OT: Memory Test Software

Jim Thompson · Oct 30, 2004

480)460-2350 | |
| E-mail Address at Website Fax

Roy Battell · Oct 30, 2004

In article <1gs5o05uqso0q8dcbri3b8uegifd7te8ap@4ax.com>, Jim Thompson
<thegreatone@example.com> writes

Seeing occasionally some strange memory error messages upon close of
Paint Shop Pro or Agent.

Looking for recommendations for software to test RAM memory in my PCs.

Thanks!

...Jim Thompson

By the time stamps looks like a bad night!

I use memtest86+ to be found at
http://www.memtest.org/

Makes a floppy boot disk - hope your machine has one of these
('obsolete' in marketing men's eyes) drives.

Roy
--
Roy Battell.
To use this address remove the digits included to remove Spam ...
Mail: news@vutrax666.co.uk
URL: http://www.vutrax.co.uk (Vutrax CAD system, FREE up to 256 pins)

Robert Monsen · Oct 30, 2004

Jim Thompson wrote:

Seeing occasionally some strange memory error messages upon close of
Paint Shop Pro or Agent.

Looking for recommendations for software to test RAM memory in my PCs.

Thanks!

...Jim Thompson

There is a package from microsoft. You write it onto a floppy or CD, and
when it boots, it continually tests. Search for Memory Test on the
microsoft site.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

Laurence Taylor · Oct 30, 2004

Jim Thompson wrote:

Seeing occasionally some strange memory error messages upon close of
Paint Shop Pro or Agent.

Looking for recommendations for software to test RAM memory in my PCs.

I used to use CkeckIt for this; it's still available, but I found that
after version 3 it went downhill a bit, so haven't investigated the
current versions.

There are other programmes also, like MicroScope and SiSoft Sandra,
that probably all work the same way.

--

rgds
LAurence

....Minds are like parachutes - they work only when open.

Clint Sharp · Oct 30, 2004

In message <xWgilEBVtzgBFwoR@moorhen.demon.co.uk>, Roy Battell
<news@vutrax666.co.uk> writes

I use memtest86+ to be found at
http://www.memtest.org/
I would second this advice, Memtest is excellent, free and very

thorough.

Makes a floppy boot disk - hope your machine has one of these
('obsolete' in marketing men's eyes) drives.
Ahh, but there's an ISO version for making a bootable CD now....

Roy

--
Clint Sharp

Guy Macon · Oct 30, 2004

Clint Sharp wrote:

Memtest is excellent, free and very thorough.

BTW, There is another product called "memtest" and one called
"memtest86", neither of which is quite as good as memtest86+
( http://www.memtest.org/ ).

Winfield Hill · Nov 2, 2004

Jim Thompson wrote...

Chuck Harris wrote:

I always take "good" diagnostic results with a grain of salt.

I'm cynical also... I'm going to start logging the faults and
see if I spot a pattern.

When you say strange memory error messages, what specifically
are these messages?

--
Thanks,
- Win

(email: use hill_at_rowland-dotties-org for now)

Chuck Harris · Nov 2, 2004

Jim Thompson wrote:

I'm cynical also... I'm going to start logging the faults and see if I
spot a pattern.

...Jim Thompson

One thing you can try on many machines, is to select a slower memory
access time for the ram. I see this capability on some BIOSs, but not
all.

-Chuck

Chuck Harris · Nov 2, 2004

Clint Sharp wrote:

In message <372dnS7qJfZyexvcRVn-2Q@rcn.net>, Chuck Harris
cf-NO-SPAM-harris@erols.com> writes

I have never found a RAM test program that could find anything but
the most catastrophic failures in RAM.

Obviously the fault has to be present for the software to find it, but I
do rate Memtest86 as one of the better ones, there are several memory
'test' programs out there that I have seen pass an obviously faulty
system and Memtest86 has pinpointed the fault (one case springs to
mind, not actually the memory, but the CPU cache had an intermittent
problem) Of course, you have to let any diagnostic have time to find an
intermittent fault, but leaving a system running diagnostics overnight
isn't a bad solution.

Hi Clint, Memtest 86 is one of the ones that I have used, but I have yet
to have it find a fault in any of the memory sticks in my machines. But
I have had memory sticks that behaved irratically in some situations, just
not with Memtest 86. And yes, I have left it run for a long, long time.

I have had machines
that exhibited flakey problems that looked like probable memory
problems. The ram test programs showed everything to be ok. But
it was only through swapping out memory sticks that the problem
finally went away.

Again, YMMV, I do rate Memtest and have been working with PCs since the
4.77MHz IBM PC so have seen more than a few 'diagnostic' utilities in my
time. Of course software is never going to be as good as a dedicated
hardware memory tester but it does cost a lot less, especially if you
only need it for one machine.

It certainly is cost effective, but it still doesn't exercise the memory as
much as a virtual memory, disk cached multitasking operating system like
linux does. Linux is truly intolerant of bad memory. But as I said, when I
used memtest86, it couldn't find the offending memory that crashed linux.
I had to do a swap job to fix the problem.

-Chuck

Active8 · Nov 3, 2004

On Tue, 02 Nov 2004 07:21:51 -0700, Jim Thompson wrote:

On 2 Nov 2004 03:07:45 -0800, Winfield Hill
Winfield_member@newsguy.com> wrote:

Jim Thompson wrote...

Chuck Harris wrote:

I always take "good" diagnostic results with a grain of salt.

I'm cynical also... I'm going to start logging the faults and
see if I spot a pattern.

When you say strange memory error messages, what specifically
are these messages?

It hasn't happened for awhile... naturally ;-) But something like
"Instruction at 0X000... failed to execute at 0X011... "

...Jim Thompson

This would be a typical bug in POS IE or any other carelessly
written program. I just stop using the program

Another error I see these days is "... the location could not be
read" - most likely a bad pointer accessing protected memory. With
Win9x, it would've just GPF'd or worse.

Best Regards,
Mike

Jim Thompson · Nov 2, 2004

On Mon, 01 Nov 2004 20:46:22 -0500, Chuck Harris
<cf-NO-SPAM-harris@erols.com> wrote:

Jim Thompson wrote:

Thanks, Robert, It works great! No RAM problems, it appears to be an
IE issue... as if that's a surprise.

...Jim Thompson

I have never found a RAM test program that could find anything but
the most catastrophic failures in RAM. I have had machines
that exhibited flakey problems that looked like probable memory
problems. The ram test programs showed everything to be ok. But
it was only through swapping out memory sticks that the problem
finally went away.

My guess is that sometimes the ram tester programs are testing the
wrong thing. For instance, suppose that the problem is related
to a timing problem that only comes into play when DMA is accessing
ram. A ram test that only tests accesses by the CPU won't find any
problem. Or, suppose the problem is related to an interleaving of
CPU, cache and DMA accesses?

I always take "good" diagnostic results with a grain of salt.

-Chuck Harris

480)460-2350 | |
| E-mail Address at Website Fax

Clint Sharp · Nov 2, 2004

In message <372dnS7qJfZyexvcRVn-2Q@rcn.net>, Chuck Harris
<cf-NO-SPAM-harris@erols.com> writes

I have never found a RAM test program that could find anything but
the most catastrophic failures in RAM.
Obviously the fault has to be present for the software to find it, but I

do rate Memtest86 as one of the better ones, there are several memory
'test' programs out there that I have seen pass an obviously faulty
system and Memtest86 has pinpointed the fault (one case springs to
mind, not actually the memory, but the CPU cache had an intermittent
problem) Of course, you have to let any diagnostic have time to find an
intermittent fault, but leaving a system running diagnostics overnight
isn't a bad solution.

I have had machines
that exhibited flakey problems that looked like probable memory
problems. The ram test programs showed everything to be ok. But
it was only through swapping out memory sticks that the problem
finally went away.
Again, YMMV, I do rate Memtest and have been working with PCs since the

4.77MHz IBM PC so have seen more than a few 'diagnostic' utilities in my
time. Of course software is never going to be as good as a dedicated
hardware memory tester but it does cost a lot less, especially if you
only need it for one machine.

--
Clint Sharp

Jim Thompson · Nov 2, 2004

On 2 Nov 2004 03:07:45 -0800, Winfield Hill
<Winfield_member@newsguy.com> wrote:

Jim Thompson wrote...

Chuck Harris wrote:

I always take "good" diagnostic results with a grain of salt.

I'm cynical also... I'm going to start logging the faults and
see if I spot a pattern.

When you say strange memory error messages, what specifically
are these messages?

480)460-2350 | |
| E-mail Address at Website Fax

OT: Memory Test Software

Jim Thompson

Guest

Roy Battell

Guest

Robert Monsen

Guest

Laurence Taylor

Guest

Clint Sharp

Guest

Guy Macon

Guest

Winfield Hill

Guest

Chuck Harris

Guest

Chuck Harris

Guest

Active8

Guest

Jim Thompson

Guest

Clint Sharp

Guest

Jim Thompson

Guest

Welcome to EDABoard.com

Sponsor

Online statistics

Forum statistics

OT: Memory Test Software

Jim Thompson

Guest

Roy Battell

Guest

Robert Monsen

Guest

Laurence Taylor

Guest

Clint Sharp

Guest

Guy Macon

Guest

Winfield Hill

Guest

Chuck Harris

Guest

Chuck Harris

Guest

Active8

Guest

Jim Thompson

Guest

Clint Sharp

Guest

Jim Thompson

Guest

Log in

Welcome to EDABoard.com

Sponsor