Win install disc.
But in order to make a driver disc, I would need a floppy drive.
I absolutely hate floppy drives. For me they are a waste of space, so I don't bother putting them into any of the computers that I build. So, now I'm stuck with my SATA drive out of commission until I can jack a floppy drive out of some old janky computer.
I happened to have a pretty new 40Gb Maxtor IDE drive sitting there that I bought as a spare a few months ago during a sale. I plugged it in, installed Windows and everything there is going fine.
I immediatly installed Half-Life 2 which my computer used to restart with. I kicked up the graphics to 1280x1024 at 32 bit with everything on high. It played perfectly.
So, I installed the Battlefield 2 demo. Once again, 1280x1024 - graphics on high. I used to be able to play the demo with the old motherboard but with the graphics all the way on low. I could never play the multiplayer option because it would constantly restart.
Well, I played multiplayer for about 3 hours last night with no problems. Every now and then, the game would freeze and I would get that sinking feeling until it would come right back up. A server running a graphic intensive game like this with 64 players has burps like that quite often.
I then installed UT2004, once again with the graphics all the way up and it played perfectly. I was getting around 80-90fps constant. I then installed 3dMark03 and it did fine. No restarts whatsoever.
Tonight, I'm going to throw Doom 3 at it with the graphics cranked up. It's so nice to actually having a working machine now. With the new mobo, I'm not getting any CRC errors like I was with the old one. It's nice to see the proc temperature drop right on down as well.
"A server running a graphic intensive game like this with 64 players has burps like that quite often."
If playing over the net, could this be more network issue than actually the server? You are, of course, at the mercy of every router between you and the server. (I did not catch if you were running the server, or playing off a remote).
I'm glad you like the ASUS board. I'm getting ready to order the ASUS A8N-SLI Deluxe with a Athlon 64 X2 (dual core); I thought my next box(es) would have Tyan boards (I am historically an Abit man), but the specs/reviews of the Asus have been stellar.
By the way, I personally have never seen a Windows install survive a mobo swap; I'm sure it does happen, but it's outside my experience (I've done some work in a PC repair shop).
I'm glad it's working better. That sounds like a rockin' system for gaming. And for posting on BadMovies.org. ;)
Ulthar,
It was the server because as soon as the little freeze ups would occur, people would start complaining about it.
As for the Windows problem, it's fine. I have a DVD-R drive so I backed up all of my music/movies/etc (16Gb) to a few DVDs cause I figured I'd lose most of it.
And, I'm quite happy with the Asus board. They are top of the line, and mine is great eventhough it's a low end board. Their support is decent, but outsourced (is it outsourcing if the company isn't American?), and their support is long distance.
I've never used Tyan boards, although I've heard good things. I might consider them if/when I build a new PC - in the distant future.