Archived from groups: alt.comp.periphs.mainboard.elitegroup (More info?)

A year or so ago I bought three K7S5A Pro boards to replace older PC's on an
as-needed basis. Got three identical cases with Austin 300W P4 capable PSs,
three Athalon 2400XP CPUs and 3 strips of Centon 512DDR. The first two
machines went together without a hitch (other than my forgetting to bump the
bus speed up to 133) using 512MB DDR and 2400 Athalon XP processors.
They've been running fine for months.

Now I've gotten to the third one and it refuses to POST. I can even pull
the RAM (and have swapped in and out known good SD and DDR modules) but
nada, zip, zilch, bupkis. Speaker is connected but never beeps. The CPU fan
spins and it's obvious the 'power on' switch works, but no signal is sent to
the CRT and no beeps are heard, even if I pull the video card or the memory
modules. Before I go fishing into two perfectly good machines to start
swapping CPUs and PS's, can anyone suggest something else to try? I've
already changed out the battery and cleared CMOS via the J1 jumper and tried
all sorts of PCI and AGP video cards. Only signs of life are the
motherboard-powered case and CPU fans.

I'm think it's a bad CPU. How often does that occur?

TIA,

Eric

Archived from groups: alt.comp.periphs.mainboard.elitegroup (More info?)

Eric Wolfe wrote:

> A year or so ago I bought three K7S5A Pro boards to replace older PC's on an
> as-needed basis. Got three identical cases with Austin 300W P4 capable PSs,
> three Athalon 2400XP CPUs and 3 strips of Centon 512DDR. The first two
> machines went together without a hitch (other than my forgetting to bump the
> bus speed up to 133) using 512MB DDR and 2400 Athalon XP processors.
> They've been running fine for months.
>
> Now I've gotten to the third one and it refuses to POST. I can even pull
> the RAM (and have swapped in and out known good SD and DDR modules) but
> nada, zip, zilch, bupkis. Speaker is connected but never beeps. The CPU fan
> spins and it's obvious the 'power on' switch works, but no signal is sent to
> the CRT and no beeps are heard, even if I pull the video card or the memory
> modules. Before I go fishing into two perfectly good machines to start
> swapping CPUs and PS's, can anyone suggest something else to try? I've
> already changed out the battery and cleared CMOS via the J1 jumper and tried
> all sorts of PCI and AGP video cards. Only signs of life are the
> motherboard-powered case and CPU fans.
>
> I'm think it's a bad CPU. How often does that occur?
>
> TIA,
>
> Eric
>
>
>
>

While you are doing this test, have you disconnected your cables to the
IDE bus and the floppy disk controller??? It is possible for a reversed
IDE cable to ground out the address bus and prevent a boot up. Also,
since it is easy to short out the underside of a motherboard with a
standoff in the wrong position, I would suggest you remove the MB from
the case until you get it running if you are not already doing that.

It is possible that you have a bad processor, but I would not conclude
that until everything (including the PS) had been eliminated. This
means clearing CMOS with the jumper for that purpose as well.

Archived from groups: alt.comp.periphs.mainboard.elitegroup (More info?)

"PJX" <me@privacy.net> wrote in message

>You are being extremely paranoid about using a working machine to
> test a suspect machine.

Ubetcha! Personal quirk along the line of "if it ain't broke, don't fix
it."

We'll see if it was the right course of action. If it was, AMD will send me
a new processor. If the Duron ran and the Athlon didn't, that's good enough
for me. I can't afford to have the working machine down right now and every
time you pop a heatsink and swap a CPU you're inviting chipped cores,
breaking the tabs off the ZIF socket, static damage and a host of other
maladies that I would rather avoid. The solution is to get a cheap MB to
use solely as a test board but that's why I bought the dirt cheap ECS K7S5A.
Somehow, when they survive all the abuse a test bench machine takes, I can't
help but turning them into workhorse machines.

My track record, up until now, has been good with ECS boards. Far better
than Giga*bite*, Abit, Tyan, Soyo and Asus - and all of those board cost a
lot more. It's going to cost me six bucks to send the CPU back to AMD. I
can live with than gamble easier than downing a working machine because I'm
an avowed klutz.

Besides, there's that old joke about swapping parts that goes like this. An
IBM SW guy, HW guy and marketing guy are driving along and get a flat tire.
The SW guy says "Let's drive it a while more and see if the problem fixes
itself." The HW guy says "Let's swap tires and see if that fixes the
problem" and the marketing guy says "We need a new car!"

Eric

Archived from groups: alt.comp.periphs.mainboard.elitegroup (More info?)

"Buffalo" <eric(nospam)@nada.com.invalid> wrote in message

> Good job.
>
> "Eric Wolfe" <b_eric_wolfe@yah00.com> wrote in message

> > Forgive the top posting but I wanted to follow up on my original post.
I
> > just got a notice from AMD that they are sending out a new processor.

Thanks! Now I have a new problem with the machine that I *would* have torn
down to test the duff processor. (Proving that you can't beat Fate!) I
decided to connect it to the network and when I configured it as a member of
a domain, that triggered the W2K Ctrl+alt+del login screen. The problem is
that the machine (a non-Pro K7S5A) will not recognize that key combo and
just sits there. I am sure the keys themselves work because they work if I
boot from a floppy or CD.

Another oddity is that when I tried to enable the power on from keyboard
feature, it won't turn on when I enter my password. The newer K7S5a Pro
board handles both tasks with no problem using the same Lite-On Classic
keyboard. I'm about to try swapping the Lite-On keyboard with a cheaper
"Windows" keyboard to see if that makes any difference. Has anyone
experienced these problems before?

Eric