memtest86, how many passes?

[John Doe]

In ordinary circumstances, normally, how many memtest86 passes
before you consider memory okay?

I got 16 passes okay before one error. So I disabled overclocking
and reduced memory timing to 2T, and it ran for 20 passes with no
errors. I'm sure it would be a wild guess, but should I expect that
error to pop up again within 20 passes if I reenable overclocking
and 1T memory timing?

Does heat cause memory errors? (without first causing permanent
physical damage). In other words, if I run the test in a cooler
environment, is it less likely to produce an error?

Thank you.

 

[JAD]

you go on testing memory in a suspect machine, makes perfect sense......

 

[John Doe]

you go on testing memory in a suspect machine,
makes perfect sense......

You really should try to increase the useful information (versus the
careless/ignorant grandstanding nonsensical trolling) in your
replies.

There's nothing unusual or difficult to understand about errors
when stressing hardware.

Methinks JAD is jealous of my "POS rig". And apparently JAD is still
having trouble figuring out how to use his imaginary friend
(the kill file).

 

[JAD]

You really should try to increase the useful information (versus the
careless/ignorant grandstanding nonsensical trolling) in your
replies.

figure it out dipstick

There's nothing unusual or difficult to understand about errors
when stressing hardware.

Methinks JAD is jealous of my "POS rig". And apparently JAD is still
having trouble figuring out how to use his imaginary friend
(the kill file).

...... and I think your memory needs some work

 

[larry moe 'n curly]

In ordinary circumstances, normally, how many memtest86 passes
before you consider memory okay?

I got 16 passes okay before one error. So I disabled overclocking
and reduced memory timing to 2T, and it ran for 20 passes with no
errors. I'm sure it would be a wild guess, but should I expect that
error to pop up again within 20 passes if I reenable overclocking
and 1T memory timing?

It once took 4-5 hours for MemTest86 v. 3.xx to find an error in one of
my a 256MB PC2100 modules. The CPU was either a 1.6 GHz Duron or
XP1800+ Athlon.

I've found that 2T can improve reliability a LOT, and in the case of 11
or 12 Kingston 512MB PC3200 modules I tested that were rated 1T, 8
failed at 1T, all but one passed at 2T.

Does heat cause memory errors? (without first causing permanent
physical damage). In other words, if I run the test in a cooler
environment, is it less likely to produce an error?

Probably, but memory doesn't that hot in use, except maybe memory used
in graphics cards, and I've never heard of it overheating so much that
it ws damaged, unless it was plugged in backwards or there was a power
supply failure. But don't you want to simulate worst-case conditions
during testing?

 

[JAD]

You really should try to increase the useful information (versus the
careless/ignorant grandstanding nonsensical trolling) in your
replies.

There's nothing unusual or difficult to understand about errors
when stressing hardware.

Methinks JAD is jealous of my "POS rig". And apparently JAD is still
having trouble figuring out how to use his imaginary friend
(the kill file).

Be assured that EVERY machine that I work on, and enable usenet groups on,
have you in it. I won't subject them to your BS. Much like this Me
machine....

 

[WooHoo2You]

Be assured that EVERY machine that I work on, and enable usenet groups on,
have you in it. I won't subject them to your BS. Much like this Me
machine....

Saving the world "one PC at a time."

--
WooHoo2You


"If a free society cannot help the many who are poor, it cannot save the few
who are rich."

-John F. Kennedy, Inaugural Address, 1961

 

[six-pack to be]

It once took 4-5 hours for MemTest86 v. 3.xx to find an error in one of
my a 256MB PC2100 modules. The CPU was either a 1.6 GHz Duron or
XP1800+ Athlon.

I've found that 2T can improve reliability a LOT, and in the case of 11
or 12 Kingston 512MB PC3200 modules I tested that were rated 1T, 8
failed at 1T, all but one passed at 2T.


Probably, but memory doesn't that hot in use, except maybe memory used
in graphics cards, and I've never heard of it overheating so much that
it ws damaged, unless it was plugged in backwards or there was a power
supply failure. But don't you want to simulate worst-case conditions
during testing?

 

[zed]

In ordinary circumstances, normally, how many memtest86 passes
before you consider memory okay?

I got 16 passes okay before one error. So I disabled overclocking
and reduced memory timing to 2T, and it ran for 20 passes with no
errors. I'm sure it would be a wild guess, but should I expect that
error to pop up again within 20 passes if I reenable overclocking
and 1T memory timing?

Does heat cause memory errors? (without first causing permanent
physical damage). In other words, if I run the test in a cooler
environment, is it less likely to produce an error?

Thank you.

You want to explore any errors you get. many times errors only
occur on a specific test e.g. test 5. you can configure memtest to
rerun test 5 over and over. This will speed up testing considerably
as you can get LOTS of test 5 runs in a few hours.

I got 2 errors out of 38 full passes overnight and they occured in
pass 18. I continued testing and there were problems. A touch more
voltage fixed the problem.

 

[Geoff]

Does heat cause memory errors? (without first causing permanent

physical damage).

Yes but for permanent physical damage, when I worked in an engineering
environment, they had to bake the boards in an oven to cause the kind of
damage you are thinking of.

-g

 

[Steven]

You really should try to increase the useful information (versus the
careless/ignorant grandstanding nonsensical trolling) in your
replies.

There's nothing unusual or difficult to understand about errors
when stressing hardware.

Methinks JAD is jealous of my "POS rig". And apparently JAD is still
having trouble figuring out how to use his imaginary friend
(the kill file).

I know two fools to stay away from already.

 

[Steven]

Yes but for permanent physical damage, when I worked in an engineering
environment, they had to bake the boards in an oven to cause the kind of
damage you are thinking of.

-g

And add mozzeralla