View Full Version : MEMTEST-86

19-12-2004, 04:50 PM
Hmm... after doing a fresh install of Win2K SP4 on a new 120GB 8MBcache HDD I started getting some random BSODs and eventually one on startup which basically screwed the OS - upon rebooting I also got a corrupted looking POST screen!

Suspecting a RAM or HDD problem I ran MEMTEST-86 which returned errors on all three of my modules (1x 512MB DDR333, 2x 512MB DDR400) and also crashed the computer part way through the tests.

I downloaded the SeaTools bootable ISO to check the HDD - the system memory check passes and so does the HDD but reports all my NTFS partitions as failed - Im doing a FULL diagnostic test at the moment...

Any comments or ideas about this would be appreciated. Is my chipset/mobo dying?

I had no problems until I put in the new HDD and 2K - and this isn't an extra HDD I swapped out a 40GB, I'm not overclocking the CPU or RAM and voltage rails and temps are stable!


Ash M
19-12-2004, 05:13 PM
Sounds like a mobo problem to me? Have you tested any of the other components on another box?

19-12-2004, 05:18 PM
Not yet - I'll be pissed if I have to replace it as the mobo is only a year old and it's an ASUS which I thought would be quite reliable...

19-12-2004, 05:41 PM
I would heed the advice of memtest86 myself, although it could very well be a motherboard problem. (You may have the black sheep of the lot.) Then again it is unlikely that all modules would be at fault. I recommend trying one stick at a time, if you haven't already done so, and retesting, and also if possible either using other RAM or using the same RAM and testing it in another system.

If it was all the RAM that was faulty, I would not be suprised that Seatools showed other random errors - even though it showed none on the memory side because it doesn't conduct such an extensive test.

19-12-2004, 05:42 PM
Reset your BIOS settings and check any jumpers on your mobo. Its possible something is being overclocked by accident.

19-12-2004, 05:53 PM
I have been testing individual modules and they are all reporting errors - into the hundreds of thousands and the result is that MEMTEST usually crashes before the test completes...

I have repeated this with the IDE devices unplugged and get the same results.

I will have to check the modules again with MEMTEST on a mate's PC tomorrow - I would rather replace the RAM modules under warranty that get a new mobo.

I will reset the BIOS but I'm pretty sure everything is AUTO/SPD.


19-12-2004, 06:20 PM
I'd test the memory in another PC to see whether the sticks were def faulty or not.

If they pass on another machine, it would indicate problems with the mobo.


19-12-2004, 06:26 PM
Yeah cheers for the comment Lo...

I will probably do that tomorrow - with the errors being generated completely randomly and on all modules and HDD related probs it will probably end up being the mobo unfortunately...

The SeaTools FULL diagnostic test did not complete and crashed with video signal lost - there have also been various POSTing problems - all random - I will probably get a A7N8X-X if I need a new one.

20-12-2004, 07:29 PM
Just an update:

I have tested my DDR400 modules on a mate's PC and they returned zero errors - this leads me to believe that my chipset or something on my mobo is dying unfortunately.

I have decided to get a GA-7N400-L (nForce 2) mobo as it is equipped with a AGP/PCI frequency lock and dual channel DDR400 ;)

21-12-2004, 06:18 PM
Well I got my new motherboard and hooked everything up without issues.

The motherboard detected my cpu as a 1250MHz Athlon? I thought that was weird but manually set the FSB in the BIOS and CPU Multi on the mobo.

After two Windows 2000 installations got corrupted I figured that the cause must have been something else and the only thing left is the CPU...

The temperatures for the CPU monitored in the BIOS is up and down and continues to increase up to about 76-80 degrees! until the PC locks up.

Keep in mind this is the second Barton 2800+ I will have to return under warranty! Must be a bad batch??!

I have been running it on stock voltages and clock settings so it's a bit of a joke really although I'm not laughing.

21-12-2004, 08:09 PM
Have you got your fan installed the right way around? I had similar issues a while back... the guy I bought the 2nd hand pc from had the heatsink and fan installed the wrong way round...

Anyway, nothing a little arctic silver and a thermaltake volcano couldn't fix. Much quieter too :)


21-12-2004, 08:15 PM
I wish it was that simple - I run a Thermaltake 3in1 cooler - it's a massive copper beast and it's definately on the right way - I am also using a thin layer of average thermal paste.

The last Barton went the same way - ran normal temps for ages then one day just started increasing and no amount of fan power would change anything - the last one cooked itself - this one just seems to freeze the PC when it gets hot enough (I assume the BIOS is halting the system). I've watched it in the BIOS just continually increase until a freeze.

Right on Christmas too - grr...

02-01-2005, 09:52 AM
Hi, Happy new year to all.

Just came across this thread as im having identical problems to yourself. I have a Shuttle SK43G with Barton 2500. After several failed attempts of getting memtest to complete I discovered that for some reason (which i have yet to discover) the bios controlled fans in my case, including the cpu fan stop working as soon as memtest starts, resulting in crashes and errors etc. Could this be the cause of your memtest problems ?

02-01-2005, 11:29 AM
Hi noob,

I've tested every component in my PC and have come to the conclusion that it's just another bad CPU - It worked for a while on low frequencies (100MHz FSB) but now just freezes the comp at POST at different stages - the CPU itself is physically looking fine though which is weird and I got MEMTEST to complete a few times so its pretty random...

Another good point is to check for motherboard pins shorting on the case which would cause lockups but everything looks fine to me and I have tried two different motherboards - in your case I doubt this is the problem as the Shuttle would use a proprietry motherboard installed in the factory I assume.