AS8 (Random) Reboots

Hiya team

Fresh build with my first deviation from Asus for a few years, and after 2 clean installs of XP-Pro, I am still getting random reboots. Normally a couple a week, but in the hope of pinning it down, it won't even get through Prime at everything default.

It's an AS8 with a 550cpu, XP90, 4 x sticks of OCZ4200, a Sata 120 with the O/S and a Maxtor 300 with the data and a TDK DVD burner. Oh... and an X800XT-PE card and an Antec Neo480 psu.

I have upped the voltages in BIOS and have the following readings using Uguru...

CPUVCORE 1.36 to 1.4
DDRDD 3.01 to 3.08
DDRVTT 1.53 to 1.56
12V line 12.28 to 12.45
FSBVTT 1.23 to 1.25
NB/AGP 1.64 to 1.68
GMCHVTT 1.26 to 1.27
5V line 5.04

cpu is sitting at 204Mhz
AGP/PCI locked at 67/34

Temps
cpu at 43ish
system at 21
psu at 35

Installation process...

Install from CD
Install inf file
Install drivers (incl Ati version 6.14.10.6553)
Install software
use PC

Any thoughts... even where do I begin, would be great.

This thing has heaps of potential... as you see from my sig, I have had it (not Prime stable) running at 4.0 for normal stuff, so I am a bit stumped.

Hello
What kind of powersupply are you running?

Fordman

A/A... Antec Neo 480 (I wonder about the psu temp at 35... quite high for a psu? I don't know, but it feels warmer than others I have had)

edit... I have replaced the Antec with a nice shiney new Enermax Noisetaker 600W and this has replicated the same problem, plus

cpu & psu temps have also gone up, so I guess the 1 x 120mm fan in the Antec is better than the 2 x smaller ones in the Enermax LOL

cpu is a good 5 degrees hotter, and when I am priming, not only does the cpu get a lot hotter (like 60 - 65 degrees... never got above about 55 before), but the psu gets up to about the same. Case stays down at about 25-30 degrees.

Interesting (plus would value any ideas... I have plenty of memory to try and a 660 cpu on the way, but other than that... might be looking for a newer board)

have you run just 2 sticks of the OCZ?

Nope.. will try that though and get back later.

Troubleshooting random problems are a pain in the buttocks. Perhaps I would start with running and memtesting one stick of RAM at a time.

Is there a reason you are running the NB/AGP and RAM voltage so high?

Thanks for that... will try memtest and see what that produces.

Re the voltages... my understanding was that OCZ thrived on volts (up to 3.0 anyway) and a slightly raised AGP helped a good overclock.

I had been running this at 4.0 (236Mhz from memory) so had tweaked those a bit.

I now have a 660 cpu on the way, so want things right for that, so may well do another clean install, start testing with memtest and work upwards from there.

Ok, guess I thought this was happening at default speed. Maybe it is now, but apparently the trouble began when it was overclocked. Ok, that makes a little more sense now. Looks like you have already tried reinstalling Windows which usually is my first suggestion. I've seen a few random reboots before and they could be a lot of different things, but they commonly tend to be the first signs of a failing motherboard or weak power supply. I didn't see what your 3.3v rail is showing, but your +12v rail is a bit higher than what I would expect. Although I wouldn't think that would cause a big problem.

Okay...

I dropped the NB/AGP back to 1.5 and the Mem back to 2.6 and the PC wouldn't even run memtest... just kept rebooting.

Wierd???

Put the voltages back up to what I call normal (NB/AGP at 2.6 and mem at 2.8) and managed to get her into dafe mode and copied all the necessaries off C drive to D drive, and then thought, bugger it... I'll try a normal boot, and here I am working on her as normal.

Most peculiar to say the least.

Sounding more and more like a hardware failure. Likely suspects are a RAM module, motherboard, or PSU.

Thanks

PSU has been changed (was Antec Truepower 480, now Enermax Noisetaker 600) so I think we can rule that one out.

Incientally, did a BIOS update (the sort from windows where it just does it all by itself) and the temps dropped right back to normal (like 36 - 40 idle) which makes me suspect Motherboard. Still wouldn't Prime though.

I will work through this afternoon one stick at a time trying memtest and see how it goes. I might also try a lighter weight video card. I have a 9800Pro here for a few days, so will see if that makes a diff.

Just checkin prices here for a P5WD2 Prem board, so might just be forced to make the move LOL

Okay...

Tried single sticks... no diff.
Tried different sticks (Corsair)... no diff
Tried different video card... no diff
Tried removing and reseating the cpu (found I had put way too much AS5 there LOL so took a heap of that off and the temps have dropped right back to 32 deg idle, which is more what I am used to.

Managed to get it to run DocMem 2.0 and it came up with an error almost straight away...

"Run time error M6103: Math - floating-point error: divide by 0"

Did a google, but couldn't find anything meaningful... any thoughts?

can you test the CPU with something like Prime or CPU Burn-In?

Well ain't that the craziest thing.... (I'm on the lappy But) she is priming fine as I type this. Cpu temp at 46deg. The unhappy thing is the Enermax Noisetaker is at 64deg, which cheeses me off a bit. System is down around a constant 27deg.

Surely an excess of AS5 wouldn't have given it all that grief? The temps (unless the AS8 reads real low) never got above 65ish and I have used throttle-watch most of the time, and it never showed up as throttling, and also, it doesn't explain the reluctance to run memtest/docmem, and the error that docmem got straight away.

I'm rambling (but happier)

I took a screenshot with Uguru, Prime & Throttlewatch all on-screen, just need some guidance as to what to do if anyone needs to see it.

Thoughts? What next? (It's still priming away)

EDIT... and then it failed again... here is what is in the "stress-test" file...

[Wed Sep 14 23:11:40 2005]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Thu Sep 15 12:51:05 2005]
Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.

Intel(R) Pentium(R) 4 CPU 3.40GHz
CPU speed: 3483.70 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.14, RdtscTiming=1
Best time for 512K FFT length: 13.767 ms.
Best time for 640K FFT length: 17.666 ms.
Best time for 768K FFT length: 21.454 ms.
Best time for 896K FFT length: 26.269 ms.
Best time for 1024K FFT length: 29.830 ms.
Best time for 1280K FFT length: 36.658 ms.
Best time for 1536K FFT length: 44.456 ms.
Best time for 1792K FFT length: 53.750 ms.
Best time for 2048K FFT length: 59.543 ms.
Best time for 2560K FFT length: 78.496 ms.
Best time for 3072K FFT length: 94.863 ms.
Best time for 3584K FFT length: 114.407 ms.
Best time for 4096K FFT length: 126.867 ms.
Best time for 58 bit trial factors: 7.864 ms.
Best time for 59 bit trial factors: 7.868 ms.
Best time for 60 bit trial factors: 7.859 ms.
Best time for 61 bit trial factors: 7.892 ms.
Best time for 62 bit trial factors: 11.017 ms.
Best time for 63 bit trial factors: 11.019 ms.
Best time for 64 bit trial factors: 12.539 ms.
Best time for 65 bit trial factors: 12.812 ms.
Best time for 66 bit trial factors: 12.838 ms.
Best time for 67 bit trial factors: 12.603 ms.
[Thu Sep 15 13:14:21 2005]
Self-test 1024K passed!
[Thu Sep 15 13:30:06 2005]
Self-test 8K passed!
[Thu Sep 15 13:47:18 2005]
Self-test 10K passed!
[Thu Sep 15 13:55:35 2005]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Thu Sep 15 17:31:09 2005]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Thu Sep 15 17:45:53 2005]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

Classic signs of an unstable processor core. The motherboard may be undervolting the Vcore. It's also possible that the processor chip is bad.

What are you memory settings? Make sure that Command Per Clock (CPC) and PAT are both disabled, and check your NB Strap (under the uGuru section) and set it to 800.

Be aware, this motherboard does have some vdroop issues, though they shouldn't affect you (especially like this) at stock.

RJARRRPCGP said:

Classic signs of an unstable processor core. The motherboard may be undervolting the Vcore. It's also possible that the processor chip is bad.

Click to expand...

Vcore ranges between 1.32 & 1.40 under load (idle is 1.32)

johan851 said:

What are you memory settings? Make sure that Command Per Clock (CPC) and PAT are both disabled, and check your NB Strap (under the uGuru section) and set it to 800.

Be aware, this motherboard does have some vdroop issues, though they shouldn't affect you (especially like this) at stock.

Click to expand...

I'll check & get back tomorrow (it's pretty late here right now) B

EDIT... just realised which thread this was.... LOL... I ditched the AS8 and went for a nice new P5WD2-P... runs my 660 fairly well I'd have to say.