Memory issues or no?: Mem Test results. HELP!

[Bookcase14]

First, here are my system specs:
AMD 64 athlon xp 3800+ 939 pin
Gigabyte k8nsnxp-939 (socket) Nforce 3 neo 2 250 motherboard
Mushkin pc3200 DDR400 dual channel RAM (2 sticks of 512)
Saphire Radeon x800 XT PE video card

Now the stats for the Mem Test program: (latest version)
Test 5: 90 errors
Some failing adresses were 0002b240958
0002b240998 at 690.0MB
Good was fdfffff39c
Bad was fdf7ffff
All errors on Count Chan 1

Test 6: 8 errors
Some failing adresses were 00029a4039c
0002ee5039c
Bad was fff7fffe

My DDR is DIMM at rows 0 1

Mem test saw my specs at AMD Athlon 64 2411Mhz
L1 Cache: 128k 19766 mb/s
L2 Cache: 512K 4911 mb/s
Mem: 1024M 2192 mb/s
Chipset: Nforce 3 250 (ECC: Disabled)
Settings: Ram-200mhz (DDR401)/CAS:2-4-2-6/Dual Channel (128 bits)


Phew...now that that's out of the way, I have had about the same ammount of errors (sometimes 140 errors) on 5 different tests. I decided to test my Ram when I started getting Machine_Check_exception BSOD when I tried to install the Sims 2 (happened consistently for 5 install tries, finally worked when i switched Hard drive it was installed on, as well as cd-rom drive)and sometimes I get crashes when I play Full SPectrum Warrior (It goes straight blue screen on me, no writing). Also, when i first tried to install XP, i consistently got "disk could not write" errors or soemthing like that, it finally randomly installed one day, after about 10 tries. Is this probably my RAM judging by the errors on mem test?Its custom built PC, i ordered the parts from respectable websites. It has been acting strange on and off.Any help would be appreciated, thank you.

 

[Navid]

It does not matter what kind of error you get from memtest. Any error suggests instability.
Are you overclocking?
Are you using tight timings for RAM?
If yes, you should reduce the overclock, or relax the timing.
If you are not overclocking and the RAM timings are relaxed already, something is wrong. You can try increasing the RAM voltage (just one step at a time).
What is the RAM voltage?

Use Motherboard Monitor to check your CPU and motherboard temperatures. What are they when under load and when idle?

Edit:
This is based on the assumption that the CPU is stable.
If the CPU is not stable, that can cause memtest to fail too. Even if the memory has no problems.

 

[Bookcase14]

I'm not overclocking anything on my computer. I called up Mushkin, they said to try upping the voltage +.3 or +.2 to 2.8 ( I think it is now) SO thats what its at now, but its been that way for about 1 week, and i still get the same issues so... What do you mean by tight timings? I don't really know how to change that, my CAS is stated up top CAS:2-4-2-6

 

[Navid]

Have you selected those numbers (2-4-2-6) in the bios or are they automatically set up by bios?
You can reduce those numbers to speed up RAM. But, that reduces the stability margin of the RAM.

Use Motherboard Monitor and check the temperatures.

 

[Navid]

Dowload CPU-Z from the link below and run it.
http://www.cpuid.com/cpuz.php

Check under the "Memory" tab and see what the timings are set to.
Look under CPU and see what the CPU frequency is.

Edit:
As a first step, it is better to first make sure that the CPU is not the cause of the problem. That is why I suggested CPU-Z and Motherboard Monitor, which will provide more information about your setup.

 

[quentinterintino]

Wow, nice hardware.

I'm assuming here that you're using memtest86:

I'm also assuming that Rows 0 and 1 are DDR1 and DDR2 as labeled on the motherboard.

If you get those erros, try to isolate the problem via these methods:

1) Remove 1 stick out of DDR2 so you only have 1 stick in DDR1, run memtest again.

if error persists, either the stick or the module is bad. to determine that, switch the stick you currently have in DDR1 with the one you previously took out, run test again.

If error persists, it looks like that module is bad.

To further isolate the problem,

Move the sticks into DDR3 and DDR4 (works the same as 1 and 2, just different locations)

Repeat the same procedure as above, but instead you'll be swapping ram in DDR3.

If that test passes, then it looks like one of your memory modules is bad, so just keep them on DDR3 & DDR4.

Hope that helps.

 

[quentinterintino]

This might be clearer:

Taken from memtest86.com:

1) Removing modules
This is simplest method for isolating a failing modules, but may only be employed when one or more modules can be removed from the system. By selectively removing modules from the system and then running the test you will be able to find the bad modules. Be sure to note exactly which modules are in the system when the test passes and when the test fails.

2) Rotating modules
When none of the modules can be removed then you may wish to rotate modules to find the failing one. This technique can only be used if there are three or more modules in the system. Change the location of two modules at a time. For example put the module from slot 1 into slot 2 and put the module from slot 2 in slot 1. Run the test and if either the failing bit or address changes then you know that the failing module is one of the ones just moved. By using several combinations of module movement you should be able to determine which module is failing.

3) Replacing modules
If you are unable to use either of the previous techniques then you are left to selective replacement of modules to find the failure.

4) Avoiding allocation
The printing mode for BadRAM patterns is intended to construct boot time parameters for a Linux kernel that is compiled with BadRAM support. This work-around makes it possible for Linux to reliably run with defective RAM. For more information on BadRAM support for Linux, sail to http://home.zonnet.nl/vanrein/badram

 

[Smilin]

You can try a CPU tester or other software based hardware tester but you can only trust the positive (failure) results they give, not the negative. Realmode software based memory testers are the only thing that comes even remotely close to being accurate. Running a software based hardware tester is like giving a clipboard to some drooling guy in a loony bin and asking him to walk around and ask everyone who's sane.

Those memory testers will not give you a false positive though - if they say something's wrong then it is.

Do what quentin said and start trying to isolate a bad stick or bad slot. You can move your timings down or maybe underclock a bit too. You might just be running at the edge of a particular speed bin on one of your components.

 

[Bookcase14]

Thanks for the suggestions guys, I'll try them out sometime soon. But would you say its more likely the processor, memory, or motherboard? Thanks.

 

[Cheetah8799]

Originally posted by: Bookcase14
Thanks for the suggestions guys, I'll try them out sometime soon. But would you say its more likely the processor, memory, or motherboard? Thanks.

If you keep getting memory errors, then it's the memory. Only way to rule out the memory as being bad is to test it in another machine.

 

[Smilin]

I'll give you say 80/20 Memory or Motherboard. I think in this case CPU may be a longshot.

Memory is just more prone to problems due to the sheer number of transistors.

But really, who knows? Only some logical troubleshooting will help eliminate variables.

 

[Navid]

Originally posted by: Bookcase14
Thanks for the suggestions guys, I'll try them out sometime soon. But would you say its more likely the processor, memory, or motherboard? Thanks.

I think it is impossible to answer that question at this point.
But, if you try the different suggestions that you have got so far, you will get more data that can help narrow down the problem.