Jump to content

Random reboots with Ultra B


Recommended Posts

Hey all, just got this rig (for free!) from a friend and planning on running it as a lan server, but having some issues with random reboots.

 

Here's a system snapshot:

 

DFI Lanparty NF2 Ultra B, Tony's 11/24 BIOS

AMD Barton 2500 Mobile IQZFA

Arctic Cooling Silent 3 Copper HSF

OCZ 2x512MB PC4000 Gold Rev 2

ATI Radeon 8500 Silent Heat Pipe

2x74GB Raptors RAID 0

Antec SP400W Power Supply

Windows XPP SP2

5.10 Driver Package

 

Settings 250x10, 1:1 2.5-3-3-7 1T

 

My problems are setting independant, as even stock settings work fine. Temps are also fine. HAL is set correctly via first install. System is rock solid for days on end, then will juat reset randomly while sitting there idling.

 

Any ideas? I've searched a bit and read a few peoples threads with no luck.

 

Cheers.

Share this post


Link to post
Share on other sites

does it pass memtest and prime95???
Yeah - it's more than 24 hours prime95 & memtest stable... but every so often it hiccups and reboots. I've also tried an OCZ Powerstream and Seasonic S12 PSU with the same problem...

 

Beginning to get a little frustrating!

Share this post


Link to post
Share on other sites

That's not a bad looking setup.

I don't know what type of chips are on that ram though but hey ^^.

 

The only things wrong I see in that description are:

Psu

Drivers

 

5.10 sw ide drivers are bad.

So don't use sw ide's for now.

 

That antec psu sucks, no offence intended.

 

Does it reboot randomly at 200mhz fsb at say 10x multi?

 

Use mps ver 1.1.

 

For the super bypass mode stuff, the 2 settings, ensure both are enabled when using 1:1, it will help keep the 2 sticks in sync with each other.

 

The vcore...

 

-->

Prime95, Memtest86+, SuperPi, S&M.

 

Check all of these.

I have seen issues not caught by all of them, in that case, prime can catch it, but it might take along time.

Run prime just over 48hrs, 3 instances of it at the same time is my preference.

One instance is just to easy... (to run 3, make 3 prime95 folders and then you can run all 3 at once)

 

Use custom fft's in prime, inplace mode.

I would use 1024-4096k as a start.

then I also use 1024-1024, and 4096-4096.

 

If you wanted to, you could run say 256, 512 and etc, for ffts.

 

Run s&m's fpu test, in loop length, for 3 loops, at 100% load, with nothign running in the background(that's cheating if you run stuff to take away cpu time).

 

Run superpi 32m.

Then run 1m for about 30min over and over and see if it screws up(when lookig for oddities).

 

 

Dl speedfan or mbm, either take in account the max diode temp(by telling mbm to track it), or use speedfan to graph it.

Diode temps can reset your rig, thermal trip(not bios wise, that thing is useless, it's built into the hardware).

 

 

Oh, and stay away from a .5x multi for now to see if it's an issue with windows and it's power management.

 

 

 

Ediut:

Oh duh, what am I thinking.

Random reboots and such can come from a lose card, or forked up card.

 

Like one that changes it's device id in windows on random, that can doit too.

Clean everything...

Try diff cards temp like if possible or whatever.

Share this post


Link to post
Share on other sites

That's not a bad looking setup.
thanks

I don't know what type of chips are on that ram though but hey ^^.

Hynix chips - useless for my NF3' date=' but great on this board.
5.10 sw ide drivers are bad.

So don't use sw ide's for now.

SW IDE drivers not installed... Using SATA anyway.
That antec psu sucks, no offence intended.
For this system, the PSU is rock solid... actually less variation on every rail than the OCZ. However I thought this too, and tried the two other PSUs listed, with same results.
Does it reboot randomly at 200mhz fsb at say 10x multi?
Like I said, any multi, and settings, even stock, I get random reboots.
Use mps ver 1.1.
Yep.
For the super bypass mode stuff, the 2 settings, ensure both are enabled when using 1:1, it will help keep the 2 sticks in sync with each other.
No problems here.
Prime95, Memtest86+, SuperPi, S&M.

 

Check all of these.

I have seen issues not caught by all of them, in that case, prime can catch it, but it might take along time.

Run prime just over 48hrs, 3 instances of it at the same time is my preference.

One instance is just to easy... (to run 3, make 3 prime95 folders and then you can run all 3 at once)

 

Use custom fft's in prime, inplace mode.

I would use 1024-4096k as a start.

then I also use 1024-1024, and 4096-4096.

 

If you wanted to, you could run say 256, 512 and etc, for ffts.

 

Run s&m's fpu test, in loop length, for 3 loops, at 100% load, with nothign running in the background(that's cheating if you run stuff to take away cpu time).

 

Run superpi 32m.

Then run 1m for about 30min over and over and see if it screws up(when lookig for oddities).

A little excessive, but still, this system has primed over 40 hours at 250x10 without issue, running prime + superpi + mp3 encoding + video decoding without issues. I literally thrashed this thing to hell to try and make it fail, and it hasn't. However just when I think it's stable, I look at it wrong and pffft! it reboots.
Dl speedfan or mbm, either take in account the max diode temp(by telling mbm to track it), or use speedfan to graph it.

Diode temps can reset your rig, thermal trip(not bios wise, that thing is useless, it's built into the hardware).

no thermal issues here, I have the NF2 mbm data file, checked with speedfan + other, no probs whatsoever.

 

Oh, and stay away from a .5x multi for now to see if it's an issue with windows and it's power management.
Never use the .5s anyway, can't calc RAM speeds.

 

Ediut:

Oh duh, what am I thinking.

Random reboots and such can come from a lose card, or forked up card.

 

Like one that changes it's device id in windows on random, that can doit too.

Clean everything...

Try diff cards temp like if possible or whatever.

LOL, first thing I checked. Thought this old AGP card was corroded or something, but it runs rock solid on another system, so thats not it. I pulled every connector, cleaned with 99% alc, replaced and still same issue.

 

Have now tried:

- New CPU HSF (thought: fan rpm sensor fubard?)

- 1 stick RAM

- PCI Video Card

- Different BIOSes

- Different PSUs

 

Only other issue could be CPU? Not freekin likely considering it ran 100% 24/7 in another rig.

 

Errrghh!

Share this post


Link to post
Share on other sites

It looks like you have covered all bases, BUT when you have eliminated everything else, then what your left with.........................is usually the answer/problem ;)

 

i was just wondering what is Tony's 11/24 BIOS i have never come across that either???/????

 

Logan

Share this post


Link to post
Share on other sites

i was just wondering what is Tony's 11/24 BIOS i have never come across that either???/????
It's on Tony's BIOS CD (here) with the 5.0.64 Silicon Image ROM.

 

I tried pretty much every BIOS on that CD, and this one gave me best results overall. I got an extra 10MHz out of this RAM at the exact same settings and voltage, and in windows, all my benchies run just a little bit faster, inlcuding a much faster HDTach and Everest run.

 

Next thing... I solved the problem! You would never, ever believe this, but I got so frustrated that I stripped the entire system down to the bare board, including removing the bridge heatsinks - everything! I then cleaned the entire board, both sides, including PCI/AGP slots IDE connectors etc with 99% rubbing alcohol. I checked pins, double-checked jumpers and shorting blocks to make sure they were properly placed and not damaged, and basically went to town on it. I then reapplied everything with ASC/AS5 and bench assembled it... And voila! No more reboots. I'm guessing the friend I got it off of did some fancy AS5 work or something that was intermittently shorting a contact on the board/CPU/socket/etc. But now she's fine! Plus it gives me the opportunity to redo all my wiring when I put it back in the case.

 

Thanks all...!

Share this post


Link to post
Share on other sites

I love it when someone just hangs tuff on a problem and solves it,instead of giving up and just blame the MB.
Thanks brother! Yeah, it feels good to figure something out on your own. I've never had a bad hardware experience (knock on wood), so I've never been one to blame the hardware. In my experience, problems are FAR more likely to be user error than hardware-related.

 

I know you used isopropyl,but what material did you use to clean w/that didn't leave fragments
I used a lint-free fiber cloth - they're made from some sort of wood/pulp material, are ultra-soft, and are awesome for electronics as they don't leave anything behind. You can get them from Canadian Tire here in Canada. Highly reccomended. Alternatively, true terrycloth towels (the small ones you get for your car) are also generally lint-free. You can buy bags of these usually for a few bucks.

 

Peace!

Share this post


Link to post
Share on other sites

Ahh i see, i never new Tmod's name was Tony, there you go learn some new stuff everyday. COngrats on Getting the reboot sorted, no need to say no more, Playah said it all already, agree 100% with that, it usually something other than hardware most of the time. Gluck man.

 

Logan

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...