Hardware Woes / Computer Science / Forums

Forums

4hv.org :: Forums :: Computer Science

« Previous topic | Next topic »

Hardware Woes

1 2 next

Move Thread

LAN_403

Part Scavenger

Sat Jun 30 2007, 02:23PM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

There's something wrong with my computer. I'm convinced that it's a hardware problem at the lowest level. Trouble is, I don't know where to start to figure out what it is. The problem is presenting itself as follows:

The computer keeps restarting at random, most often when Windows XP is booting. Oftentimes the only way to get into the OS is by booting into safe mode and doing system restore. Windows repair *sometimes* will go all the way through, but most often I get copy errors and the "blue screen of death." The computer boots best when it is off over night, given time to cool, etc. I also get errors saying my hard drive is corrupt, more often something about memory, and other random errors. If I ever FINALLY get into Windows, it usually works fine except Internet Explorer fights with my AntiVirus, which I don't think is connected. *Every once in a while* you'll get a random restart.

Failing the System Restore/WinXP Repair, I got another hard drive that I know is good, formatted it, and tried to install WinXP again. Still copy errors, random restarts and blue screens. I never got it installed.

So, how do I figure out if it's the memory, the CPU, etc? I've built many (25+) computers, and I've never had any trouble like this. When I first started working on it, the CPU fan was fulla crap, so it's possible the CPU is dead. But then again, I thought MOBOs nowadays had overheat shutdown. I don't want to buy a new one only to find out it's something else. So how do I test the memory and CPU? I'd like to test the power supply too, to make sure it can supply enough power. It's 350W, IIRC.

I'm using a BioStar P4M800Pro-M7 (It's not very expensive) and a Intel P4 processor.

Thanks.

Marko

Sat Jun 30 2007, 04:39PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Have you tried entering BIOS and leaving it as such, does it reboot then?

I had one computer wich was randomy freezing similar way, although at sparser rate. It was never fixed. Some computers look like they form a mind of their own, and nothing in world can help them. (Not to mention how I hate that).

I would recommend you to try borrowing another motherboard and seeing if it works, if it doesn't the the same with CPU. That's the only way you can accurately find hardware problems, and for some reason I believe it's the motherboard here.

Ken M.

Sun Jul 01 2007, 01:55AM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

I doubt your CPu is toasted, for the reason you stated but also because if I'm not mistaken if theres no cpu the pc won't even turn on, if it does you surely won't get to the windows screen.

Perhaps a section of your ram has been\is corrupt which may be causing the issue doubtful, and if your ram was dead you'd get a bios beep error (a series of beeps that indicate a problem that happened on POST).

Your copy of windows maybe corrupt of have a serious but not obvious disc flaw, it may load properly once or twice but then it won't work at all, happened to a game disco f mine 1 time, the data that was burnt onto the disc waSn't all there, but it still let me load the game, but caused issues while playing. Solution get a different disc and try it, i think you said someone you know does pc work and you had access to discs.

You may have a bad data cable from HDD to mobo.

All these are suggestions, also one thing I've learned with pc's is definitely DO NOT assume, I've assumed stuff before and turns out it was something easy or at least semi easy to fix.

Dave Marshall

Sun Jul 01 2007, 02:56AM

Registered Member #16 Joined: Thu Feb 02 2006, 02:22PM
Location: New Wilmington, PA
Posts: 554

I'd wager good money its one of two things.

1) Your RAM. Try swapping it with a known good stick or, if you have more than one stick in there, run it for a bit with one removed, then the other. The bad stick will become very apparent. Alternately, download an Ubuntu install CD and boot off that to run memtest.

2) Your powersupply. Intermittent power supplies can be maddeningly hard to detect because they can present a variety of ways. Unstable voltage can cause no end of different errors from freezes, to BSODs, to a supposedly missing or corrupt hard drive. If you've got another one, swap it out and see if the problem persists. The fact that it runs ok when first started in the morning is a gigantic neon sign indicating heat problems somewhere.

Check your heatsinks also. Chipset, CPU, etc. Make sure nothing has come loose or is lacking thermal grease.

If none of that works, then we have to dig a little deeper.

Dave

Alex

Sun Jul 01 2007, 03:40AM

Geometrically Frustrated
Registered Member #6 Joined: Thu Feb 02 2006, 04:18AM
Location: Bowdoin, Maine
Posts: 373

Dave Marshall wrote ...
[...]download an Ubuntu install CD and boot off that to run memtest.

You could also use the 44kB memtest iso.

In addition to checking your heatsinks and fans, look for a program to monitor your CPU temperature. It'd be nice to know what temp the system is at before it crashes.

Part Scavenger

Sun Jul 01 2007, 05:59PM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

Hmph. I can't get Ubuntu to boot... Still working on that. I know this disc is good! I do not have reset problems when just booted to the BIOS. Nothing is hot either. It's not the hard drive for sure.

Here's some stats:
After several crashes, I went to the BIOS and read the info:
CPU Temp = 36C
System Fan Speed = 2410
3.3V = 3.21V
12V = 11.58V
CPU Fan Speed = 2909
CPU VCore = 1.36V
+5(V) = 5.10-5.13V

Is the low voltage normal?

Woot! Ubuntu is booting!

The attached picture is the reason I think it is the CPU. I guess my parent's don't exactly clean behind their computer... Uuuck!!!

Furthermore, the default BIOS setting is no overheat shutdown on the CPU.

I wanted to make certain that this was the problem before they spent $100 bucks on a new one. But why does it work fine for a while and then not hardly at all? Everything except the PSU is about 6mo old.

Ubuntu just crashed on the drive format... Burning MemTest.iso...

Ken M.

Sun Jul 01 2007, 06:49PM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

The low voltages on the cpu are correct. Just reading the cpu temp from bios doesn't really work to well since you must restart the pc which means it may not have been on long enough. As for a cpu temp program, I don't tink using 1 designed for another mobo would work but if anyone thinks it might, see if you can't download asus probe from the net, it'll display fan speeds, cpu temp, chipset temp, mobo temp, all the voltages, and it'll sound an alarm when a certain thing goes wrong i.e temp too high, fan too slow, voltage too high or low.

thedatastream

Sun Jul 01 2007, 07:44PM

Registered Member #505 Joined: Sun Nov 19 2006, 06:42PM
Location: Yorkshire!
Posts: 329

I have a copy of Ultimate Boot CD

in my CD stack which is useful for diagnosing hardware issues. You can give your CPU a workout, check disks and memory.

From your symptoms, it sounds almost identical to a PC I fixed for my neigbour's lad. One of his mates had decided to overclock it by turning the memory up from 333 to 400. I used the memory checker utilities on Ultimate Boot CD to hammer the memory for a while and it wasn't long before the errors started to pile up. Turned down the memory to the correct speed and it worked fine. Returned to user with a BIOS password applied to prevent future service calls!

Part Scavenger

Sun Jul 01 2007, 08:04PM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

Hmm. I actually have that, but It's a Windows program which I can't get into. Does it come in boot form?

Ubuntu ALWAYS fails "debootstrap error" while installing the base system on either "dash" or "bash". Including the first time I mentioned... I was, in fact, not still formatting.

I've been running the MemTest. > 4000 errors in 40 min! 95+% of them happen in the "Moving inversions, random pattern" On a single run (about 10 min) I have 1400 errors in this test alone, but very few errors on other tests. Tests 3 and 7 generally get around 10-30 errs. The rest contain no errors. Errors almost exclusively happen in the 195-220MB range on the "Failing address" readout. Description of tests:

The "BadRam" readout always contains twenty sectors, but they are not the same sectors between reboots, however they are almost always exactly the same between tests. I have no idea what this all means... and I don't have access to another DDR2 stick that I know of.

I tried different PSUs, doesn't seem to make any difference, only the 12 and 5V rails varied about 0.08Volts. The rest was exactly the same. I've been checking the CPU temp often also, I've never seen it over 37C.

Ken M.

Sun Jul 01 2007, 11:30PM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

Hmm, I'd definitely say your memory's bad, I think you could use a ddr1 stick, but at least try Dave's idea, of just pulling out 1 ram stick and see where it goes, then put stick 1 back and pull out stick 2, and repeat upto stick four. If you only have one stick, then I'd say your pretty well out of luck.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob