How did I kill it - Althlon XP-M Overclock

I've just seemingly killed my first processor due to overclocking -
something I've done to ever processor I've owned since an Intel P90!

It was an Athlon XP-M 3000+. I'd wound it up to about 2400MHz (13.5 x
178), but got some errors after an hour of Prime 95, so put it back to ~
2300MHz (14 x 166). Core voltage was 1.7V. This is on an Asus A7V8X
mobo.

It ran Prime95 successfully for ~ 2hours, before I needed to go out.
Temperatures had peaked at about 65C (too high?), at one point, but were
typically 60-61C. I switched off the machine.

Four hours or so later, the machine would not start. Discs span up, but
there was no beep, and no signal to the monitor.

I cleared the CMOS, but to no avail.

Replacing the old CPU brought the machine back to life.

Any ideas what killed it? I could understand it dying in action, so to
speak, but to run successfully, shut down OK, and then completely fail
to restart sounds really odd.

Thanks for any help.

Ian
--
Ian Riches
Bedford, UK

It ran Prime95 successfully for ~ 2hours, before I needed to go out.
Temperatures had peaked at about 65C (too high ?

Yep, too high, though remember alot of failures occur right at startup
when voltage is first supplied to the cpu and the fine connecting wires burn
out, hitting high temp weakens them before the end. You can check the max
temp for it at: http://users.erols.com/chare/elec.htm

rms

rms wrote:
It ran Prime95 successfully for ~ 2hours, before I needed to go out.
Temperatures had peaked at about 65C (too high ?

Yep, too high, though remember alot of failures occur right at startup
when voltage is first supplied to the cpu and the fine connecting wires burn
out, hitting high temp weakens them before the end. You can check the max
temp for it at: http://users.erols.com/chare/elec.htm

rms


If the silicon die is exposed, I'd check for chipped
edges or a crack in the die. It is possible for a
heating/cooling cycle, to finish the thing off. So
check for visible faults first. Some of those chips
survive quite nicely with chips off the edges of the
silicon die, so a small chip doesn't guarantee death,
but a chip or crack could spread over time.

Paul

On Mon, 27 Oct 2008 08:42:25 +0000, Ian Riches wrote:

I've just seemingly killed my first processor due to overclocking -
something I've done to ever processor I've owned since an Intel P90!

It was an Athlon XP-M 3000+. I'd wound it up to about 2400MHz (13.5 x
178), but got some errors after an hour of Prime 95, so put it back to ~
2300MHz (14 x 166). Core voltage was 1.7V. This is on an Asus A7V8X
mobo.

It ran Prime95 successfully for ~ 2hours, before I needed to go out.
Temperatures had peaked at about 65C (too high?), at one point, but were
typically 60-61C. I switched off the machine.

Four hours or so later, the machine would not start. Discs span up, but
there was no beep, and no signal to the monitor.

I cleared the CMOS, but to no avail.

Replacing the old CPU brought the machine back to life.

Any ideas what killed it? I could understand it dying in action, so to
speak, but to run successfully, shut down OK, and then completely fail
to restart sounds really odd.

65C doesn't exceed the max temp, but it is way too hot IMO. Generally, it
should run under 60C under load. And like a light bulb, failure usually
occurs during rapid cooling or heating, meaning when powering up or
powering down. You could try lowering the default settings for the cpu and
you may get it to work at a slower speed. But not worth the effort IMO
when you can get a newer, much faster, system these days for peanuts.

--
Want the ultimate in free OTA SD/HDTV Recorder? http://mythtv.org
My Tivo Experience http://wesnewell.no-ip.com/tivo.htm
Tivo HD/S3 compared http://wesnewell.no-ip.com/mythtivo.htm
AMD cpu help http://wesnewell.no-ip.com/cpu.php