Memtest stops at pass 78% and other problems....

[Louise]

I'm trying to track down partial failures - some programs work most of
the time, some never work and some work sometimes.

This happened all of a sudden and the only even I know of was a brief
power failure or deep brownout. Whichever it was, the lights dimmed to
almost nothing and came back up within 30 seconds or maybe even less. I
have a UPS but I don't think it even got set off.

My computer worked fine for the rest of the night but when I shut down
and turned it on in the AM, some programs weren't loading properly, some
wouldn't load at all, and some were fine.

This is an ASUS P4C800E Deluxe with 2 gig of Kingston ram (4 dimms), and
a 120gig Seagate SATA hard drive. I checked the cables and they are
tight.

I ran chkdsk /f /r. I ran a virus checker in addition to the one
installed. I defragged.

Finally I restored a Ghost image from about 2 weeks ago. This erratic
behavior continued even with the restored image.

Seatools reports critical error on the NTFS file system even after I ran
chkdsk.

I just ran memtest and everything tested perfectly. But suddenly, at
pass 78% and test #8, it just stopped. It didn't report any errors, it
just stopped. Is this normal or does this indicate a problem with the
memory?

What else should I check?

TIA

Louise

Any thoughts - suggestions?

 

[pcbutts1]

I'm trying to track down partial failures - some programs work most of
the time, some never work and some work sometimes.

Sometimes you're a lot better off not owning anything.

This happened all of a sudden and the only even I know of was a brief
power failure or deep brownout.

How do you know of was a brief power failure or deep brownout?

Whichever it was, the lights dimmed to almost nothing and came back up
within 30 seconds or maybe even less.

Maybe somebody.

I have a UPS but I don't think it even got set off.

Cool, get me a ups.

My computer worked fine for the rest of the night but when I shut down
and turned it on in the AM, some programs weren't loading properly, some
wouldn't load at all, and some were fine.

I need more sleep!

This is an ASUS P4C800E Deluxe with 2 gig of Kingston ram (4 dimms), and
a 120gig Seagate SATA hard drive.

You have spyware.

I checked the cables and they are tight.

Are they really tight, Louise?

I ran chkdsk /f /r.

You're going to get angry because I'm going to rat on you.

I ran a virus checker in addition to the one installed.

You have spyware.

I defragged.

You're sounding a little desperate because you're trying to build yourself
up by tearing everybody else down.

Finally I restored a Ghost image from about 2 weeks ago.

Could you be more imprecise?

This erratic behavior continued even with the restored image.

The GOOD kind.

Seatools reports critical error on the NTFS file system even after I ran
chkdsk.

One thing after another.

I just ran memtest and everything tested perfectly.

Just checking out.

But suddenly, at pass 78% and test #8, it just stopped.

It's really shocking how many people are mean and sadistic.

It didn't report any errors, it just stopped.

It staggers the imagination, doesn't it?

Is this normal or does this indicate a problem with the memory?

I have no idea. Is that normal or does that indicate a problem with the
memory?

What else should I check?

You call yourself a woman?

TIA Louise Any thoughts - suggestions?

Get ****ed in advance.

 

[philo]

I just ran memtest and everything tested perfectly. But suddenly, at
pass 78% and test #8, it just stopped. It didn't report any errors, it
just stopped. Is this normal or does this indicate a problem with the
memory?

What else should I check?

if memtest froze ...then your memory is prob bad.
get some new meomory and give that a try

 

[spodosaurus]

I'm trying to track down partial failures - some programs work most of
the time, some never work and some work sometimes.

This happened all of a sudden and the only even I know of was a brief
power failure or deep brownout. Whichever it was, the lights dimmed to
almost nothing and came back up within 30 seconds or maybe even less. I
have a UPS but I don't think it even got set off.

Then your UPS is not functioning properly, doing who knows what to the
current coming in through it. Replace it. There is no lag time in a
working UPS.

My computer worked fine for the rest of the night but when I shut down
and turned it on in the AM, some programs weren't loading properly, some
wouldn't load at all, and some were fine.

Again, if your UPS was working this brownout wouldn't have affected the
computer (assuming no surge was involved). Brownouts eat hard drives
Have you downloaded (on a different computer, obviously) the program
from your hard drive manufacturer to test your drive with?

This is an ASUS P4C800E Deluxe with 2 gig of Kingston ram (4 dimms), and
a 120gig Seagate SATA hard drive. I checked the cables and they are
tight.

It would take one hell of a 'brownout' to knock cables free!

I ran chkdsk /f /r.

And the results were?

I ran a virus checker in addition to the one
installed. I defragged.

Finally I restored a Ghost image from about 2 weeks ago. This erratic
behavior continued even with the restored image.

Time to figure out which pieces of hardware got cooked.

Seatools reports critical error on the NTFS file system even after I ran
chkdsk.

What else does seatools report? Did you do a full scan with it?

I just ran memtest and everything tested perfectly. But suddenly, at
pass 78% and test #8, it just stopped.

So it didn't get through a full pass. What's your CPU temperature, by
the way?

It didn't report any errors, it
just stopped. Is this normal or does this indicate a problem with the
memory?

What else should I check?

The memory modules individually, not together.

TIA

Louise

Any thoughts - suggestions?

--
spammage trappage: replace fishies_ with yahoo

I'm going to die rather sooner than I'd like. I tried to protect my
neighbours from crime, and became the victim of it. Complications in
hospital following this resulted in a serious illness. I now need a bone
marrow transplant. Many people around the world are waiting for a marrow
transplant, too. Please volunteer to be a marrow donor:
http://www.abmdr.org.au/
http://www.marrow.org/

 

[spodosaurus]

if memtest froze ...then your memory is prob bad.
get some new meomory and give that a try

Or the CPU is overheating or the motherboard is borked. Testing each
module one at a time is the first step, with all other devices
disconnected from the board (including the hard drive which has already
got errors).

--
spammage trappage: replace fishies_ with yahoo

I'm going to die rather sooner than I'd like. I tried to protect my
neighbours from crime, and became the victim of it. Complications in
hospital following this resulted in a serious illness. I now need a bone
marrow transplant. Many people around the world are waiting for a marrow
transplant, too. Please volunteer to be a marrow donor:
http://www.abmdr.org.au/
http://www.marrow.org/

 

[Louise]

Then your UPS is not functioning properly, doing who knows what to the
current coming in through it. Replace it. There is no lag time in a
working UPS.


Again, if your UPS was working this brownout wouldn't have affected the
computer (assuming no surge was involved). Brownouts eat hard drives
Have you downloaded (on a different computer, obviously) the program
from your hard drive manufacturer to test your drive with?


It would take one hell of a 'brownout' to knock cables free!


And the results were?


Time to figure out which pieces of hardware got cooked.


What else does seatools report? Did you do a full scan with it?

Running chkdsk /f /r doesn't give results in Win XP Pro (or at least not
on my machine . It just does its thing and then it's gone and you're
back at the desktop.

No viruses were found

Seatools reported no error other than a critical fault in the NTFS file
system, index, (can't remember what else). It suggested running all the
defrag, chkdsk etc. Which I did. Seatools was run in full scan and no
other errors were reported.

CPU - 45C
Motherboard - 31C

Vcore Voltage 1.552
3.3v - 3.24 to 3.26
5v - 4.95 to 4.97

12v - 11.96 to 11.97

How are these numbers?

These figures were taken off the motherboard because my other monitoring
program is one of the many programs that isn't functioning.

As per some suggestions here, I have turned up the fan on the CPU (have
an adjustable fan-mate on it). It is now as high as it will go.

I am going to re-run memtest.

BTW, on the Asus board I have, memory is in pairs - I can take out one
pair and check it again. That will be my next step if this fails.

Any and all thoughts greatly appreciated.

Louise

 

[spodosaurus]

But was this a full check using seatools? IIRC there are at least two
different seatools tests. Also, were you using the seatools from within
windows or the seatools from a floppy or cdrom in DOS mode? Seeing as
you already have a ghost backup you might as well use the seatools
floppy and run its version of a surface scan to really make sure the
hard drive is okay.

Running chkdsk /f /r doesn't give results in Win XP Pro (or at least not
on my machine . It just does its thing and then it's gone and you're
back at the desktop.

Yes, it does. It records the results to a log file, you just have to go
dig it up. Second result from a google search:

http://www.computing.net/windowsxp/wwwboard/forum/132863.html

No viruses were found

Seatools reported no error other than a critical fault in the NTFS file
system, index, (can't remember what else). It suggested running all the
defrag, chkdsk etc. Which I did. Seatools was run in full scan and no
other errors were reported.

CPU - 45C
Motherboard - 31C

Vcore Voltage 1.552
3.3v - 3.24 to 3.26
5v - 4.95 to 4.97

12v - 11.96 to 11.97

How are these numbers?

When are you getting the CPU temps? Is it from the BIOS right after a
freeze occurs when you have to reboot?

These figures were taken off the motherboard because my other monitoring
program is one of the many programs that isn't functioning.

As per some suggestions here, I have turned up the fan on the CPU (have
an adjustable fan-mate on it). It is now as high as it will go.

I am going to re-run memtest.

BTW, on the Asus board I have, memory is in pairs - I can take out one
pair and check it again. That will be my next step if this fails.

You don't HAVE to have them in pairs to run memtest, and since you're
having problems running memtest then you SHOULDN'T be running them in
pairs. You only need the pairs for running the memory in dual channel.
Please try testing individual modules.

Any and all thoughts greatly appreciated.

Louise

--
spammage trappage: replace fishies_ with yahoo

I'm going to die rather sooner than I'd like. I tried to protect my
neighbours from crime, and became the victim of it. Complications in
hospital following this resulted in a serious illness. I now need a bone
marrow transplant. Many people around the world are waiting for a marrow
transplant, too. Please volunteer to be a marrow donor:
http://www.abmdr.org.au/
http://www.marrow.org/

 

[w_tom]

As others have noted, the UPS should have made brownouts
completely irrelevant. Furthermore, brownouts do not damage
hardware. This assumes your power supply contains essential
functions that even existed 30+ years ago. Many today use MBA
concepts to buy power supplies only on price. Then when the
missing functions result in hardware damage, the naive blame
the brownout. Nonsense. That would be a classic example of
human failure because the human used what is taught in the
business schools rather than first learn how hardware works.

Minimally acceptable power supply starts at about $65
retail. Obviously, that did not say a $65 power will contain
all essential functions. Furthermore, lights must be less
than 40% intensity - and still the power supply provides a
computer with sufficient power. Properly constructed power
supplies will also perform long enough after power is lost so
that a UPS can eventually switch over to batteries. Another
essential function.

So first, how good is the power supply in that system?

Second, as spodosaurus noted, the UPS should have made the
brownout completely irrelevant. What happens when the UPS
power cord is removed from wall receptacle? Does computer keep
working? Computer should if the UPS is minimally functional.

Now for the memory test. Memory is not just tested at room
temperature. It is even better tested in a room at 100 degree
F or with selected ICs heated by hair dryer on high. Hair
dryer should slightly burn your skin before it gets anywhere
near to being too hot for memory. But memory that fails at
those warmer temperatures will typically be failing later at
70 degree F. Heat is how you find defective memory (or
related ICs) before the intermittent becomes a hard error.

Fourth, you are apparently taking voltage measurements using
the on-board monitor hardware. Does not matter how many
programs are used. It is still the same uncalibrated
hardware. Hardware is only a monitor. Its readings are not
reliable until first confirmed and compensated for using a 3.5
digit multimeter. Voltage could be a problem. 3.3 at 3.21
volts is reason to suspect things that even the meter cannot
detect. 3.14 is considered unconditional failure. Your
numbers don't report useful until that source's integrity is
verified by a meter.

Fifth, forget the nonsense about heat. Heat is often cited
first by those who never really learned how hardware works.
For example, what is the maximum normal temperature for the
CPU? How many degree difference between your CPU and max
temperature? That alone says any accusation about heat was
bogus. But again, how do we find defective hardware? Heat it
to normal but high temperatures. Those who just know without
first learning why would have blames the heat rather than
intermittent and getting worse hardware. Ignore all that
nonsense about fans. If the one chassis fan is spinning and
if CPU temperature is that low, then you have more than enough
cooling even for a 100 degree F room. Notice I also include
numbers with this post so that junk science reasoning is not
possible.

It is possible that the memory location used by memtst is
intermittent. Swap memory modules. Then memtst could report
that error in the other memory module. Use heat to find
intermittent semiconductor failures using the diagnostics.

 

[manny]

Minimally acceptable power supply starts at about $65 retail.

Nowadays, less than half that, provided plain grey is chosen
over white trash flash.

Obviously, that did not say a $65 power will contain all
essential functions.

Actually it probably will have all the essentials - decent
construction and protection from under/over voltage and
over current/power/emperature.

Now for the memory test. Memory is not just tested at room
temperature. It is even better tested in a room at 100 degree
F or with selected ICs heated by hair dryer on high. Hair
dryer should slightly burn your skin before it gets anywhere
near to being too hot for memory.

That could explain why so many memory modules from non chip
makers fail. According to www.hardwaresecrets.com/article/171/3
Kingston relies primarily on ordinary PCs for its testing, and
not merely to control special testing hardware.

 

[Shep©]

I just ran memtest and everything tested perfectly. But suddenly, at
pass 78% and test #8, it just stopped. It didn't report any errors, it
just stopped. Is this normal or does this indicate a problem with the
memory?

Memtest is a flawed program.I do not recommend using any software
programs for testing RAM.
If you have tow sticks of RAM try each separately in different slots
and see if you can replicate faalts.If you cannot then it's not the
RAM.If you can by elimination you can spot the bad stick or mother
board slot.
HTH

 

[larry moe 'n curly]

Memtest is a flawed program. I do not recommend using any
software programs for testing RAM.

How does an ordinary person test RAM without using software? Most
computer shops charge to test it, and apparently the <$1,000 hardware
they typically use doesn't do as thorough a job as MemTest86 (not
MemTest) or Gold Memory.

Evaluations of memory test software and hardware can be found here:

www.realworldtech.com/page.cfm?ArticleID=RWT120901222920

and here:

www.realworldtech.com/page.cfm?ArticleID=RWT052001232443

MemTest did OK but not as well as MemTest86, but Gold Memory
(www.goldmemory.cz), a shareware program, found even error that
MemTest86 missed.

I've had a couple of modules that passed either MemTest86 or Gold
Memory but not both, so I don't rely on just one diagnostic. Most
recently, I had a Kingston 512MB PC3200 module that passed MemTest86
for 4-5 hours but failed Gold Memory in about ten minutes.

 

[Shep©]

How does an ordinary person test RAM without using software? Most
computer shops charge to test it, and apparently the <$1,000 hardware
they typically use doesn't do as thorough a job as MemTest86 (not
MemTest) or Gold Memory.

You snipped this from my post,
If you have two sticks of RAM try each separately in different slots
and see if you can replicate faalts.If you cannot then it's not the
RAM.If you can by elimination you can spot the bad stick or mother
board slot.

and it's not rocket science to do

HTH

 

[JAD]

Memtest is a flawed program. I do not recommend using any
software programs for testing RAM.

How does an ordinary person test RAM without using software? Most
computer shops charge to test it, and apparently the <$1,000 hardware

I have been asked to donate 5 - 10 bucks to test it.........MUCH better than
a new module and find its the slot or the controller.

they typically use doesn't do as thorough a job as MemTest86 (not
MemTest) or Gold Memory.

Utter nonsence...


When you post a 'article' for 'proof' make sure you know where those
articles come from. RTW prints anything from anybody that's in the
business...even from employees- volunteers of any company or brain trust.
Also 2 'different' sources would better uphold your view. Using only
software to determine a module to be good or defective is not as bad today
as it was when RAM costs were much higher, but if its a 1 gig module, for
SURE its going to a hardware tester before I start tossin them.



Evaluations of memory test software and hardware can be found here:

www.realworldtech.com/page.cfm?ArticleID=RWT120901222920

and here:

www.realworldtech.com/page.cfm?ArticleID=RWT052001232443

MemTest did OK but not as well as MemTest86, but Gold Memory
(www.goldmemory.cz), a shareware program, found even error that
MemTest86 missed.

I've had a couple of modules that passed either MemTest86 or Gold
Memory but not both, so I don't rely on just one diagnostic. Most
recently, I had a Kingston 512MB PC3200 module that passed MemTest86
for 4-5 hours but failed Gold Memory in about ten minutes.

 

[larry moe 'n curly]

I have been asked to donate 5 - 10 bucks to test it.........MUCH
better than a new module and find its the slot or the controller.

Utter nonsence...

Have they ever found flaws that a 12-hour test with Memtest86 and Gold
Memory failed to spot?

When you post a 'article' for 'proof' make sure you know where those
articles come from. RTW prints anything from anybody that's in the
business...even from employees- volunteers of any company or brain trust.

What's RTW? The source I gave, www.realworldtech.com, seemed to run
its own tests and even went to the trouble of using some known
defective modules supplied by Micron.

Also 2 'different' sources would better uphold your view.

Then what did RTW say? I can't find any other memory diagnostic
reviews.

 

[JAD]

I have been asked to donate 5 - 10 bucks to test it.........MUCH
better than a new module and find its the slot or the controller.

Utter nonsence...

Have they ever found flaws that a 12-hour test with Memtest86 and Gold
Memory failed to spot?

When you post a 'article' for 'proof' make sure you know where those
articles come from. RTW prints anything from anybody that's in the
business...even from employees- volunteers of any company or brain trust.

What's RTW? The source I gave, www.realworldtech.com, seemed to run
its own tests and even went to the trouble of using some known
defective modules supplied by Micron.

i figured you would jump the typo...... NM you believe what you
want..,,,,,,,,

Also 2 'different' sources would better uphold your view.

Then what did RTW say? I can't find any other memory diagnostic
reviews.