System Problem

[aae991]

My system:

AMD Dual 1800+MP
1 GIG DDR
Gainward Ti200
Dual 40 gig HD, one drive Win2k, the other Win98SE
DV Storm video capture card
Adaptec 2910 SCSI card
SB Live

The problem:

I got my system in October 2001 and everything worked GREAT until 3 weeks ago when the Ti200 crashed. The builder replaced the Ti200 with a new one. Ever since then, the system spontaneously re-boots or sometimes never can fully boot before re-booting. At first it was isolated to the Win2k OS. Now Win98SE is re-booting as well. The OS is determined by whichever drive I assign as the master.

Any ideas what may be going on? The system builder admits they’re a bit stumped.

Thanks!

 

[Quetzacoatl]

I didn't know adaptec made RAID cards! Well, this is a longshot, but you could have corrupted hard drives...try reflashing the bios on your mobo as well. If it is rebooting on its own, it could be the power supply of the case. You want at least 300Watt, preferably 350Watt. Other than that...may just be that the two OS don't work well together. Are you trying to boot as Windows 2000 with like 98SE a backup? Too many unknown factors to make a certain guess

"When there's a will, there's a way."

 

[IIB]

he'll need alot more then 300w for a dual system...

This post is best viewed with common sense enabled

 

[aae991]

I forgot to add a couple of things...

1) 465W Enerex power supply
2) The SCSI card is just for some peripherals.
3) I've got a Promise Fastrack setup for video editing with 2 additional 60 GIG drives.
4) I choose my OS through the setup screen, choosing which drive will be the master, which one the slave

The system did work perfectly for over 4 months... Not now!

 

[IIB]

did the room temperture changed during these 4 months (say summer happened)... your hard drive or cpu might over heat... you can check your CPU temp, but not sure about the Hard drive... what kind of cooling do you use for your case?

This post is best viewed with common sense enabled

 

[aae991]

No major temp changes... I've got a good cooling system for my system. CPU temps staying below 50C. 3 rear exhaust fans, 3 incoming air fans. However, when the system came in to have the original Ti200 replaced, they admitted the CPU cooling unit became slightly loose on one of the CPUs. If this did allow a temporary surge in CPU temperature, what would it take to damage the 1800+. And if it was damaged, would spontaneous re-boots be the result?

 

[Schmide]

Have you done the single drive boot with minimal peripherals? Could your AGP be sharing an IRQ with some of the drive controllers? Have you asked your BIOS to reload IRQs?

All errors are undocumented features waiting to be discovered.

 

[aae991]

In Win2k, AGP is on its own at IRQ 18. In Win98, AGP is sharing an IRQ. How do I ask the BIOS to re-laod IRQs in Win2k (or 98)?

 

[Schmide]

Its not in all BIOSes but its under PNP/ PCI Configuration/ Force Update E something something. I doubt that’s your problem since the IRQ was alone in Win2k. Are your chipset drivers up to date? AGP mini-port driver? AGP registry patch?

All errors are undocumented features waiting to be discovered.

 

[aae991]

I'll have to check that stuff. I'm at work now and the computer is at home

 

[o1die]

Have you or the builder changed the ti200 driver since replacing it? If that's the case, I would remove the current driver and re-install the same driver version you were using before the ti200 was replaced. Gainward drivers sometimes run at faster default settings than other drivers, and this can cause freeze-ups and other problems with some devices and games.

 

[Tiberius13]

Here is a silly question for you.... wtf do you mean by "the Ti200 crashed"????

How does a graphics card just 'crash' after running fine for months?

If it just suddenly 'died'... maybe whatever caused that sudden death (a power surge?) has damaged something else in the system? The ram, cpu, motherboard, ???

<font color=green><b>More salt than just a grain you will need with posts of mine. - Yoda©®</b></font color=green>

 

[aae991]

Thanks for all the responses...

The drivers for the Ti200 were version 2.1.83. When those did not work, I loaded the newest ones from NVidias' site. As far as the video card dying, one day it was fine, the next day lots of "dancing" horizontal lines started appearing on my monitor. They'd come and go, until finally I got a black screen. The system is hooked up to a 950W UPS that appears to be operating just fine - so there should not have been any power surge.

 

[aae991]

Okay, the dealer has found the problem... In spite of the 950W UPS I have hooked up, it appears a power surge occurred. The Enerex power supply has burn marks inside on the power cords. Here are my questions now... Could a surge occur that's powerful enough to cause burning inside a 465W power supply - even through a potentially faulty 950W UPS? If the UPS was bad, could it occur this severely. And one more question, if the initial surge killed the video card, would a cascade effect occur whereby the system becomes more and more unstable - leading to the frequent re-boots.

Either way, I think I'm looking at an expensive re-build.

P.S. I also learned a lesson the hard way - after laying out $2800 on the system, I bought a re-furbished UPS instead of coming up with the bucks for a new, albeit expensive, UPS with a guarantee. I feel pretty stupid right now! Now I'm going to pay an even bigger price!

 

[Tiberius13]

*OUCH*

I had a hunch something like that happened because you said the graphics card just suddenly 'died' and then these weird problems started. I guess the refurbished UPS didn't provide a warranty/gaurantee to replace any products that got fried while hooked up to it eh?

It sound to me like you should just not use it... heck, I never have issues and I don't even use a UPS... just a cheap power bar that supposedly has surge protection.

As far as getting your system working again. Hopefully it is just 1 other component that was hurt by the apparant surge. If replacing the power supply gets everything working fine again (might be a warranty fix on the psu?) you are off to the races!

If you still have problems, then it's time to start swapping components. The real kicker is if more than 1 component is causing your problems..... but for now you are best to assume that isn't the case! (the other way to test each piece of your system other than swapping one piece of it at a time to see if that's the weak link would be to put one piece of the wonky system into another stable system and see if it remains stable... if you test each piece that way it should identify any problematic hardware)

Best of luck!

<font color=green><b>More salt than just a grain you will need with posts of mine. - Yoda©®</b></font color=green>

 

[Guest]

Find a friend who can loan you some memory and try swapping that out first.

Just the fax ma'am!

 

[lhgpoobaa]

some power surges can be amazingly strong
lightning strikes for example...
tough luck on your part, hopefully the warantee on the ups can protect u...
though if u read the fine print i bet there is some exemption clause

Morally destitute, Emotionally bankrupt but a proud and respected member of Toms Forums!

 

[aae991]

Once again, thanks for the responses. When I got to the dealer, he showed me that the power cable from the psu to the motherboard had fried on the connector end with 2 of the pins - the 5 volt pins. Knock-on-wood, but with a new Enermax 465W psu and good incoming voltage, everything looks good. I did a render in the 3D landscape generator program Terragen - which used to cause a rapid re-boot - and it finished without a hitch. The best guess now is not a surge, but an under-voltage issue with over-loaded circuits. I'm going to move around the power requirements in my computer room and buy a UPS with voltage regulation. Probably a Tripp-Lite with 940W.

Thanks again for your advice.

P.S. The original Ti200 was not fried after all! Once the Enermax was compromised, nothing was getting enough voltage! I'll gladly pop for the new psu if everything else is okay!!!