Memtest86 Crash at 7th test

Hi,

<prologue>

I do have some stability problem on my machine which randomly reset
itself or crash during windows boot up or while executing some
applications. I believed it may come from memory misbehaviour so i
downloaded and tried memtest86, last stable release, zip iso download,
booting it from cdrom.

<problem definition>

Results of my tests are as following.

First run: unattended memtest standard test ended with a shutdown, no
data collected about errors.

Second run: attended memtest standard test ended with a shutdown
during 7th test, no error reported so far.

Third run: manual launch of memtest tests 1 to 6 for a few cycles, no
errors reported at all.

Fourth, Fifth and Sixth run: manual launch of the seventh test, it
deterministically crashed at 3% completion.

<Conclusion>

While understanding that probably my ram may be working correctly and
the program, too, is fine and the crashes are probably due to the
stability issue i was trying to fix, i ask to anybody who has a good
comprehension of hardware and memtest functionalities to diagnosis my
problem.

Why memtest86 work correctly through the first six test and miserably
fail and crashes regularly during the seventh?
Which functionalities of my machine (cpu, cache, bus, other) are used
during the seventh test which are certainly not used during previous
tests?
Which component may be the responsible for the crash?

Thank you for your time.
Gabriele Palma.

wrote:
> Hi,
>
> <prologue>
>
> I do have some stability problem on my machine which randomly reset
> itself or crash during windows boot up or while executing some
> applications. I believed it may come from memory misbehaviour so i
> downloaded and tried memtest86, last stable release, zip iso download,
> booting it from cdrom.
>
> <problem definition>
>
> Results of my tests are as following.
>
> First run: unattended memtest standard test ended with a shutdown, no
> data collected about errors.
>
> Second run: attended memtest standard test ended with a shutdown
> during 7th test, no error reported so far.
>
> Third run: manual launch of memtest tests 1 to 6 for a few cycles, no
> errors reported at all.
>
> Fourth, Fifth and Sixth run: manual launch of the seventh test, it
> deterministically crashed at 3% completion.
>
> <Conclusion>
>
> While understanding that probably my ram may be working correctly and
> the program, too, is fine and the crashes are probably due to the
> stability issue i was trying to fix, i ask to anybody who has a good
> comprehension of hardware and memtest functionalities to diagnosis my
> problem.
>
> Why memtest86 work correctly through the first six test and miserably
> fail and crashes regularly during the seventh?
> Which functionalities of my machine (cpu, cache, bus, other) are used
> during the seventh test which are certainly not used during previous
> tests?
> Which component may be the responsible for the crash?
>
> Thank you for your time.
> Gabriele Palma.
>
>
It still could be a memory issue. I've found most of my problems have
shown up during tests 5 and/or 7. Sometimes increasing VDIMM .1 volt or
changing memory timings has solved it. If your running more than 1
stick try running the sticks individually using the slot closest to the
CPU. If all is OK, try a single stick in each of the slots. Since you
say it errors out on test 7 consistently, you should be able to run just
test 7 and get through it pretty quickly
Rob C.

> It still could be a memory issue. I've found most of my problems have
> shown up during tests 5 and/or 7. Sometimes increasing VDIMM .1 volt or
> changing memory timings has solved it. If your running more than 1
> stick try running the sticks individually using the slot closest to the
> CPU. If all is OK, try a single stick in each of the slots. Since you
> say it errors out on test 7 consistently, you should be able to run just
> test 7 and get through it pretty quickly
> Rob C.

I will surely follow your advice.
I'm using two slots (512+256) and i may enjoy more details about
modifying VDIMM tension and memory timings as i don't know anything
about it.
Do you have any additional information, link or references about it
(not too much technical, just a troubleshooting).
If may be useful in some way, my motherboard is a A7S333.

A couple of question, how should i troubleshoot if the problem is not
in the ram but in the first motherboard slot?
Would the pc start if i put one ram in the second slot and none in the
first?
One more, the computer is old and dust may be a problem, how should i
clean it to be safe from statics?

The strange issue here is that i thought that a program designed to
find memory error should be able to find and report it without
crashing, so if it crashes this should mean that the problem is
somewhere else, but i have no insight or experience on the topics.

Thank again. I'm going trying banks separately.

wrote:
>> It still could be a memory issue. I've found most of my problems have
>> shown up during tests 5 and/or 7. Sometimes increasing VDIMM .1 volt or
>> changing memory timings has solved it. If your running more than 1
>> stick try running the sticks individually using the slot closest to the
>> CPU. If all is OK, try a single stick in each of the slots. Since you
>> say it errors out on test 7 consistently, you should be able to run just
>> test 7 and get through it pretty quickly
>> Rob C.
>>
>
> I will surely follow your advice.
> I'm using two slots (512+256) and i may enjoy more details about
> modifying VDIMM tension and memory timings as i don't know anything
> about it.
> Do you have any additional information, link or references about it
> (not too much technical, just a troubleshooting).
> If may be useful in some way, my motherboard is a A7S333.
>
> A couple of question, how should i troubleshoot if the problem is not
> in the ram but in the first motherboard slot?
> Would the pc start if i put one ram in the second slot and none in the
> first?
> One more, the computer is old and dust may be a problem, how should i
> clean it to be safe from statics?
>
> The strange issue here is that i thought that a program designed to
> find memory error should be able to find and report it without
> crashing, so if it crashes this should mean that the problem is
> somewhere else, but i have no insight or experience on the topics.
>
> Thank again. I'm going trying banks separately.
>
>
According to the manual for that board you won't be able to adjust the
VDIMM any. The manual shows there are lot of different things to check
in reference to the memory such as if your operating the board in Jumper
Free mode and the frequency of your CPU vs. frequency of the DRAM. The
board shouldn't care which slots you use, so the first can be empty. It
is best to resolve the memory issue before booting into the OS so you
won't corrupt any data! If you don"t have the manual you can download a
copy from ASUS.com. If you do make any adjustments, timing or
otherwise, do them one at a time, so you can see if they impact
anything. The memory timing numbers should most likely be adjusted to a
higher setting, if need be.
With the cleaning, if the board is in the case make sure the PC is
disconnected from the AC source. Make sure you know where everything is
situated in case something gets knocked loose (always seems to happen)!
As long as you keep one hand on the case, so that your grounded to it,
you should be OK.
Rob C.

I have a P5W DH Deluxe and have FSB over clocked to 334, If I don't throttle
back my memory, I get all kinds of system problems, one of which is Windows
finding corrupt files on boot, but if I slack off on the memory speed these
problems go away. I have DDR2-800 and run it about DDR2-768, which on boot
is PC2-5300. Are you overclocking? Memory at defaults, if so try manual
settings. I've had the Windows memory test program crash on me about 3-4
test. Memtest86 will run just fine for me. Unless I throttle back.

<> wrote in message
news: oups.com...
>> It still could be a memory issue. I've found most of my problems have
>> shown up during tests 5 and/or 7. Sometimes increasing VDIMM .1 volt or
>> changing memory timings has solved it. If your running more than 1
>> stick try running the sticks individually using the slot closest to the
>> CPU. If all is OK, try a single stick in each of the slots. Since you
>> say it errors out on test 7 consistently, you should be able to run just
>> test 7 and get through it pretty quickly
>> Rob C.
>
> I will surely follow your advice.
> I'm using two slots (512+256) and i may enjoy more details about
> modifying VDIMM tension and memory timings as i don't know anything
> about it.
> Do you have any additional information, link or references about it
> (not too much technical, just a troubleshooting).
> If may be useful in some way, my motherboard is a A7S333.
>
> A couple of question, how should i troubleshoot if the problem is not
> in the ram but in the first motherboard slot?
> Would the pc start if i put one ram in the second slot and none in the
> first?
> One more, the computer is old and dust may be a problem, how should i
> clean it to be safe from statics?
>
> The strange issue here is that i thought that a program designed to
> find memory error should be able to find and report it without
> crashing, so if it crashes this should mean that the problem is
> somewhere else, but i have no insight or experience on the topics.
>
> Thank again. I'm going trying banks separately.
>

On 3 Mar 2007 06:06:31 -0800, wrote:

>Hi,
>
><prologue>
>
>I do have some stability problem on my machine which randomly reset
>itself or crash during windows boot up or while executing some
>applications. I believed it may come from memory misbehaviour so i
>downloaded and tried memtest86, last stable release, zip iso download,
>booting it from cdrom.
>

memtest86 is VERY old.

This is the current standard: memtest86+ last updated 1-14-2007

http://www.memtest.org/#downiso

Check out their forums also for various issues. Esp with bios.