Need help here !

My PC has 2 DDR2 memory modules, 1GB each, for a total of 2GB RAM.

The OS is WinXP.

For the past few days, the PC kind of halting. Sometimes the cursor
kind of not responding, sometimes some other things happened,
indicating that something is terribly wrong --- most probably a MEMORY
error.

I downloaded 3 memtest programs,
1 from http://hcidesign.com/memtest/,
1 from http://www.memtest86.com/download.html,
1 from http://www.memtest.org/

First test -

Booted up the PC with the Memtest bootCD from www.memtest86.com.
Ran the test overnight.

No error !

Went back to Windows, the halting problems cropped up. So I ran a
second test.

Second test -

Booted up the PC with the Memtest bootCD from www.memtest.org.
Ran the test overnight again.

Again, No error !


Went back to Windows, again the halting problems !! So the 3rd test.

I drop XP to safe mode with command prompt, then call up 2 instances
of the memtest from http://hcidesign.com/memtest/, each checking
918MB.

Result - Errors !

Memory Errors Detected !
Copying between 4c21c71 and 4c2193a did not result in accurate copy.

Memory Errors Detected !
Pair 19897713 does not store values accurately.

Memory Errors Detected !
Copying between 5547871 and 5547832 did not result in accurate copy.

Memory Errors Detected !
Copying between 4c32471 and 4c323a2 did not result in accurate copy.

Memory Errors Detected !
Copying between 36f8071 and 36f8060a did not result in accurate copy.

Memory Errors Detected !
Copying between 4c32471 and 4c322d9 did not result in accurate copy.

Conclusion:
Something wrong with the DDR2 memory module !

Problem:
Which module giving out errors?

I got one of the DDR2 module out. Ran the memtest from http://hcidesign.com/memtest/
-- no error.

Switch the sticks to test the other DDR2 module. Ran the memtest from
http://hcidesign.com/memtest/ -- again no error !

Now I'm totally confused. Why when I ran the test with both DDR2
modules, I got errors, and no error when I ran the modules one at a
time??

So I ran the bootCD memtest test again, this time with one module at a
time. Again no error on both runs !

What am I doing wrong here? How am I going to locate which module that
is giving out errors? Can anyone help out please?

Thank you !

On Thu, 10 Apr 2008 21:37:42 -0700 (PDT),
penang@freemail.c3.hu wrote:

This generally means there isn't a hard fault in either
memory module, instead a general instability of your
motherboard with multiple modules installed. Is your bios
set to "auto" or "SPD" for memory speed and timings? If
not, try that, and check on a motherboard bios update from
the manufacturer, sometimes they address memory problems.

Beyond this you can instead get the specs for your memory to
determine what speed, timings, and voltage it's supposed to
need and try manually setting these values in the bios. If
that doesn't work, increase the voltage a little bit and
retest. If that doesn't work, try relaxing memory timings
some (higher numbers).

You didn't tell us exactly what memory you have, what
timings it's spec'd for, nor what timings you are running.
"CPU-Z" can show you what info it read from the memory prom
and often what speed it's running at for a reference.


If you have more than two memory slots you might also try
leaving a different one empty, or if it's running in dual
channel mode try in single channel mode - though it may be a
performance hit but besides raising voltage all the other
typical things you'd change to regain stability are also
minor performance hits. The other option is manually
setting a slower memory bus frequency, whether it be a ratio
or % or speed numbers in the bios.

Also examine the board for failed capacitors, bad power can
cause errors. Are you sure this system with this memory
previously ran ok or is it possible it had errors since the
first moment these memory modules were both installed?

penang@freemail.c3.hu wrote:
read your mainboard's specifications carefully.
some boards will require that one slot be less capacity than the
other(s) when paired with one or more module, or have other niggling
details that you might need to ponder over.


--
To the States or any one of them, or any city of the States,
Resist much, obey little,
Once unquestioning obedience, once fully enslaved,
Once fully enslaved, no nation, state, city of this earth,
ever after-ward resumes its liberty.

-Walt Whitman, 1860

penang@freemail.c3.hu wrote:
So the only time this error pops up is from within windows?

I'm assuming you let the memtest cdrom do it's thing for more than one
pass, right?

Ari

--
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/

On Apr 11, 4:13 am, spodosaurus <spodosaurus@_yahoo_.com> wrote:

Actually there's no error message by XP. It's just the halting
problems that brought up my suspicions.

The "Error message" I got is from the memtest program that I got from
hcidesign.

Yep. Ran the memtest cdrom thing overnight, for 4 nights. 2 nights
each for memtest86 v 3.4 and for memtest86+ v 2.01

No error whatsoever from both versions.

On Apr 11, 4:06 am, catchme <some...@somewhere.net> wrote:

Re-read the owner's manual, no such mention at all. Got online and
check the mobo manufacturer's site, also no such mention.

The mobo got 2 slots for DDR2 module. Maximum is 2GB X 2, for a total
of 4GB.

On my PC, I populate it with 1GB X 2.

On Apr 10, 10:26 pm, kony <s...@spam.com> wrote:
There are only 2 memory slots and I use both of them.

Have also tried to alter the speed. The mobo is rated for 800MHz front-
side-bus, and both the DDR2 modules are also rated 800Mhz. I have
tried different configurations -- from 400MHz to 533MHz to 667MHz but
all come down with errors.

I've even tried disabling the L1/L2 cache trying to pinpoint the
problem. Even with that severe performance hit, still the error comes
up.

I've also tried disabling the PCI settings on the BIOS. No luck
either.

At first I thought it was the RAM module. Then I thought it was the
mobo. Now I dunno what to think.



I eyeballed the mobo. No sign of capacitor fault. I even tried
changing the power supply, but still no luck.

I'm going to get 2 new DDR2 modules this weekend. I will run the tests
again and will see if the problem is still there. If it's there, then
it's the mobo. If the problems go away, then it is in the RAM module
(one or both).

penang@freemail.c3.hu wrote:
*sigh*

The hcidesign program runs under windows, right?


--
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/

On Apr 11, 9:16*am, spodosaurus <spodosaurus@_yahoo_.com> wrote:
I like the HD test suggestion by Bob, but I have had problems like
yours in the past setting up a PC, I had actually 4gb in there, and i
had game crashes, windows problems, the whole works. I was stumped for
ever and about to give up, when i came across a new memory test,
called Orthos, you can just google search it, "Orthos memory test" the
application also will over clock CPU, which in your case you won't
probably need, and you can set options so it won't, I had mem test
come clean over 36hours w/o errors, this ran for about 10minuts, and
picked up errors, i hope this helps, good luck

penang@freemail.c3.hu wrote in
news:b2f6d0cd-cca3-4605-b705-03e640cb794a@u36g2000prf.googlegroups.com:

Have you tried disabling your startup programs to see if that is the
cause. First eliminate software and then pinpoint hardware problems.

--
(setq (chuck nil) car(chuck) )

On Fri, 11 Apr 2008 07:55:17 -0700 (PDT),
penang@freemail.c3.hu wrote:



That should work, and usually motherboards are now *meant*
to run two equal modules (in dual channel mode), certainly 2
x 1GB.

On Fri, 11 Apr 2008 08:02:58 -0700 (PDT),
penang@freemail.c3.hu wrote:


It's not necessarily so simple. If you replaced those
modules with some that default to different timings or have
more margin over JEDEC values (assuming the other modules do
meet spec), it could still be a motherboard problem that
causes this fault and yet the higher margins on the
replacement modules might mask, eliminate the instability.
For practical purposes that is an acceptable result, but for
returning the memory that doesn't work to the manufacturer
for a warranty replacement it could be futile as you might
get back exact same or equivalent modules.

On Fri, 11 Apr 2008 12:29:39 -0700 (PDT), zach
<zachvek@hotmail.com> wrote:

<snip>


While Orthos would pick up on limited memory errors, it is
mostly a test of processor and a failure is most typically
an indication of instable or insufficient processor voltage,
overheating northbridge, overheating CPU, or overclocking
pushed too far (incompatible manually set variables or
overclocking as related to a need for more cooling or higher
than stock voltage).

Main point being, if Memtest86+ runs that long w/o error
then orthos fails when one isn't overclocking and is using
bios defaults, it tends to be an overheating or failing
motherboard problem. Other things are possible, but
probably less likely.

On Thu, 10 Apr 2008 21:37:42 -0700 (PDT),
penang@freemail.c3.hu wrote:

While it seems that one of the programs found memory errors,
I would not be so certain that is the cause. If there were
a fault in the one testing under windows, in the design of
the program, that could result in it indicating errors that
wouldn't have occurred otherwise.

Generally, if a system is stable enough it can run
memtest86+ (one of the other 2 test programs) for several
hours without any faults appearing, your errors are
infrequent enough that they would not cause the kind of
symptoms you're seeing. It might be possible, but other
things seem possible like a failing network connection or
hard drive (whether drive be physically failing or a bad
cable connection, etc, an interruption in disk subsystem).
Be sure to check all windows logs including event viewer, to
set windows to not reboot on error, and to write down stop
code messages if you get bluescreens.

It's also possible you have more than one problem. Memory
errors just don't usually cause hesitation but then
everything still seems to keep working, it tends to cause
crashing instead, not just OS, GUI stalls. If your video
card has a fan you might check to ensure it's working, and
generally check temps as well, inspect whole system.

On Thu, 10 Apr 2008 21:37:42 -0700 (PDT), penang@freemail.c3.hu put
finger to keyboard and composed:

IME RAM testing software often doesn't properly test refresh. Refresh
errors usually show up in normal operation because, unlike most RAM
diagnostic software, the OS may not read a memory location immediately
after writing it, thus giving the data some time to decay.

See http://www.memtest86.com/tech.html

================================================== ====================
Test 9 [Bit fade test, 90 min, 2 patterns]

The bit fade test initializes all of memory with a pattern and then
sleeps for 90 minutes. Then memory is examined to see if any memory
bits have changed. All ones and all zero patterns are used. This test
takes 3 hours to complete. The Bit Fade test is not included in the
normal test sequence and must be run manually via the runtime
configuration menu.
================================================== ====================

FWIW, I notice that the failing memory addresses all end in "71" which
makes me wonder whether you have an addressing error. It's a pity that
HCI Design didn't see fit to report the bit differences in the failing
data patterns. :-(

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

penang@freemail.c3.hu wrote:

You get an errors like can't write memory to location #### ?

This can be caused by DEP, turn it off for the program(s) your having
problems with. http://www.tech-recipes.com/windows_tips566.html
--

Amiga Tribute, with help from GLaDOS
http://youtube.com/watch?v=9mg6wrYCT9Q

pen...@freemail.c3.hu wrote:

Another diagnostic is Gold Memory, from www.goldmemory.cz. For me,
ver. 5.07 has found more errors than the newer versions have,
including when MemTest86 has found none (and vice-versa).

spodosaurus wrote:
This is no help with your present difficulty, but it should help
avoid similar problems in future. You should have bought a system
with ECC memory (and the capability to use it). That converts
unreliable memory into quite reliable memory, and avoids all the
intermittent errors due to cosmic reays etc.

Just remember the phrase ECC and don't let anyone talk you out of
it.

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.


** Posted from http://www.teranews.com **

CBFalconer wrote:
Don't be a dick.

spodosaurus wrote:
Such as the ignorant ramblings of spodosaurus.

--
[mail]: Chuck F (cbfalconer at maineline dot net)
[page]: <http://cbfalconer.home.att.net>
Try the download section.


** Posted from http://www.teranews.com **