Could Memtest be wrong??

[jake]

I had a computer go into a pattern of restarts to the boot choice
screen but from there it could not go into safe mode and would
just reboot. I pulled the harddrive and put it in another system
and did a chkdsk /f on it and fixed several problems. Then it
worked fine. I decided on a lark just to run memtest to see if
maybe a bad memory stick was the problem. I got hundreds of
errors in a very short time. That shocked me since the Kingston
sticks (2 - 512mb sticks) had been totally reliable for 3 months.

So I pulled one stick to find the bad one. Again, errors right away.
So I swapped sticks just to make sure. Again, hundreds of errors
right away.

So I stuck in two other sticks I have had. Again, hundreds of errors.

What the heck!!! So I pulled the memtest cd and started normally.
It starts fine and looks fine.

What could be doing this??

Thanks much!!!!

system spec's: XP on ASUS A8N32-SLI with eVGA 7800GT
and AMD 64 X2 4400+ and Zalman P4 fan. The mem sticks
are Kingston PC3200 400MHZ CL2.5 DDR

 

[John Doe]

So I pulled one stick to find the bad one. Again, errors right away.
So I swapped sticks just to make sure. Again, hundreds of errors
right away.

So I stuck in two other sticks I have had. Again, hundreds of errors.

What could be doing this??

Something other than the memory sticks.

 

[Frank McCoy]

Something other than the memory sticks.

Like a bad motherboard.
Bad caps is the sort of thing that comes to mind.

 

[Franc Zabkar]

I had a computer go into a pattern of restarts to the boot choice
screen but from there it could not go into safe mode and would
just reboot. I pulled the harddrive and put it in another system
and did a chkdsk /f on it and fixed several problems. Then it
worked fine. I decided on a lark just to run memtest to see if
maybe a bad memory stick was the problem. I got hundreds of
errors in a very short time. That shocked me since the Kingston
sticks (2 - 512mb sticks) had been totally reliable for 3 months.

So I pulled one stick to find the bad one. Again, errors right away.
So I swapped sticks just to make sure. Again, hundreds of errors
right away.

So I stuck in two other sticks I have had. Again, hundreds of errors.

What the heck!!! So I pulled the memtest cd and started normally.
It starts fine and looks fine.

What could be doing this??

Thanks much!!!!

system spec's: XP on ASUS A8N32-SLI with eVGA 7800GT
and AMD 64 X2 4400+ and Zalman P4 fan. The mem sticks
are Kingston PC3200 400MHZ CL2.5 DDR

FWIW, I've had problems with an older version of Memtest running on
both a 486 machine and a socket 7 box. In the latter case Memtest fell
over when testing the RAM shared by the onboard graphics. IIRC, in
both cases I was able to get Memtest to run without error if I
excluded the problem RAM addresses.

FWIW, you might like to try Microsoft's memory diagnostic tool:
http://oca.microsoft.com/en/mtinst.exe

BTW, you didn't say whether you were using the latest version of
Memtest86+. V1.70 "adds detection for Turion 64 X2".

- Franc Zabkar

 

[spodosaurus]

I had a computer go into a pattern of restarts to the boot choice
screen but from there it could not go into safe mode and would
just reboot. I pulled the harddrive and put it in another system
and did a chkdsk /f on it and fixed several problems. Then it
worked fine. I decided on a lark just to run memtest to see if
maybe a bad memory stick was the problem. I got hundreds of
errors in a very short time. That shocked me since the Kingston
sticks (2 - 512mb sticks) had been totally reliable for 3 months.

So I pulled one stick to find the bad one. Again, errors right away.
So I swapped sticks just to make sure. Again, hundreds of errors
right away.

So I stuck in two other sticks I have had. Again, hundreds of errors.

What the heck!!! So I pulled the memtest cd and started normally.
It starts fine and looks fine.

What could be doing this??

ummmm....the motherboard...maybe....perhaps...

Thanks much!!!!

system spec's: XP on ASUS A8N32-SLI with eVGA 7800GT
and AMD 64 X2 4400+ and Zalman P4 fan. The mem sticks
are Kingston PC3200 400MHZ CL2.5 DDR

--
spammage trappage: remove the underscores to reply
Many people around the world are waiting for a marrow transplant. Please
volunteer to be a marrow donor and literally save someone's life:
http://www.abmdr.org.au/
http://www.marrow.org/

 

[jake]

BTW, you didn't say whether you were using the latest version of
Memtest86+. V1.70 "adds detection for Turion 64 X2".

I was using the most recent release of 3.3.

Isn't that the one to use??

 

[jake]

FWIW, I've had problems with an older version of Memtest running on
both a 486 machine and a socket 7 box. In the latter case Memtest fell
over when testing the RAM shared by the onboard graphics. IIRC, in
both cases I was able to get Memtest to run without error if I
excluded the problem RAM addresses.

FWIW, you might like to try Microsoft's memory diagnostic tool:
http://oca.microsoft.com/en/mtinst.exe

BTW, you didn't say whether you were using the latest version of
Memtest86+. V1.70 "adds detection for Turion 64 X2".

I got Memtest86+. V1.70. Test 6 turns red for all addresses (as far
as I can tell).

I can't get Microsoft's tool to run. It starts, but won't run on that
computer.

 

[JAD]

Could Memtest be wrong??


ABSOPOSTIVELY! especially if your running it in a suspect machine!

 

[jake]

Could Memtest be wrong??


ABSOPOSTIVELY! especially if your running it in a suspect machine!

What exactly do you mean by suspect machine? Test 6 of memtest86+
version 1.7 goes RED. I have tried 4 different memory sticks and they
all go red for test 6. I have a very difficult time believing that all the
sticks
are bad.

On the other hand, the machine works well. I did get a crash one time from
Kernel Data Inpage Error for win32.K.sys and could not get into safemode.
It just kept rebooting to start safemode. I pulled the drive and put it in
a
different computer and did a chkdsk /f that fixed some clusters. Since
then,
it has ran fine. I only got all these indications when I tried to do a
memory
test. I still cannot get a memory test to run correctly.

If I replace the mobo (A8N32-SLI), I would replace with a cheaper board
since I don't use SLI. Any recommendations??

 

[JAD]

What exactly do you mean by suspect machine? Test 6 of memtest86+
version 1.7 goes RED. I have tried 4 different memory sticks and they
all go red for test 6. I have a very difficult time believing that all the sticks
are bad.

In order to effectivly test something that is part a certain enviroment, and that
enviroment is not working properly, you have to remove the test subject OUT of that
enviroment and test in something that is known to be good.

On the other hand, the machine works well. I did get a crash one time from
Kernel Data Inpage Error for win32.K.sys and could not get into safemode.
It just kept rebooting to start safemode. I pulled the drive and put it in

sounds like hard drive or simply a bogus windows install, even simpley a driver issue

a
different computer and did a chkdsk /f that fixed some clusters. Since then,
it has ran fine. I only got all these indications when I tried to do a memory
test. I still cannot get a memory test to run correctly.

you could try and set 'default' memory timings in the bios. or change them to something
slower.

If I replace the mobo (A8N32-SLI), I would replace with a cheaper board
since I don't use SLI. Any recommendations??

SLi is a hokey marketing scam...i don't think you need a new mother board

 

[John Doe]

In order to effectivly test something that is part a certain
enviroment, and that enviroment is not working properly, you have to
remove the test subject OUT of that enviroment and test in something
that is known to be good.

That is not true if you are looking for a positive test result. You
aren't going to get a positive test result from a bad part if the
machine is failing otherwise.

Learn a new trick, JAD.

 

[JAD]

That is not true if you are looking for a positive test result. You
aren't going to get a positive test result from a bad part if the
machine is failing otherwise.

Learn a new trick, JAD.

'always' true when trying to find FAULT asswipe...which is what he was trying to
do....... you smell funny

speaking of tricks...hows your mom?

 

[John Doe]

'always' true when trying to find FAULT asswipe...

If you get a positive result and you still think that part is suspect,
you are an idiot.

which is what he was trying to do.......

You were speaking in general terms, idiot.

speaking of tricks...hows your mom?

Okay then... learn a new original trick.

 

[Franc Zabkar]

I was using the most recent release of 3.3.

Isn't that the one to use??

Memtest86+ (http://www.memtest.org/) appears to be a third party
enhancement of Chris Brady's original work. The author claims that he
has access to the latest hardware so I suspect that his version would
be more up-to-date.

Can you post your error messages? Do these errors point to a
particular bit? Can you make these errors go away if you choose the
most conservative memory timings in your BIOS/CMOS setup? Is the CPU
cache disabled during your tests? Is it an ECC issue ???

The fact that "all addresses" fail a particular test suggests to me
that the fault probably lies with memtest86.

Maybe you are witnessing the following problem:

"There are some systems that cause Memtest86 to be confused about the
size of memory and it will try to test non-existent memory. This will
cause a large number of consecutive addresses to be reported as bad
and generally there will be many bits in error. If you have a
relatively small number of failing addresses and only one or two bits
in error you can be certain that the errors are valid. Also
intermittent errors are without exception valid. Frequently memory
vendors question if Memtest86 supports their particular memory type or
a chipset. Memtest86 is designed to work with all memory types and all
chipsets. Only support for ECC requires knowledge of the chipset."

- Franc Zabkar

 

[Bob M]

I had a computer go into a pattern of restarts to the boot choice
screen but from there it could not go into safe mode and would
just reboot. I pulled the harddrive and put it in another system
and did a chkdsk /f on it and fixed several problems. Then it
worked fine. I decided on a lark just to run memtest to see if
maybe a bad memory stick was the problem. I got hundreds of
errors in a very short time. That shocked me since the Kingston
sticks (2 - 512mb sticks) had been totally reliable for 3 months.

So I pulled one stick to find the bad one. Again, errors right away.
So I swapped sticks just to make sure. Again, hundreds of errors
right away.

So I stuck in two other sticks I have had. Again, hundreds of errors.

What the heck!!! So I pulled the memtest cd and started normally.
It starts fine and looks fine.

What could be doing this??

Thanks much!!!!

system spec's: XP on ASUS A8N32-SLI with eVGA 7800GT
and AMD 64 X2 4400+ and Zalman P4 fan. The mem sticks
are Kingston PC3200 400MHZ CL2.5 DDR

Go into the BIOS and change your memory timing from 1T to 2T. Then
rerun Memtest. It'll probably pass now. I've seen this happen when
memory could not handle running 1T timings.

Bob

 

[Franc Zabkar]

ummmm....the motherboard...maybe....perhaps...

IMO that would be unlikely. AMD64 CPUs have a direct path to system
RAM, ie the memory controller is embedded in the CPU, and not on the
northbridge chip. One possible problem could be the supply voltage for
the memory modules, but this would tend to produce random errors, or
total failure.

- Franc Zabkar

 

[jake]

Go into the BIOS and change your memory timing from 1T to 2T. Then rerun
Memtest. It'll probably pass now. I've seen this happen when memory could
not handle running 1T timings.

This is a large BIOS with many options. Do you happen to know where that
setting is?

And thanks very much for this suggestion. I now have that computer at
another
location so it will be a couple of days before I can get back to you on the
success.
Do you know what it is that test 6 is the one to fail so dramatically??

thanks again

 

[Bob M]

This is a large BIOS with many options. Do you happen to know where that
setting is?

And thanks very much for this suggestion. I now have that computer at
another
location so it will be a couple of days before I can get back to you on the
success.
Do you know what it is that test 6 is the one to fail so dramatically??

thanks again

It'll be listed in the memory timings section. I'm not familiar with
your motherboard but it will be listed in the MB manual.

Bob

 

[Franc Zabkar]

Do you know what it is that test 6 is the one to fail so dramatically??

Nobody can offer you meaningful advice until you post the actual error
messages. Then we'll know if you have a data or addressing issue, or
whether the errors are related to caching or ECC or whatever. Remember
that the memory controller is internal to the CPU, not the chipset, so
you may even have a CPU issue. The most likely explanation, however,
is that Memtest doesn't understand your hardware.

Here is an alternative memory tester:

DocMemory PC RAM Diagnostic & Docmemory SPD Reader
http://www.simmtester.com/page/products/doc/docinfo.asp

The SPD (Serial Presence Detect) Reader attempts to read the memory
specs stored in the small 8-pin EEPROM chip on each of your memory
sticks. The SPD reader "works for DDR1/DDR2 with Intel and SIS Chipset
only", but the PC RAM diagnostic should work with all CPUs and
chipsets AFAICT.

- Franc Zabkar

 

[been there done that]

For all you people in googleville, this guy is correct. Go to the
forum associated with the memtest86 website and read what the
experts say for more information. It's a bios reported memory map
error issue.