final answer to CPU

  • Thread starter Thread starter philo
  • Start date Start date
P

philo

a month or two back i had posted about a p-III (1ghz) that was crashing
and i had replaced all the components except for the cpu and motherboard.

since i was at the end of my rope...i put the closest substitute cpu in
the machine ( a celron-433) and the machine worked fien after that...
so it seemed that no matter how unlikely...the cpu was flakey

anyway i finally got another P-III board that I can play with...
so i put the (in question) P-III in and let it run.
it ran fine...so it seems the cpu was good.


So now I've concluded that the mobo must have been bad...but
good enough to function at a reduced clock rate.

It was an IBM mobo with *no* option to set clock speed so I was no able
to clock it down manually
 
philo said:
a month or two back i had posted about a p-III (1ghz) that was
crashing and i had replaced all the components except for the cpu and
motherboard.
since i was at the end of my rope...i put the closest substitute cpu
in the machine ( a celron-433) and the machine worked fien after
that... so it seemed that no matter how unlikely...the cpu was flakey
Click to expand...

BZZZZT!!!! Wrong answer. The odds of a CPU that has been working fine going
"flakey" without physical damage (or having the voltage ramped up or
something equally user-caused) are about 100,000:1. The chances of the
motherboard 'going flakey' are about 50:1. It it's already gone flakey, the
odds are 2000:1 that it's the mobo.
anyway i finally got another P-III board that I can play with...
so i put the (in question) P-III in and let it run.
it ran fine...so it seems the cpu was good.
Click to expand...

CPUs almost *never* fail in normal usage. Always suspect the CPU last unless
the heatsink fell off in use or you jabbed it with a screwdriver/bent some
pins etc.
So now I've concluded that the mobo must have been bad...but
good enough to function at a reduced clock rate.
Click to expand...

Not reduced clock-rate, reduced power requirement of the CPU. The capacitors
and MOSFETs that supply the vcore were probably flakey. Any bulging/leaking
caps? Even if there weren't it doesn't mean they were fine. A lot of caps
from that era fail, not all of them show physical signs.
It was an IBM mobo with *no* option to set clock speed so I was no
able to clock it down manually
Click to expand...

Probably wouldn't have helped.

Not many mobos were made that support both a PPGA CPU (Celeron 433) and a
FC-PGA CPU (P3 1GHz). Therefore the board must have been old, made when
FC-PGA CPUs were new on the market, to be backwards-compatible.

If it was a Slot 1 board rather than a Socket 370 board then boards that
support both types of CPU were slightly more common. In fact a lot of them
were backwards-compatible.
 
Probably wouldn't have helped.

Not many mobos were made that support both a PPGA CPU (Celeron 433) and a
FC-PGA CPU (P3 1GHz). Therefore the board must have been old, made when
FC-PGA CPUs were new on the market, to be backwards-compatible.

If it was a Slot 1 board rather than a Socket 370 board then boards that
support both types of CPU were slightly more common. In fact a lot of them
were backwards-compatible.
Click to expand...


..i originally thought the odds of a cpu being flakey were rather
small...anyway

the mobo is probably a real piec of junk...
it's not really that old...and it was not designed for the older/slower
cpu...as when it boots up...
there is a bios error about "no rom bios for cpu"
however it does work.

the internet cafe has a total of five of those IBM computers...
in addition to the already mentioned problem...
they have also lost one HD and one ram module. (and one TFT monitor smoked)

also most of the mice and keyboards had to be replaced...
but for public usage machines...i guess that is at least tolerable
 
~misfit~ said:
BZZZZT!!!! Wrong answer. The odds of a CPU that has been working fine going
"flakey" without physical damage (or having the voltage ramped up or
something equally user-caused) are about 100,000:1. The chances of the
motherboard 'going flakey' are about 50:1. It it's already gone flakey, the
odds are 2000:1 that it's the mobo.
Click to expand...

I don't know about that, I've seen a few dead chips. Then again, when
people give ME a fast computer, they've usually given up on it. (I work
at a computer refurbisher and recycler)

There are plenty of ways to kill a CPU that will leave the motherboard
alive. No heat sink, no working fan, too small of a heat sink, badly
applied heat sink, too much dust accumulation. Static electricity,
overclocking. I've seen all of it happen. The scary thing is that about
half the time, the problem can be pinned on an actual technician who
should know better.

I've rarely seen a dead P3, but I've seen lots of dead Athlons, and
P4s. They're easier to overheat.
 
I don't know about that, I've seen a few dead chips. Then again, when
people give ME a fast computer, they've usually given up on it. (I work
at a computer refurbisher and recycler)

There are plenty of ways to kill a CPU that will leave the motherboard
alive. No heat sink, no working fan, too small of a heat sink, badly
applied heat sink, too much dust accumulation. Static electricity,
overclocking. I've seen all of it happen. The scary thing is that about
half the time, the problem can be pinned on an actual technician who
should know better.

I've rarely seen a dead P3, but I've seen lots of dead Athlons, and
P4s. They're easier to overheat.
Click to expand...

CPU's *can* die of course...

my original post (a month or two ago) was concerning what I thought
might have been a semi-working cpu. That possibility seemed very
unlikely...and as it turned out...the cpu was OK...it must have been a
mobo problem.

although there is probably an exception out there somewhere...
in general...a CPU is either good or bad
 
I don't know about that, I've seen a few dead chips. Then again, when
people give ME a fast computer, they've usually given up on it. (I work
at a computer refurbisher and recycler)

There are plenty of ways to kill a CPU that will leave the motherboard
alive. No heat sink, no working fan, too small of a heat sink, badly
applied heat sink, too much dust accumulation. Static electricity,
overclocking. I've seen all of it happen. The scary thing is that about
half the time, the problem can be pinned on an actual technician who
should know better.

I've rarely seen a dead P3, but I've seen lots of dead Athlons, and
P4s. They're easier to overheat.
Click to expand...


They can be killed, but it's a bit of a variable to use not
knowing what state the system is in. If an owner were to
innocently claim it's not overly dusty, fans work, heatsink
didn't fall off or any other obvious sign of a problem then
it is unlikely the CPU failed without an external cause.
 
I don't know about that, I've seen a few dead chips. Then again, when
people give ME a fast computer, they've usually given up on it. (I
work at a computer refurbisher and recycler)

There are plenty of ways to kill a CPU that will leave the motherboard
alive. No heat sink, no working fan, too small of a heat sink, badly
applied heat sink, too much dust accumulation. Static electricity,
overclocking. I've seen all of it happen. The scary thing is that
about half the time, the problem can be pinned on an actual
technician who should know better.
Click to expand...

Exactly, that's why I included "without physical damage (or having the
voltage ramped up or something equally user-caused)" above. I consider a bad
dust build-up to be user-caused. Either due to placement of the PC or lack
of maintainence.
 
1- The motherboard aren´t compatible with Coopermine core (ur celeron is
Mendicino) or power design does not meets PIII 1Ghz requirements
2- Voltage regulators or capacitors are faulty (look at capacitors, they
may be leaky). []´s

--


Mariano Faria - AZTech Hardware (Brasil)
(e-mail address removed) (MSN Messenger)
ICQ# 17542630
Skype: marianofaria
 
Back
Top