• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Memtest+ flawless unless immediately after blue screen?

C

Ceromus

n00b
Joined
Oct 5, 2007
Messages
13
Running:
E8500 currently running at stock speeds (problem still occurs), was at 3.8ghz voltage was only at 1.275.
MSI p35 platinum motheboard
tried both Patriot and Crucial Ballistix.
gefoce 8800 gt
System is about 6 months old. Problems started occuring last week.


I have been trouble shooting this PCs problems for a few days now. Still trying to narrow it down could use some help here.

This is a little unexpected so hopefully one of you will recognize these symptoms.

My computer has had various blue screens in the last few days. Mostly its MACHINE_CHECK_EXCEPTION or CRITICAL_STRUCTURE_CORRUPTION. It passes memtest with flying colors...or so I thought! I decided to run orthos over night and I set it to reboot upon bsod. I also inserted the memtest cd so as soon as it rebooted it would start running memtest. Well as soon as it crashed and memtest launched it spammed hundreds of errors. Powered off, turned back on, passed memtest with flying colors.

Any ideas here? Thanks. My next step will be to try a different CPU.



Incase you happen to be able to debug MCE's the parameters for the MACHINE_CHECK_EXCEPTION was:

Arg1: 0000000000000005
Arg2: fffffadfc8c71240
Arg3: 00000000b2000030
Arg4: 0000000008000e0f

and

Arg1: 0000000000000001
Arg2: fffffadf90c71240
Arg3: 00000000b2000000
Arg4: 0000000006000e0f

Update:
Sometimes file checks such as MD5 will come back incorrect but when I reboot they are fine. What could cause that?

I recently recieved another blue screen:
DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)
An attempt was made to access a pageable (or completely invalid) address at an
interrupt request level (IRQL) that is too high. This is usually
caused by drivers using improper addresses.
If kernel debugger is available get stack backtrace.
Arguments:
Arg1: fffffade9c8b1018, memory referenced
Arg2: 0000000000000002, IRQL
Arg3: 0000000000000001, value 0 = read operation, 1 = write operation
Arg4: fffffadf8eeac11e, address which referenced memory

referencing inspect.sys which is comodo. I have had comodo for a while now though.
 
Last edited:
thats usually a sign that your voltage on your cpu is to low.. has nothing to do with memory..

i would bump the cpu voltage up 1 notch see if the problem goes away.. if that fails to work bump your nb voltage up 1 setting see if that helps.. but its not the memory.. usually you will see an IRQ less then/or equal error or something similar to that..
 
This is not a recently built system though.

Also im running stock settings right now. So its the CPUs default speed and voltage. Shouldnt really have to raise the voltage on it unless somethings failing. I may give it a shot though once I determine what part of the comp is acting up.

What aspect of what I said has nothing to do with memory? Those blue screens do not eliminate memory as a culprit to my knowledge. However since ive tested another memory its pretty safe to say its not the case. Im thinking something is wrong with the chipset.
 
[Testing the RAM]
Download Memtest86+ v2.11 or whatever the latest version is, unzip it, burn the ISO file to a CD, and then boot from it, just like you would do with the XP/Vista install CD. Let Memtest+ run for at least three hours on each stick of RAM separately as well as test the RAM all together. Go for a full 24 hours if you want to be completely sure that the RAM is not a problem. If you start seeing errors, than your RAM is defective or you have incorrect settings for the RAM.

[Testing the Hard drive]
Download the CD image of Hitachi Drive Fitness Test, burn the ISO file to a CD, and then boot from it, just like you would do with the XP/Vista install CD. Test the hard drive and see if any problems are found. DFT will run on most manufacturers' hard drives. Alternatively, you can use Seagate's SeaTools for DOS to test a Seagate or Maxtor drive. For a Western Digital drive, you could use Data Lifeguard Tools for DOS to test a Western Digital drive.

[Testing the CPU]
Use Prime 95, OCCT, Orthos or Intel Burn Tool to stress test the CPU

[Testing the GPU]
Use Furmark or ATI Tool to stress test the GPU

[Testing the PSU]
Buy and use a digital multi-meter to monitor the voltages coming from the PSU. Follow the instructions for "Using a multi-meter to check voltages": http://www.bfgpower.com/troubleshooting.html#DMM
 
Sounds like a simple case of memory overheating to me. Seems to me that the heat from your cpu heatsink is 'spilling' onto your ram sticks making them hotter and causing them to fail.

Memtest+ isn't particularly good at stressing ram under real world conditions (like during gaming) where the whole system gets hot and consequently the ram gets hotter. Memtest doesn't really push the cpu so it doesn't get particularly hot and so your ram isn't exposed to real world conditions. AFAIK memtest only uses 1 core.

I've seen many cases where memtest will run happily, but games and windows based memory testers will fail.

Run prime95 in blend mode for an extended period and see if you get any errors. Use the latest version of prime95 that will do proper multicore threading.


p.s. if Prime95 fails in blend, your northbridge could be responsible, not necessarily your ram. I'd underclock your ram if it fails to rule it out and work from there.
 
I have never heard of testing each stick of ram individually. What would the purpose of that be if both sticks together won't error aside from some unknown condition? In my experience dual channel brings out errors in situations where they go unnoticed in single channel tests.

I have done tests on the hard drives.

I run orthos.

The layout of my board and the type of heatsink I have would make it pretty much impossible for the heat to be spilling out into the heatsink like that. I havent completely ruled out memory even though I have tried different sticks I could simply be experiencing multiple problems and was thrown off my crashes with the other ram.

I havent really crashed in games yet im sure it will happen though.

I will knock it out eventually...i hope :confused:
 
Run memtest when your computer's nice and cool. Take a blow dryer set on low and direct it at your CPU, NB, and RAM. If you start getting errors, you know it's a heat problem.
 
This ended up being 2 issues.

1. My CPU has become less tolerant for the OC that I had it at.

2. Unknown memory issues. When I switched to Corsair Value XMS memory at 1.9Volts its fine. However the previous memory both Crucial and Patriot work just fine in the other computers (so far). Other comps are p45 chipset this one is p35.

I am still uncertain about the situation regarding the reboot and excess memory failure in memtest. I wonder if the voltage setting for memory in the bios is accurate.
 
You must log in or register to reply here.
Back
Top