A7N8X-E failure?

  • Thread starter Thread starter Salty
  • Start date Start date
S

Salty

I noticed that my memory wasn't running at full speed and thought I
would speed it up in the BIOS. Shouldn't of attempted it when home
from work, tired and kids about, but I did. Anyway I think I increased
the FSB to 200MHz (I have Athlon XP 2800)and so when I tried to boot
got a POST about CPU overclocking. Tried to restart again and got the
same. I now dont get any POST and PC won't start. I have tried
resetting the CMOS. How can I get it started? If not what is the most
likely thing to have been killed?
 
I thought that I would still get a POST message to
say something?
Click to expand...

Huh? If the cpu is gone, that is like your brain is dead but you want the
patient to say, 'hey doctor, my brain no longer functions'.

Whether OS code is executing or bios code or any other compiled code, you
got to have a cpu to execute it.

-g
 
I noticed that my memory wasn't running at full speed and thought I
would speed it up in the BIOS. Shouldn't of attempted it when home
from work, tired and kids about, but I did. Anyway I think I increased
the FSB to 200MHz (I have Athlon XP 2800)and so when I tried to boot
got a POST about CPU overclocking. Tried to restart again and got the
same. I now dont get any POST and PC won't start. I have tried
resetting the CMOS. How can I get it started? If not what is the most
likely thing to have been killed?
Click to expand...

I didn't get any post on my "A7N8X" one time when fooling around, but
this fixed it for me.

Step 1.
Unplug PC
Remove CMOS battery
Set CMOS jumper to Clear
Remove CPU

Step 2.
Set CMOS jumper to Normal
Replace CMOS battery
Plug in PC and power up.
(I think at this point all I got was beeps)

Step 3.
Unplug PC
Install CPU + HS/fan
Plug in PC and power up.

hth,
Ed
 
Salty said:
I noticed that my memory wasn't running at full speed and thought I
would speed it up in the BIOS. Shouldn't of attempted it when home
from work, tired and kids about, but I did. Anyway I think I
increased the FSB to 200MHz (I have Athlon XP 2800)and so when I
tried to boot got a POST about CPU overclocking. Tried to restart
again and got the same. I now dont get any POST and PC won't start.
I have tried resetting the CMOS. How can I get it started? If not
what is the most likely thing to have been killed?
Click to expand...

First, your cpu is not dead. It takes considerably more to do that.

Tell me how you resetted CMOS. Did you use the jumper on board? Was power
cable removed from the pc at the time, if not, pull and repeat.
Did you place back the "clear cmos" jumper as it was before? If not, do so.

One possibility is your bios data might have gone bad.

I've been there myself, overclocking too much, but a proper clear cmos
brought me back on track again.

In any case, if you do get your pc up and running again, leave memory speed
in sync with fsb. Do not set to 200, if you use 166MHz fsb. Performance will
be worse.
 
I used the jumper on board and replaced it afterwards. I left the
power cable connected, but no power in it for an earth. I will have a
go again and remove the power cable from the board, etc. Thanks for
the advice I will see how I get on. I will try this before removing
the CPU as I don't want the hassle of redoing the thermal paste, but
might have to anyway.

Gilbert, one of the POST messages is "No CPU installed". It must be
able to do this without a CPU installed for a brain.

Thanks for the help.
 
I used the jumper on board and replaced it afterwards. I left the
power cable connected, but no power in it for an earth. I will have a
go again and remove the power cable from the board, etc. Thanks for
the advice I will see how I get on. I will try this before removing
the CPU as I don't want the hassle of redoing the thermal paste, but
might have to anyway.

Gilbert, one of the POST messages is "No CPU installed". It must be
able to do this without a CPU installed for a brain.

Thanks for the help.
 
I thought that I would still get a POST message to
say something?
Click to expand...

Huh? If the cpu is gone, that is like your brain is dead but you want the
patient to say, 'hey doctor, my brain no longer functions'.

Whether OS code is executing or bios code or any other compiled code, you
got to have a cpu to execute it.

-g[/QUOTE]

The Winbond Vocal POST chip contains its own processor. It can
run on a completely stripped motherboard. If you take a bare
A7N8X-E Deluxe, with no components installed on it, connect a
PSU and a power switch, then plug amplified speakers into the
green Lineout connector, you'll hear "No CPU installed".

The Winbond chip has its stored program in an adjacent eight
pin DIP chip. The eight pin chip is a serial flash memory, and
contains voice samples and the computer program the Winbond chip
runs for itself.

Paul
 
The Winbond Vocal POST chip contains its own processor. It can
run on a completely stripped motherboard.
Click to expand...

Any idea what the processor is? Intel, motorola, or ?

I wonder what the issues are in building a board like that because the
processor executes code initially then, at some point, hands execution off
to the cpu.

-g
 
The Winbond Vocal POST chip contains its own processor. It can
run on a completely stripped motherboard.
Click to expand...

Any idea what the processor is? Intel, motorola, or ?

I wonder what the issues are in building a board like that because the
processor executes code initially then, at some point, hands execution off
to the cpu.

-g[/QUOTE]

It doesn't share with the AMD CPU. It has a fixed task, which is
to monitor some hardware error signals (such as the signal that
tells it no CPU is in the ZIF socket), and it also has a queue
for software events delivered by the main CPU. When there are
no other components installed in the board, about all it can
do, is tell you "No CPU installed". The vast majority of
other error messages are only emitted at the command of the
main CPU (which is AMD AthlonXP in this case).

It is very slow - an instruction executes in 400 microseconds.
And this datasheet, like others from Winbond, doesn't do the
part justice. It doesn't describe how to program the device.
That is something I had to figure out, by looking at a couple
of files that come with the Voice Editor software (and I
didn't understand very much of what I saw). That
software is downloadable from the Asus site.

Datasheet for the Winbond chip:
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/W83791SDSGa.pdf

The two Voice Editor files are listed here:
http://groups.google.ca/group/alt.comp.periphs.mainboard.asus/msg/af9eb7c50228db72

The main processor communicates with it via I2C, which is the
same thing as SMBUS. SMBUS only runs at 10Kbits/sec, which again
means that communications from the Athlon processor, to the
Winbond chip, is slow as molasses.

The Winbond chip has a timer, and can deliver a Voice message if
the timer times out. Some of the tests that the main processor
carries out on the motherboard, involve setting the timer inside
the Winbond, and then starting the test. If the code on the
main processor crashes, the timer on the Winbond expires, and
a voice message is delivered. If the main processor test code
completes OK, the main processor clears the timer on the Winbond
chip, so no error message is emitted. So, while the Winbond chip
is semi-intelligent, it cannot run the whole motherboard. For
the rest of the error messages, the main processor writes a number
in the queue of the Winbond, and the Winbond then plays that
particular voice sample.

The part is dirt cheap, and appears on a lot of the "deluxe"
versions of Asus motherboards. It is handy for debugging, but
not as good as being able to see port 80 error codes.

Paul
 
Well, I have reset the CMOS and still nothing. The hard drive starts,
all fans work but no POST and the monitor is not started. I could do
with finding the fault. Would a CPU failure cause this? or is it most
likely to be the board?

Ta.
 
Which BIOS version do you have ?
I ask because I had that problem with v ersion 1009 (1109?)
and fried my CPU. That version was removed a few days
later from Asus site.
 
Salty said:
I used the jumper on board and replaced it afterwards. I left the
power cable connected, but no power in it for an earth. I will have a
go again and remove the power cable from the board, etc. Thanks for
the advice I will see how I get on. I will try this before removing
the CPU as I don't want the hassle of redoing the thermal paste, but
might have to anyway.

Gilbert, one of the POST messages is "No CPU installed". It must be
able to do this without a CPU installed for a brain.

Thanks for the help.
Click to expand...
I would suggest you disconnect the AC power cord from the case and then
press and hold the power button for a few seconds to drain any residual
charge contained within the capacitors. Then move the CMOS jumper to
the clear position for a few seconds and finally return CMOS jumper to
normal location. Reconnect the AC and give it another try.

hth,

PB
 
Salty said:
Well, I have reset the CMOS and still nothing. The hard drive starts,
all fans work but no POST and the monitor is not started. I could do
with finding the fault. Would a CPU failure cause this? or is it most
likely to be the board?

Ta.
Click to expand...

If one of the messages is "No CPU installed", I would remove
the heatsink and have a look at the CPU. Pull the CPU from the
socket and reseat it. (Use an ESD wrist strap or similar
protection against static electricity.) Check for burn marks
on the heatsink, which would suggest the thing overheated. Use
some fresh paste and give it another try.

If it still won't start, see if you can borrow a S462 processor
somewhere, to test out the board.

Early A7N8X boards used to have problems with the flash chip
getting corrupted, and reflashing the chip could bring a
board back to life (sometimes referred to as the "BIOS - Save
- Death" bug). The A7N8X-E doesn't seem to have that problem,
or at least I've never seen any complaints for the A7N8X-E like
there were for the earlier boards.

The later Asus AMD boards have COP (CPU overheat protection),
so the computer should be able to shut off in the event the
heatsink is not making good contact. It really shouldn't be
possible to fry the CPU, as COP reads the thermal diode on
the CPU to detect overheating. It would take a COP failure,
and a CPU overheat event, to burn a CPU on your board.

Paul
 
Thanks for everyones help. The problem lay in one of the memory
modules. It must have failed. I thought that the POST might have
helped there! I don't know what happened as the memory should have
been able to take the higher clockspeed as it was its rated one. May
be a coincidence.

Thanks
 
Back
Top