• 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 test 6 errors.. need some help

T

three4seven

[H]ard|Gawd
Joined
Oct 11, 2004
Messages
1,154
anybody know what memtest test 6 errors mean. I only get errors in test 6

For some reason, my computer which has been running fine for about 5 months starting resetting on me randomly...maybe.. once every week or so. I ran memtest and memtest only shows 1 or 2 errors about every other pass in test #6 only. I read about turning off legacy USB support and have done that already, but still yield the same results.

Specs:

x2 3800+ AC pro 64 (939)
a8n-e mobo
2x 160gb raptors RAID 0
2x 300gb hd
1x 250gb hd
480 watt antec truepower 2.0
7900gt co ac 5 rev 3
dvr 111d burner
2 sticks of twinmos 512mb (2.5-3-3-8 @ 201mhz)
2 sticks of corsair 1gb (2.5-3-3-8 @ 201mhz)

I though it would be the twinmos... so I removed one (from slot 4) and ran test 6 only... 2 passes, no errors, removed the other stick and put the other stick back in (also in slot 4)... 2 passes.. no errors... put them both in.... the setup doesn't recognize the twinmos ram at all and gives many errors in memtest.

remove the twinmos and put 2x512 ballistix ram in place of the twinmos... ran memtest ... 13 hours and 9 passes later... i get almost the same results. The ballistix are rated to run at 2-2-2-7 i believe, but i loosened the timings to match my corsairs at 2.5-3-3-8. Still getting 1 or 2 test 6 errors every other pass or so.

so my question is.... could it be my PSU not being able to put out enough juice? Faulty hd (ran scandisk, no errors)? Could it be my mobo? Could it be I need to up the voltage on the ram (running at 2.7v right now)

any insight would be greatly appreciated.
 
The problem is most likely the ram
Either get it RMA'd. slow it down via FSB or timings or do as you suggested, increase the voltage slightly.
Before RMAing, try it in another pc to make sure the fault persists.

To see if you are pushing the psu too hard, unplug anything you dont need and see if it improves.
The power supply for the ram is generated on the motherboard and isnt a great power drain unless something is faulty.
As other ram works fine, its unlikely to be a motherboard fault.

Some motherboards are sensitive to particular memory modules so its good to see if there have been any BIOS updates for the motherboard that fix memory issues.
Check to see if others have had similar problems with your memory and motherboard.
 
thanks for the reply... im trying each stick individually now and perhaps up the voltage to 2.75. I tried the twinmos on another system... no errors for 5 or so passes, but 1 pass shoes 12 errors in test 5.

i looked at the PSU... there should be no sweatin the power consumed by my compnonents.

x2 3800 is about 125w
7900gt is about 60w
at max, 2 raptors runs about 25w (read/write)
i assume 3 other hd's are about the same.. so 35 w

so PSU shouldn't even get close to reaching its limit, the TP2-480 is known to have good efficiency

going through each stick individually now (ballistix and crucial) to see which stick may be faulty... its a shame that it takes an insane amount of time for each pass in memtest.

thanks for you help nenu
 
alright.. im confused now... i just tested each stick for errors individually.... zero errors come up ... 3 passes each...

when all four are put together though.. it yields errors in test 6 only again...

failling addresses is diff each time, but it always fails on :

good: f f f f f f f b
bad : f f f f f f f f


i just cannot seem to figure out the problem... and its driving me crazy
 
If you have enabled 1T timing in the CMOS, this wont work for 4 memory sticks.
You must select 2T for 4 sticks to work.
 
im hopelessly lost..

nenu.. thanks again for ur help...

my config is already at 2t.

i've tried everythign.. isolated each stick. then i began removing all my hd's and peripherals until all i had was the cd drive, psu, cpu and vidcard... and it still gave me the error (1 or 2 errors in test 6)... i've tried swapping slots on all the dimms... no go... upped the voltages.. loosened the timings... everything...

no idea what it could be....
 
Sorry to hear the problem is persisting.

I had a similar problem when mixing ram on a friends motherboard.
Each pair of memory sticks would work fine but when put together to give 4 sticks, it was very unstable.
The culprit turned out to be the memory timings being quite different on the 2 pairs of memory sticks.

When the pc boots it will read one of the memory modules SPD and set the timings according to that.
If a speed setting on the chosen module is too fast for another module in the system, it can become unstable or not work at all.

The problem was a little more severe than that as we found that one pair worked best at rated speed and a little faster but became unstable when some settings were slowed down.
The opposite was the case with the other pair of sticks, they would work as rated speed and a bit slower on some settings.

We ended up testing the settings for a lot of timings for each memory pair and then found which settings worked for both pairs.
There was a narrow band of usable values for a few settings but it worked :)

Its a bit long winded but might be worth trying.
 
ok.. after taking a couple days off.. i was at it again trying to figure out what is wrong w/ my setup.... the freezing has gotten worse... the computer gets the BSOD within 4 minutes of booting into windows

i have tried all sorts of timings and tests, isolation... even tried out microsofts memory diagnostic... which pulled up stride6 errors once every 3 passes. The test 6 errors (maybe one or two every other pass or so) are persistent.

I've tried overclocking.. underclocking... loosening timings ... reduce ht multipliers... everything... nothing seems to work.

there was a single boot where the computer didn't get the BSOD for atleast 20 minutes until i turned it off... (mem was running at ddr200, and ht at 200mhz). I tried this setting again after testing variations of speeds... got the BSOD within 5 minutes.

I am absolutely clueless as to what to do here. .. I can't even pinpoint if its the memory anymore or if its some other piece of hardware causing the problem.

Any other things i should try out?. Please don't hesitate to let me know. I toyed with the idea that the CMOS battery perhaps is dying and is causing the instability (I have yet to test this theory). I ruled out PSU cause i tried pulling everything out except CPU, cd, vidcard, mem.. and still gave me the memtest 6 errors. By that method.. also ruled out, soundcard, raid ide card, firewire card, any of the 6 hds. Whats left is... mobo, vidcard, PSU (maybe), and the memory.

Thank you nenu for all the suggestions you have given me thus far. I greatly appreciate it.
 
I wouldn't imagine its the PSU. Usually if you get errors like that, it is either the RAM or mobo. Maybe you can test it out on a friends rig? Maybe you can swap out the mobo if you are lucky and try that.
 
In order to properly isolate the problem, you should try 1 stick at a time in the B1 (I think) slot...

If they all work, then it will get more tedious... you will have to start testing them in other slots... I am aware that you can only run 1 stick in certain slots with that board, so it might be hard to perfectly diagnose it
 
thanks for all the replies..

the memory seems to be fine.... I tried isolating the problem... tested each single stick in first slot (which i assume is b1) and ran memtest test 6 for three to five passes for each... all passed.. no errors.

when I put them all together though (2 x1gb corsair valueselect, 2x 512mb crucial ballistix or 2x 512 twinmos) it rings up errors.

as for swapping out the mobo... i would like to say this is the absolute last resort or if I find out it is absolutely the mobo. Swapping out the mobo will take considerable time and effort lol.

thanks again for all the suggestions.
 
do any of those 3 pairs work well individually?

if so, then it's not so much that anything is wrong as much as motherboards are very picky about what you put when you load 4 dimms.

that's why I didn't even try adding my 2x512's when I upgraded to 2x1024's...

surprised you even ran it for 5 months without a problem.

If you can run the 2GB corsair kit stable, then I'd say just do that and try to be happy.
 
i can't remember if i ever just tried a pair by themselves... the idea of only 2gb vs 3gb and letting my ballistix go to waste makes me sad .. haha... but i suppose you are right.. i should give that a try and see how it goes..

thanks
 
alrighty.. what i thought was okay ram.. was actually my corsair giving me the test 6 errors...removed that... and I found what I thought to be the culprit causing the memtest errors and system reboots....

but I was wrong lol.... slapped in some new memory... no memtest errors on either test five or 6 after about 10 passes.

few minutes after it gets into windows (every single time i boot windows)... i get the BSOD...

DRIVER_IRQL_NOT_LESS_OR_EQUAL

which after some researching has to do w/ memory?... but i am seriously doubtin its a memory issue this time... as it passed memtest 6-7 times w/o any errors. It occasionally crashes before it even gets into windows. but i haven't been able to capture the BSOD on that yet...

so it seems.. the memtest errors and the BSOD restarting issue was independent of each other...

::sigh::... this is depressing..
 
Hmmm...

Well if I was in your shoes, I would try 2 sticks of whatever RAM is working in Memtest perfectly... JUST TWO STICKS. if you have problems, try the other slots, and if it works then, it might just be one of your DIMM slots is problematic on your board.

If that doesn't solve things, try removing any unnecessary things from the system (i.e. extra HDs, soundcards, whatever) and reinstall windows.

See if everything is stable from there.

Really, the best way to prove it's your motherboard for sure is to try another power supply as well, and another hard drive if that's old too.

You may want to consider updating your BIOS to the latest version to see if that helps.

It's not going to be easy to know what is wrong for sure unless you have some spare parts lying around.
 
You must log in or register to reply here.
Back
Top