Going Nucking Futs! Need help! (Kind of long) - Tilted Forum Project Discussion Community

KrazyKracka · 10-27-2004, 03:36 PM

Greetings everybody,

Here's my Problem: A bout 3-4 months ago, I helped my friend build a new PC. Two weeks ago, he calls me up and tells me he needs cash badly and asks if I would like to buy the PC. Since I know all the details about it, I said "absolutely". Here's where it gets tricky: After I took delivery, I added 200 gigs of HD space (80 gig master and 120 gig slave disk), added 1 gig of RAM to the machine (PC 2700 SD DDR), My own CD RW drive and a new DVD RW Drive, along with Win XP PRO SP2, and a PCI wireless G card. about 5 Days ago, my system crashed, just as if someone pulled the power plug, and restarted. I restarted, sent an error report, which it told me that the crash was caused by a graphics device driver. It got progressively worse over the next few days, to a point where the system would not even boot.

I formatted the Primary hard disk and reinstalled the OS. The system worked good for 36 hours, where it crashed again and gave me the same error message. To make a long story short, I did this SEVERAL times over the past few days. I ended up diagnosing the graphics card as being bad. The card is a NVIDIA GeForce 5200, with 256MB RAM. I have an older 32 MB PNY technologies card, which I used in its place, and it seemed to work OK. Until TODAY, when it crashed again! This time there was no error report option. I turned down the hardware acceleration to none, and the system seems much more stable. It has crashed once with this card and settings. Has anyone here had any type of problem like this? My 'board is an Elitegroup model K7VTA3, and it runs an AMD Athlon XP 2400 + processor, currently running at 2.10 GHz. If anyone can offer any insight, I would be especially grateful*.

Thanks in Advance!

*gratitude to be returned in backrubs

Aerundel · 10-27-2004, 03:55 PM

Have you changed the graphics driver? If it isn't the newest, update it. If it's the newest, try using an older one. Or if you're brave, try a (leaked) beta driver even newer than the latest official driver. http://www.guru3d.com

Redjake · 10-27-2004, 03:58 PM

this might sound funny, but are the CPU temps ok? I know if your CPU gets really hot, like mine does (OCed a lot), STUPID errors can start to be generated. Just all-out weird stuff happening. how's the temperatures looking?

KrazyKracka · 10-27-2004, 04:04 PM

Aerundel, thanks, but I have tried using both old and new drivers, without any positive results. I get the same performance (unwarned crashing) with any driver I install for the card. Also, with that Nvidia Geforce card, it would cause "stop" error messages during the boot cycle, and one of my boot sectors actually became corrupted, hence the constant reformatting for 5 days straight.

Redjake, my CPU is currently running at 37 degrees C. I was wondering this myself, because the problem is sporadic, there's no method to it.

Thanks to both of you I appreciate your suggestions.

Lasereth · 10-27-2004, 05:49 PM

The only thing I can suggest is the beginning of the hardware-exchange process. I'd try to find another motherboard to use first. Just switch every component in the system until it stops doing it. It sounds tedious but it's really one of the fastest ways to find out what's wrong with the system. If you don't have extra components, just take out individual pieces that aren't needed (like the extra HDD, CD-ROM, some RAM, etc.) and see if it fucks up without those pieces in.

-Lasereth

jonjon42 · 10-27-2004, 05:58 PM

hmm too many hds? hard drives can get quite hot sometimes...also do memtest86 overnight and make sure your memory is good (google memtest or get the knoppix cd)

RAGEAngel9 · 10-27-2004, 10:37 PM

Another guess is the power supply is too weak to handle everything.
You have some good things runnign there and if you happened to get an older or weaker power supply, that could cause the problems.
Just a guess.

Rawb · 10-28-2004, 12:00 AM

Since all we can do right now is grasp at straws. We just have to start somewhere and work from there. Go ahead and start by reducing the voltage and Mhz of the CPU to their factory default settings, be sure that your PCI bus is locked back to the normal 33Mhz. Try running that for a while. If that doesn't work, my next recommendation is to run a program called memtest86 which you can find at: http://www.memtest86.com/ This program will take a while to run, let it run at least overnight so that your ram and CPU can get up to normal working tempature for a while. After all of that is done, we can start playing the hardware swapout game.

Ashton · 10-28-2004, 02:28 AM

Since all you did was add ram and a hard drive I would pull teh new ram and test it with the ram that was in it when you got it, bad ram can cause all kinds of problems.

KrazyKracka · 10-28-2004, 03:15 AM

Hey guys, thanks for all of your replies! As far as the process of elimination goes, I ran the system with its original RAM and still had the problem. I eliminated my slave 120 gig HDD, and still had the problem, I then removed EVERYTHING that the PC didn't need, so it was running ONE CD ROM drive, and the Primary HDD. I STILL had the problem! YAY! (I tried using different CD-ROM drives as well). Since I have installed the 32 MB card and reduced the hardware acceleration, I am starting to notice a pattern here. When I first boot up the PC (after being off for several hours), it will run for about 30 seconds, then crash. It will then reboot, run for about 2 minutes and crash again (did this yesterday as well) and now, being the 3rd time I've booted, it has been running approximately 15 minutes and appears stable. WTF???

I've also noticed at a "cold" boot up, that the monitor "flares" up, meaning its brightness and intensity will vary subtly, just enough to make me notice that it's going on. I also was wondering about the power supply, but it is rated at 450W continuous. However I am also running 6 fans on the machine as well (not counting the one built in to the power supply), one on the processor, 2 for the PCI cards,
one for the hard drives, one at the back of the 'board, and one at the top of the case. Could these be putting excessive straing on the power supply even though it's rated at 450W? I also unplugged 2 of the fans at first, but I again had the problem.

Note: This morning when I booted up it displayed "BIOS runcheck (or sumcheck) error", and prompted me for a floppy disk, I restarted, and it booted normally.

Is this info at all useful to you? I'll give any more information I can, but right now I'm going to download that memtest and get some sleep!

TheProf · 10-28-2004, 06:39 AM

Hello,

I had a problem with my machine always resetting. Took quiet a lot of hair-pulling before I figured out it was a CPU temperature problem.

Some suggestions:
1) Pull out all the fans and clean off the dust from them - including the CPU fan. Same with all the heatsinks. And the power-supply. Sounds like they're doing a lot of work, and even a little dust will act as a nice thick insulation layer.
2) Check your voltages using your BIOS. Compare that with your motherboard specs and see if they are correct. If not, the power-supply may be wonky.
3) Relating to the power-supply: a quick visit to http://www.jscustompcs.com/power_supply/ shows me that you're probably drawing more than 350 Watts of power with all those components, which may mean you're taxing your 400 watt power-supply (and therefore generating more heat). The flipside is that more fans to cool the case down will result in more power being drawn.
4) Does the computer generate any beeps when you're turning it on? Those BIOS codes tell you quite a lot in terms of what's going wrong. It was because of those codes I was finally able to solve my problems.
5) I feel heat is your issue. Try downloading http://www.almico.com/speedfan.php and running it to check your temps, fan speeds, etc. It is an EXCELLENT piece of free software.

Good-Luck

KrazyKracka · 10-28-2004, 12:14 PM

Thanks Prof, I'm going to try that out and see what happens. I was erroneous in my statement about the power supply, I checked the rating and it's 450W MAX output, but I can't find a continuous rating on it. I cleaned the CPU fan and heatsink already... as of today there are no crashes, so I'm going to check the voltages and see what's going on. Thanks again everyone!

billege · 10-29-2004, 12:05 AM

Bad RAM or overheating can and will cause some very odd, hard to trace, errors. Start swappin' out known-good parts and take notes. You'll find it.

KrazyKracka · 10-29-2004, 03:09 AM

And the plot thickens:

Today I was inspecting my GeForce 5200 Card, and I found a bent transistor. I straightened it out and thought "well, why not?". The PC booted... and it's showing no graphics problems as of yet. :yay: HOWEVER, The monitor still "flares" up at first startup, followed by the crash 30 seconds in to the system, followed by another crash, followed by stable performance. I did get a BIOS ROM sumcheck error again, but as I am writing this, the system is being very stable, with the 256MB card and FULL hardware acceleration. I'm still going nucking futs!

I have never flashed a BIOS before, and I'm considering it now. Is this an easy project?

PS: I used the fan speed/voltage/temp utility, and everything is within normal specs, so I'm either looking now at the power supply, OR the BIOS. Am I on the right track?

wakelagger · 10-29-2004, 06:33 AM

Is your computer still running with SP2? Was it running wih SP2 before your friend gave the computer to you?
From what I've heard, some drivers that are supposed to work with WinXP do not work with service pack 2.
I have heard almost nothing good about SP2, and I'm personally waiting for the bugs to be ironned out before I make the switch myself.

And flashing your bios isn't too hard, just follow the directions down to a T. If not, your motherboard won't be much more than a expensive dinner plate.

enderfax · 10-29-2004, 07:27 AM

1. Reset the BIOS using the motherboard pins. Check your manual for specifics.
2. Check your motherboard for swollen and/or bursting capacitors.
3. Flash the BIOS: As stated, this is easy. Most all companies now have flash programs that you just download, install, load the flash file, and run. You used to have to reboot to a floppy and do all this weird shit, and could possibly fuck up your situation, but it's quite easy now.
4. 2.10GHZ is not stock for an XP2400+. That's over-clocked about 100mhz, which *should* be fine. Some can handle it, some can't. Every CPU is different. My old 2400+ could handle 2.25GHZ easily, until I changed motherboards. Now it weirds out at anything past 2.10, so I run it at 2086mhz. Was your friend running it at this speed?
5. In the BIOS check your CPU/Memory voltages. CPU should be running at 1.65-1.7 volts. If it's lower your system could be tripping out, if higher things can get REALLY messed up.

The checksum error is a pretty big giveaway that it's something with the motherboard. Weither it's puked out, or just needs some tweaking, I don't know.

merlin · 10-29-2004, 01:00 PM

a cpu doesn't have to be too hot to be unstable; try bringing it back to stock speeds and voltages. i would def clear the cmos (usually a jumper on your mobo). also, it didn't seem like you take out ALL of the unnecessary components, you still had fans and a cd drive in. but if you're really stumped, i'd start the hardware swap. debugging is always annoying, so good luck =)

KrazyKracka · 10-29-2004, 02:20 PM

To answer a couple of the questions: My friend ran the processor at 2.0 GHz, however when he told me he tried different things, he had crashes all the time at first. Currently, I personally have not adjusted the clock speed on the processor, I simply selected "load optimized defaults" and the processor is running now at 2.1 GHz.

Yes, My PC is running with SP2. I didn't hear anythign about compatibility issues for SP2 yet. In fact, I never ran the system WITHOUT SP2, I always went straight to Windows Update and downloaded all my updates again before playing.

I just checked the voltages and everything is fine... it was 1.63. Temperatures were all within spec too. Thanks for the advice on flashing the BIOS... I'm also going to load up the "fail-safe" defaults and see how stable my system is.

enderfax · 10-30-2004, 12:38 AM

I'd bump up the voltage a notch. More then likely not the problem, but it couldn't hurt.

KrazyKracka · 10-30-2004, 03:16 AM

ok thanks I'll give that a shot.

slimsam1 · 10-30-2004, 07:05 AM

It sounds like a RAM issue to me.

Rawb · 10-30-2004, 03:08 PM

How did the memtest run go? Were there any errors?

KrazyKracka · 11-05-2004, 12:58 AM

Sorry I took so long to reply, I've been really busy this past week.

No errors with memtest. But I found the problem! FINALLY! YAY! After all those headaches... it was my SURGE PROTECTOR! YEAH!!!!! whoo-hoo *rolls eyes*...

Thanks to all of you for your input, it's nice to know there are people willing to help.

10-27-2004, 03:36 PM	#1 (permalink)
KrazyKracka Crazy Location: PA	Going Nucking Futs! Need help! (Kind of long) Greetings everybody, Here's my Problem: A bout 3-4 months ago, I helped my friend build a new PC. Two weeks ago, he calls me up and tells me he needs cash badly and asks if I would like to buy the PC. Since I know all the details about it, I said "absolutely". Here's where it gets tricky: After I took delivery, I added 200 gigs of HD space (80 gig master and 120 gig slave disk), added 1 gig of RAM to the machine (PC 2700 SD DDR), My own CD RW drive and a new DVD RW Drive, along with Win XP PRO SP2, and a PCI wireless G card. about 5 Days ago, my system crashed, just as if someone pulled the power plug, and restarted. I restarted, sent an error report, which it told me that the crash was caused by a graphics device driver. It got progressively worse over the next few days, to a point where the system would not even boot. I formatted the Primary hard disk and reinstalled the OS. The system worked good for 36 hours, where it crashed again and gave me the same error message. To make a long story short, I did this SEVERAL times over the past few days. I ended up diagnosing the graphics card as being bad. The card is a NVIDIA GeForce 5200, with 256MB RAM. I have an older 32 MB PNY technologies card, which I used in its place, and it seemed to work OK. Until TODAY, when it crashed again! This time there was no error report option. I turned down the hardware acceleration to none, and the system seems much more stable. It has crashed once with this card and settings. Has anyone here had any type of problem like this? My 'board is an Elitegroup model K7VTA3, and it runs an AMD Athlon XP 2400 + processor, currently running at 2.10 GHz. If anyone can offer any insight, I would be especially grateful. Thanks in Advance! gratitude to be returned in backrubs Last edited by KrazyKracka; 10-27-2004 at 03:38 PM.. Reason: too much content

10-27-2004, 03:58 PM	#3 (permalink)
Redjake I'm a family man - I run a family business. Location: Wilson, NC	this might sound funny, but are the CPU temps ok? I know if your CPU gets really hot, like mine does (OCed a lot), STUPID errors can start to be generated. Just all-out weird stuff happening. how's the temperatures looking? __________________ Off the record, on the q.t., and very hush-hush.

10-27-2004, 05:49 PM	#5 (permalink)
Lasereth Knight of the Old Republic Location: Winston-Salem, NC	The only thing I can suggest is the beginning of the hardware-exchange process. I'd try to find another motherboard to use first. Just switch every component in the system until it stops doing it. It sounds tedious but it's really one of the fastest ways to find out what's wrong with the system. If you don't have extra components, just take out individual pieces that aren't needed (like the extra HDD, CD-ROM, some RAM, etc.) and see if it fucks up without those pieces in. -Lasereth __________________ "A Darwinian attacks his theory, seeking to find flaws. An ID believer defends his theory, seeking to conceal flaws." -Roger Ebert

10-27-2004, 05:58 PM	#6 (permalink)
jonjon42 Psycho Location: inside my own mind	hmm too many hds? hard drives can get quite hot sometimes...also do memtest86 overnight and make sure your memory is good (google memtest or get the knoppix cd) __________________ A damn dirty hippie without the dirty part....

10-29-2004, 12:05 AM	#13 (permalink)
billege Watcher Location: Ohio	Bad RAM or overheating can and will cause some very odd, hard to trace, errors. Start swappin' out known-good parts and take notes. You'll find it. __________________ I can sum up the clash of religion in one sentence: "My Invisible Friend is better than your Invisible Friend."

10-27-2004, 03:55 PM	#2 (permalink)
Aerundel Upright	Have you changed the graphics driver? If it isn't the newest, update it. If it's the newest, try using an older one. Or if you're brave, try a (leaked) beta driver even newer than the latest official driver. http://www.guru3d.com

10-27-2004, 04:04 PM	#4 (permalink)
KrazyKracka Crazy Location: PA	Aerundel, thanks, but I have tried using both old and new drivers, without any positive results. I get the same performance (unwarned crashing) with any driver I install for the card. Also, with that Nvidia Geforce card, it would cause "stop" error messages during the boot cycle, and one of my boot sectors actually became corrupted, hence the constant reformatting for 5 days straight. Redjake, my CPU is currently running at 37 degrees C. I was wondering this myself, because the problem is sporadic, there's no method to it. Thanks to both of you I appreciate your suggestions.

10-27-2004, 10:37 PM	#7 (permalink)
RAGEAngel9 Insane Location: Vermont	Another guess is the power supply is too weak to handle everything. You have some good things runnign there and if you happened to get an older or weaker power supply, that could cause the problems. Just a guess.

10-28-2004, 12:00 AM	#8 (permalink)
Rawb Crazy Location: Salt Town, UT	Since all we can do right now is grasp at straws. We just have to start somewhere and work from there. Go ahead and start by reducing the voltage and Mhz of the CPU to their factory default settings, be sure that your PCI bus is locked back to the normal 33Mhz. Try running that for a while. If that doesn't work, my next recommendation is to run a program called memtest86 which you can find at: http://www.memtest86.com/ This program will take a while to run, let it run at least overnight so that your ram and CPU can get up to normal working tempature for a while. After all of that is done, we can start playing the hardware swapout game.

10-28-2004, 02:28 AM	#9 (permalink)
Ashton PORNHOUND Location: California	Since all you did was add ram and a hard drive I would pull teh new ram and test it with the ram that was in it when you got it, bad ram can cause all kinds of problems.

10-28-2004, 03:15 AM	#10 (permalink)
KrazyKracka Crazy Location: PA	Hey guys, thanks for all of your replies! As far as the process of elimination goes, I ran the system with its original RAM and still had the problem. I eliminated my slave 120 gig HDD, and still had the problem, I then removed EVERYTHING that the PC didn't need, so it was running ONE CD ROM drive, and the Primary HDD. I STILL had the problem! YAY! (I tried using different CD-ROM drives as well). Since I have installed the 32 MB card and reduced the hardware acceleration, I am starting to notice a pattern here. When I first boot up the PC (after being off for several hours), it will run for about 30 seconds, then crash. It will then reboot, run for about 2 minutes and crash again (did this yesterday as well) and now, being the 3rd time I've booted, it has been running approximately 15 minutes and appears stable. WTF??? I've also noticed at a "cold" boot up, that the monitor "flares" up, meaning its brightness and intensity will vary subtly, just enough to make me notice that it's going on. I also was wondering about the power supply, but it is rated at 450W continuous. However I am also running 6 fans on the machine as well (not counting the one built in to the power supply), one on the processor, 2 for the PCI cards, one for the hard drives, one at the back of the 'board, and one at the top of the case. Could these be putting excessive straing on the power supply even though it's rated at 450W? I also unplugged 2 of the fans at first, but I again had the problem. Note: This morning when I booted up it displayed "BIOS runcheck (or sumcheck) error", and prompted me for a floppy disk, I restarted, and it booted normally. Is this info at all useful to you? I'll give any more information I can, but right now I'm going to download that memtest and get some sleep!

10-28-2004, 06:39 AM	#11 (permalink)
TheProf Crazy	Hello, I had a problem with my machine always resetting. Took quiet a lot of hair-pulling before I figured out it was a CPU temperature problem. Some suggestions: 1) Pull out all the fans and clean off the dust from them - including the CPU fan. Same with all the heatsinks. And the power-supply. Sounds like they're doing a lot of work, and even a little dust will act as a nice thick insulation layer. 2) Check your voltages using your BIOS. Compare that with your motherboard specs and see if they are correct. If not, the power-supply may be wonky. 3) Relating to the power-supply: a quick visit to http://www.jscustompcs.com/power_supply/ shows me that you're probably drawing more than 350 Watts of power with all those components, which may mean you're taxing your 400 watt power-supply (and therefore generating more heat). The flipside is that more fans to cool the case down will result in more power being drawn. 4) Does the computer generate any beeps when you're turning it on? Those BIOS codes tell you quite a lot in terms of what's going wrong. It was because of those codes I was finally able to solve my problems. 5) I feel heat is your issue. Try downloading http://www.almico.com/speedfan.php and running it to check your temps, fan speeds, etc. It is an EXCELLENT piece of free software. Good-Luck

10-28-2004, 12:14 PM	#12 (permalink)
KrazyKracka Crazy Location: PA	Thanks Prof, I'm going to try that out and see what happens. I was erroneous in my statement about the power supply, I checked the rating and it's 450W MAX output, but I can't find a continuous rating on it. I cleaned the CPU fan and heatsink already... as of today there are no crashes, so I'm going to check the voltages and see what's going on. Thanks again everyone!

10-29-2004, 03:09 AM	#14 (permalink)
KrazyKracka Crazy Location: PA	And the plot thickens: Today I was inspecting my GeForce 5200 Card, and I found a bent transistor. I straightened it out and thought "well, why not?". The PC booted... and it's showing no graphics problems as of yet. :yay: HOWEVER, The monitor still "flares" up at first startup, followed by the crash 30 seconds in to the system, followed by another crash, followed by stable performance. I did get a BIOS ROM sumcheck error again, but as I am writing this, the system is being very stable, with the 256MB card and FULL hardware acceleration. I'm still going nucking futs! I have never flashed a BIOS before, and I'm considering it now. Is this an easy project? PS: I used the fan speed/voltage/temp utility, and everything is within normal specs, so I'm either looking now at the power supply, OR the BIOS. Am I on the right track?

10-29-2004, 06:33 AM	#15 (permalink)
wakelagger Insane Location: Toronto	Is your computer still running with SP2? Was it running wih SP2 before your friend gave the computer to you? From what I've heard, some drivers that are supposed to work with WinXP do not work with service pack 2. I have heard almost nothing good about SP2, and I'm personally waiting for the bugs to be ironned out before I make the switch myself. And flashing your bios isn't too hard, just follow the directions down to a T. If not, your motherboard won't be much more than a expensive dinner plate.

10-29-2004, 07:27 AM	#16 (permalink)
enderfax Upright	1. Reset the BIOS using the motherboard pins. Check your manual for specifics. 2. Check your motherboard for swollen and/or bursting capacitors. 3. Flash the BIOS: As stated, this is easy. Most all companies now have flash programs that you just download, install, load the flash file, and run. You used to have to reboot to a floppy and do all this weird shit, and could possibly fuck up your situation, but it's quite easy now. 4. 2.10GHZ is not stock for an XP2400+. That's over-clocked about 100mhz, which should be fine. Some can handle it, some can't. Every CPU is different. My old 2400+ could handle 2.25GHZ easily, until I changed motherboards. Now it weirds out at anything past 2.10, so I run it at 2086mhz. Was your friend running it at this speed? 5. In the BIOS check your CPU/Memory voltages. CPU should be running at 1.65-1.7 volts. If it's lower your system could be tripping out, if higher things can get REALLY messed up. The checksum error is a pretty big giveaway that it's something with the motherboard. Weither it's puked out, or just needs some tweaking, I don't know.

10-29-2004, 01:00 PM	#17 (permalink)
merlin Tilted	a cpu doesn't have to be too hot to be unstable; try bringing it back to stock speeds and voltages. i would def clear the cmos (usually a jumper on your mobo). also, it didn't seem like you take out ALL of the unnecessary components, you still had fans and a cd drive in. but if you're really stumped, i'd start the hardware swap. debugging is always annoying, so good luck =)

10-29-2004, 02:20 PM	#18 (permalink)
KrazyKracka Crazy Location: PA	To answer a couple of the questions: My friend ran the processor at 2.0 GHz, however when he told me he tried different things, he had crashes all the time at first. Currently, I personally have not adjusted the clock speed on the processor, I simply selected "load optimized defaults" and the processor is running now at 2.1 GHz. Yes, My PC is running with SP2. I didn't hear anythign about compatibility issues for SP2 yet. In fact, I never ran the system WITHOUT SP2, I always went straight to Windows Update and downloaded all my updates again before playing. I just checked the voltages and everything is fine... it was 1.63. Temperatures were all within spec too. Thanks for the advice on flashing the BIOS... I'm also going to load up the "fail-safe" defaults and see how stable my system is.

10-30-2004, 12:38 AM	#19 (permalink)
enderfax Upright	I'd bump up the voltage a notch. More then likely not the problem, but it couldn't hurt.

10-30-2004, 03:16 AM	#20 (permalink)
KrazyKracka Crazy Location: PA	ok thanks I'll give that a shot.

10-30-2004, 07:05 AM	#21 (permalink)
slimsam1 Crazy	It sounds like a RAM issue to me.

10-30-2004, 03:08 PM	#22 (permalink)
Rawb Crazy Location: Salt Town, UT	How did the memtest run go? Were there any errors?

11-05-2004, 12:58 AM	#23 (permalink)
KrazyKracka Crazy Location: PA	Sorry I took so long to reply, I've been really busy this past week. No errors with memtest. But I found the problem! FINALLY! YAY! After all those headaches... it was my SURGE PROTECTOR! YEAH!!!!! whoo-hoo rolls eyes... Thanks to all of you for your input, it's nice to know there are people willing to help.