Question about memtest failure and some help

T

the_transporter

Limp Gawd
Joined
Aug 19, 2004
Messages
223
I have 4x1GB Corsair XMS2 (specs in sig) and I have been having trouble installing XP (won't read files). I ran memtest (boot from CD) and it looked horrible. It gave me errors on the second test and it crashed (had to manually press reset button). This was really scary. I took out two sticks of ram and still the same aforementioned error with memtest.

I have been reading lately about SATA DVD drives and how people were having problems with them as opposed to an IDE dvd drive. This is a really dumb question, but is it possible for a SATA DVD drive to cause memtest to produce errors (like it's reading the tests wrong from the disc)? Because I really find it hard to believe that my ram will just die for no reason. I ran memtest before on these sticks and they were fine. I just hope it's not my ram, but it sure does look like it :(. It's like I can't believe my ram (non-generic or budget type) is damaged so bad that it caused memtest to crash instead of halting when encountering an error.

Also, please note that PRIOR to me doing the above, Windows ran great. I played Crysis, ran 3dmark06 and no problems or crashes at all. Windows started to act up after I did some registry scans this morning and I think I had a virus, so that's what prompted the reinstall of windows.

Any idea with regards to this scenario?
 
Update:

I ran Memtest 3.4 and I think I downloaded a bad version because it gave me nasty errors. I downloaded Memtest 3.3 instead and it ran with no errors. What I want to know is, is there a possibility that running a bad version of memtest can damage my ram? I hope not. Thanks.

I ran Memtest outside of Windows, during system bootup
 
No, you cannot damage your RAM with Memtest. In fact, I would suggest testing each module individually for about 3 passes per module using Memtest. Also, make sure you are getting your Memtest ISO from www.memtest.org and get version 1.70.

It is possible your copy of Windows is scratched or blemished. Also, can you try it with a different optical drive?

And, while we're at it, please list all your bios settings you have set for memory, CPU, and all voltages you have set.
 
Just as an FYI ther are two versions of Memtest 86. The first and original version is Memtest 86 and the latest working version is 3.4 however I have noticed the bug that you talked about. Using 3.3 it takes almost 30 minutes to complete one full pass on the standard test.

There is also a Memtest86+ which is based off of the original Memtest86 but it is a little more streamlined and one full pass takes about 13 minutes. The latest version of this one is 1.7 and can be fount at this link: http://www.memtest.org/

Carter
 
Thanks for the replies.

The Memtest I got was off this website: http://www.memtest86.com/

Is the .org the official website for Memtest?

So I'm a little confused. It looks that 1.7 takes the less time, but I don't mind waiting. I just want to make sure all my sticks are clean. So basically, which would be the better tester to ensure this? 1.7 or 3.3? I'm assuming 3.4 is bugged, so I won't use that. I will test each individual module tonight when I get home. Also does it matter which mem slot I put it in? Can the mem slot cause errors? Or if there's an error, it's usually the ram and not the slot?

With regards to the Windows install, I was able to install it successfully this morning after I used an IDE DVD drive instead of the SATA DVD drive. My SATA DVD is probably dead after my 3rd use and 4 months of ownage :eek: I'm gonna throw it out or something.

Mike, these are my settings:

E4300 @ 2.9GHz (332x9) 1.35v
4x1GB Corsair XMS2 PC5400 running at 333, 4-4-4-12, 1.9v
 
the_transporter said:
Mike, these are my settings:

E4300 @ 2.9GHz (332x9) 1.35v
4x1GB Corsair XMS2 PC5400 running at 333, 4-4-4-12, 1.9v
Click to expand...
If you have any stability issues, you may need to increase your chipset voltages a bit due to the load of 4 x 1gb. However, at 333 you may be OK with stock.
 
the_transporter said:
Thanks for the replies.

The Memtest I got was off this website: http://www.memtest86.com/

Is the .org the official website for Memtest?

So I'm a little confused. It looks that 1.7 takes the less time, but I don't mind waiting. I just want to make sure all my sticks are clean. So basically, which would be the better tester to ensure this? 1.7 or 3.3? I'm assuming 3.4 is bugged, so I won't use that. I will test each individual module tonight when I get home. Also does it matter which mem slot I put it in? Can the mem slot cause errors? Or if there's an error, it's usually the ram and not the slot?

With regards to the Windows install, I was able to install it successfully this morning after I used an IDE DVD drive instead of the SATA DVD drive. My SATA DVD is probably dead after my 3rd use and 4 months of ownage :eek: I'm gonna throw it out or something.

Mike, these are my settings:

E4300 @ 2.9GHz (332x9) 1.35v
4x1GB Corsair XMS2 PC5400 running at 333, 4-4-4-12, 1.9v
Click to expand...

Memtest86+ 1.7 will tell you if you have RAM errors with one pass I would use that it has worked for me and I swear by it.
 
You must log in or register to reply here.
Back
Top