worrying if stability test succeeds but winzip extract fails...

"V" wrote in message
...
Hi,

I've got a PC which passes the prime95 test (W2K blend, running for many
tens of hours), and passes memtest86 also running for hours (tested 9
passes).
However if I attempt to extract a huge ZIP file 700MB it gives CRC errors
always in different places.

Memory is PC3200-DDR400 (Kingston - Winbond)
CPU is AMD Athlon 2500+ (Barton core)
VIA KT600 chipset

These are my test results (core voltage 1.65V, RAM voltage: default
(1.55V)
for all tests). I have repeated these tests several times:

CPU FSB 333MHz (1833MHz) - Memory 333MHz:
Prime95: pass
memtest86: pass
Winzip extract: pass
CPU FSB 333MHz (1833MHz) - Memory 400MHz: fail
Prime95: pass
memtest86: pass
Winzip extract: FAIL: CRC errors
CPU FSB 400MHz (2203MHz) - Memory 333MHz: pass
Prime95: pass
memtest86: pass
Winzip extract: pass
CPU FSB 400MHz (2203MHz) - Memory 400MHz: fail
Prime95: pass
memtest86: pass
Winzip extract: FAIL: CRC errors

One conclusion I can draw is that prime95 and memtest86 alone are not
sufficient to test the stability of a PC...

--
V

The conclusion I would draw seems pretty clear; your memory has a problem
rrunning at 200 MHz. Perhaps the memory timing is too tight, or you need a
bit more voltage to the memory for it to run at 200 MHz. 2.55v may be the
default, but memory generally requires more to run stable at 200 MHz.

It may also be a problem with the chipset. Try increasing chipset voltage
if available.

"dr_hardwrae" schreef in bericht
et...

The conclusion I would draw seems pretty clear; your memory has a problem

rrunning at 200 MHz. Perhaps the memory timing is too tight, or you need a

bit more voltage to the memory for it to run at 200 MHz. 2.55v may be the

default, but memory generally requires more to run stable at 200 MHz.

Thanks all for your replies.

Increasing SDRAM voltage to 2.65V makes the problem go away...

I would not expect these problems with Kingston memory...

--

V

"V" wrote in message
...
"dr_hardwrae" schreef in bericht
et...

The conclusion I would draw seems pretty clear; your memory has a
problem

rrunning at 200 MHz. Perhaps the memory timing is too tight, or you need
a

bit more voltage to the memory for it to run at 200 MHz. 2.55v may be
the

default, but memory generally requires more to run stable at 200 MHz.

Thanks all for your replies.

Increasing SDRAM voltage to 2.65V makes the problem go away...

I would not expect these problems with Kingston memory...

--

V

As mentioned, this is pretty common with *all* memory. The Kingston you
have uses Winbond chips (excellent chips used with TwinMos memory modules
and others), which almost always require 2.65v or more to run at 200 MHz and
above.
Anyway, glad to hear the problem is resolved. Enjoy!

"V" wrote in
:

Hi,

I've got a PC which passes the prime95 test (W2K blend, running for
many tens of hours), and passes memtest86 also running for hours
(tested 9 passes).
However if I attempt to extract a huge ZIP file 700MB it gives CRC
errors always in different places.

memtest86 is not really accurate to detect memory stability. Tho it can
find memory defect quite sufficient enough. Errored out on extraction
usually means your memory timing is too aggressive for your current
memory/vdmm voltage.

--
Watch your back when it's time to stab.

Ken

In article ,
says...
"V" wrote in
:

Hi,

I've got a PC which passes the prime95 test (W2K blend, running for
many tens of hours), and passes memtest86 also running for hours
(tested 9 passes).
However if I attempt to extract a huge ZIP file 700MB it gives CRC
errors always in different places.

memtest86 is not really accurate to detect memory stability. Tho it can
find memory defect quite sufficient enough. Errored out on extraction
usually means your memory timing is too aggressive for your current
memory/vdmm voltage.

SoundBlaster card?, which MOBO?

BoroLad

In article ,
says...
"V" schreef in bericht
...
One conclusion I can draw is that prime95 and memtest86 alone are not
sufficient to test the stability of a PC...

Now I am absolutely and completely lost. Although increasing the memory
voltage to 2.65V helped to make the problem vanish completely, and reducing
to 2.55V made it reappear, I have now found another "solution":
Removing the SATA HDD. There was one SATA HDD and one PATA HDD (both WD) on
the mobo (A7V600).

After I had done this, the PC turned out to be completely stable again,
although the SATA HDD was not used as the source nor the destination, nor
the temp for the winzip extract process.

CPU FSB 414MHz (2174MHz) @2.65V - Memory 414MHz (/2) @2.55V (!!!)
Prime95: pass
memtest86: pass
Winzip extract: pass

Because I want to know for sure that it is the SATA HDD, I am currently
attempting to reproduce the instability by reintroducing the same SATA HDD,
with no "luck".
So somehow the SATA HDD was causing the instability, but how? It was not
used...

ask for TomG or BigBadger