Bad ram?...or something else

[psykoon]

I just built a new PC-

Windows XP 64bit SP2
Antec 430W Earthwatt
Intel e2180 with AC Freezer 7 Heat Sink
Gigabyte P35-D3SL rev 2.0 with F7 Bios
Corsair XMS2 TWIN2X2048-6400
Nvidia GeForce 8800 GS
Seagate 250GB HD (SATA)
IDE DVD Burner
120mm Fan

I ran Prime95 before overclocking any thing and I kept getting "FATAL ERROR: Rounding was 0.5, expected less than 0.4" within minutes of testing.

I went into the Bios and set the memory to 5-5-5-15 and upped the DDR2 voltage to +.2V. Also, I disabled Legacy USB as I read that could cause issues.

Ran Prime 95 again and still got the same errors.

After this I ran Memtest v2.01 and got errors during test 5 and test 6. I tested each stick individually. On each pass one stick consistantly had a bunch of errors on test 6, the other had errors on test 5 and a few on test 6. (See images)

Errors on Test 6
Errors Test 5 and 6

I tested each stick in each of my motherboard's memory slots and the errors were always the same on each pass.

Does this mean my memory is bad and needs to be RMA'd? or is there something else going on?

 

[DSF]

Is there any way you could get another stick of known good memory to test with?

 

[psykoon]

Is there any way you could get another stick of known good memory to test with?

I wish I could...I dont know of anyone who has an extra stick lying around.

I was considering buying a stick of this from newegg Kingston Memory

Its cheap, on the approved list of memory for my motherboard, and would hopefully help me determine which component is at fault.


I think I've narrowed it down to either the memory or the motherboard. I dont believe its the power supply (although I have not put a multimeter on it because I'm not entirely sure how.) I think the CPU and cooling is fine because I ran the Prime95 small-fft test for a few hours and did not get any errors.

I suspect it is the memory because each stick has its own distinct set of errors with every pass through memtest. Also, these errors always occur no matter which slot I put them in.

It seems to me that if it was a motherboard issue...the errors would be more random and would also change depending on which slot I put them in the motherboard. Then again I am still new to this so I am not sure what to think.

 

[LOUISSSSS]

have u tried to run MEMTEST86+ v2.0, HCI MEMTEST, and Prime95 v25.6 with that memory DIMM in a different slot? it can just be a bad slot, maybe up your MCH (NB) voltage by 0.10v

 

[Yellowbeard]

Based on what you have posted, we'll RMA the memory for you if you want to try that. If you cannot test on another machine or borrow some known good memory your options are limited. However, we rarely have 2 bad modules in 1 set which makes me wonder if it is another issue.

Let me know if I can help.

 

[psykoon]

have u tried to run MEMTEST86+ v2.0, HCI MEMTEST, and Prime95 v25.6 with that memory DIMM in a different slot? it can just be a bad slot, maybe up your MCH (NB) voltage by 0.10v

Yes I tried all that. It doesnt matter which slot the memory is in. One DIMM always gets errors during test 5 and couple in test 6 (with error bits always at 00008000), while the other always gets errors during test 6 (with error bits at 00000080). I didnt try increasing the MCH voltage as I thought that wasn't necessary until I started increasing the FSB.

If you cannot test on another machine or borrow some known good memory your options are limited. However, we rarely have 2 bad modules in 1 set which makes me wonder if it is another issue.

Yea unfortunately, almost everyone I know either has a computer that 5+ years old and doesnt use DDR2, or they have a laptop. Also, as I said before, it seems like its most likely the memory cuz each DIMM has a different set of errors. If it was another component, I would guess that the errors would be similar for each stick of memory.