New MB; can't even get into BIOS setup

[Steve Conover]

Thanks; I tried all that, and just finished unplugging,
disassembling, and reboxing everything. Cleaned up all the misc
trash, vacuumed the carpet, and put everything in the closet.

Four days of wrestling with it is my limit. ASUS help desk said
it's one of three things: not enough power supply, bad ram, or
bad mb. $200 worth of new hardware eliminated the first two
possibilities, so it must be the third one. I just now sent ASUS
a query asking next steps for getting the mb replaced.

If I never hear "System failure due to CPU overclocking" again,
it will be too soon.

Thanks for all of your suggestions.

Steve

It has most likely been suggested somewhere in this long posting and
answers, but have you tried removing the MB from the case
and placeing it on card board ( I usually set the sheet foam packing from
the mb box on cardboard then set the MB on this. With this just setup the
essentials,
Power supply connections, 1 stick of memory, 1 floppy drive, video board,
keyboard, and mouse,. Reset or clear the bios. Remove cpu fan and cpu-clean
surface of cpu and fan and then reinstall cpu and fan (with thin even coat
of heatsink compound. After all of this try to boot and see what happens.
Sometimes just a little speck of solder or something on the mb will short to
the case and all kinds of weird problems can happen. If you can get as far
as the opening readings on the video screen hit the delete key and go into
bios and check all settings one page at a time.. Pay close attention to the
primary bios screen and set all HD settings to "Auto ", and setup floppy
drive just for now and then reboot. Let us know after you have done this and
then you may need to go page by page in the bios (with at least 1 HD
installed-no cd or DVD for now). Check the soft menu for the cpu if you can
reach it (although on your board it is supposed to reset to default settings
after the type of error message that you are getting). give it a try and see
what happens...Aloha

Delete the ZZZ.
It keeps spam
off the server.

 

[Steve Conover]

One month ago, I was trying to assemble a new system, but
couldn't get past the vocal POST message "System failed due to
CPU overclocking." Help from several of you, and subsequent
step-by-step testing led me to conclude it was a faulty MB
(P4C800-E Deluxe).

MB is back from its RMA trip; packing slip said "Problem: BIOS
Chksum Error BIOS version Error. Repair Result: Flash BIOS to
the lastest [sic] version."

I thought, "That's good, they found and fixed the problem. Now
let's build my new, speedy system."

Cautiously following the step-by-step process I'd learned before,
here's what happened:

1. (MB, PSU, Power switch): vocal POST says, "No CPU installed."

Me: "OK, now install the CPU, Arctic Silver 5 goop, and heatsink,
then we're off and running."

2. (plus CPU+goop+heatsink): vocal POST says, "System failed CPU
test."

Me: "Maybe it needs RAM to get past that one."

3a. (plus 1GB RAM): vocal POST says, "System failed CPU test."

Me: "Maybe it's bad RAM; remove it, and try the cheap RAM I
bought last month when I thought it was bad RAM."

3b. same vocal POST message.

Me: "Uh-oh, did ZipZoomFly sell me a bad CPU? This is not good,
because they have a 30-day limit on returns, and I'm way past
that."

Could it be anything besides the CPU? Does Intel have an RMA
process for P4's?

Thanks,
Steve


Delete the ZZZ.
It keeps spam
off the server.

 

[Paul]

One month ago, I was trying to assemble a new system, but
couldn't get past the vocal POST message "System failed due to
CPU overclocking." Help from several of you, and subsequent
step-by-step testing led me to conclude it was a faulty MB
(P4C800-E Deluxe).

MB is back from its RMA trip; packing slip said "Problem: BIOS
Chksum Error BIOS version Error. Repair Result: Flash BIOS to
the lastest [sic] version."

I thought, "That's good, they found and fixed the problem. Now
let's build my new, speedy system."

Cautiously following the step-by-step process I'd learned before,
here's what happened:

1. (MB, PSU, Power switch): vocal POST says, "No CPU installed."

Me: "OK, now install the CPU, Arctic Silver 5 goop, and heatsink,
then we're off and running."

2. (plus CPU+goop+heatsink): vocal POST says, "System failed CPU
test."

Me: "Maybe it needs RAM to get past that one."

3a. (plus 1GB RAM): vocal POST says, "System failed CPU test."

Me: "Maybe it's bad RAM; remove it, and try the cheap RAM I
bought last month when I thought it was bad RAM."

3b. same vocal POST message.

Me: "Uh-oh, did ZipZoomFly sell me a bad CPU? This is not good,
because they have a 30-day limit on returns, and I'm way past
that."

Could it be anything besides the CPU? Does Intel have an RMA
process for P4's?

Thanks,
Steve


Delete the ZZZ.
It keeps spam
off the server.

Is the 2x2 ATX12V power connector plugged in ?

That might be enough to stop it.

"System failed CPU test" is based on a timer used
by the Winbond Voice POST chip. The Voice POST chip
has its own crude internal processor and executes a
program stored in the same eight pin DIP chip, as the
voice samples. The timer in the Voice POST chip can be
set up by that code, and if it times out before the CPU
wakes up and clears the timer, you get the System
Failed CPU Test message. So anything which can delay
the execution of BIOS code, would be enough to prevent
the board from getting past that voice message.

Another example of a possible problem, would be a
stuck reset switch. The board does not need a
reset switch to work, so you can disconnect it. If
the reset switch is stuck in the depressed ON position,
it leaves the board in reset, and that might be enough
to do it. (But, you would think a stuck reset would
also stop the Winbond chip to, so maybe this is
a dumb idea.)

I don't know of an easy way to be able to tell whether
the processor is doing anything or not. If you owned
a BIOS POST card (a PCI/ISA card with two digit hex
display on it), you could check to see if the processor
is writing to port 80. Other than that, you would need
to see visible signs of life, like drives being
accessed, or the video card waking up etc.

You don't mention what kind of processor you've got, but
if the board were to have an older BIOS in it, and you
had a Prescott or a Celeron D, it could be the BIOS
doesn't support the processor. You can find out what
combinations work on the cpusupport web page:

http://www.asus.com.tw/support/cpusupport/cpusupport.aspx

If the slip in the box said the BIOS was updated, you
have no way of knowing whether that was done or not. Since
it sounds like you didn't send back the processor with the
motherboard, how do they know what they are fixing ?
Maybe you can assume that if there is a paper label stuck
on the BIOS chip, then that is the version of BIOS you've
got. Consult the cpusupport web page based on the paper
label, and see if the combo of processor and BIOS will
work.

If you have a spare P4 processor, test the motherboard with
that. If the board will run with a spare P4, then it means
your Zipzoomfly processor hasn't killed the board. And,
that means you could take the Zipzoomfly processor to a
friend's house, and test it on another motherboard, assuming
you don't have a mobo to test the processor with.

Unless you've zapped the new processor with static, or broken
a pin off, it is highly unlikely the new processor is busted.
I assume if the processor is a retail boxed unit, that the
plastic packaging was intact, and the seals were still on it.
If so, then the odds of the processor being dead are like
10000:1 . This is more likely to still be a BIOS problem,
and the slip in the box could be a lot of BS. If you can get
the board to run with another P4 (a Northwood, for example),
then you can flash the board up to the latest BIOS yourself.

HTH,
Paul

 

[Steven Hilgendorf]

One month ago, I was trying to assemble a new system, but
couldn't get past the vocal POST message "System failed due to
CPU overclocking." Help from several of you, and subsequent
step-by-step testing led me to conclude it was a faulty MB
(P4C800-E Deluxe).

MB is back from its RMA trip; packing slip said "Problem: BIOS
Chksum Error BIOS version Error. Repair Result: Flash BIOS to
the lastest [sic] version."

I thought, "That's good, they found and fixed the problem. Now
let's build my new, speedy system."

Cautiously following the step-by-step process I'd learned before,
here's what happened:

1. (MB, PSU, Power switch): vocal POST says, "No CPU installed."

Me: "OK, now install the CPU, Arctic Silver 5 goop, and heatsink,
then we're off and running."

2. (plus CPU+goop+heatsink): vocal POST says, "System failed CPU
test."

Me: "Maybe it needs RAM to get past that one."

3a. (plus 1GB RAM): vocal POST says, "System failed CPU test."

Me: "Maybe it's bad RAM; remove it, and try the cheap RAM I
bought last month when I thought it was bad RAM."

3b. same vocal POST message.

Me: "Uh-oh, did ZipZoomFly sell me a bad CPU? This is not good,
because they have a 30-day limit on returns, and I'm way past
that."

Could it be anything besides the CPU? Does Intel have an RMA
process for P4's?

Thanks,
Steve

If it's a Boxed (retail) processor, Intel should honer the three (3)
year warranty directly through Intel by following the instructions in
your enclosed warranty booklet, however, if it's an OEM/Tray processor
any warranty available would be handled directly through your retailer
or place of purchase:

http://support.intel.com/support/processors/warranty/qna.htm

Steven

 

[Steve Conover]

Paul,

Thanks a lot for your help. Although I did discover <blush> that
I had failed to plug in the 2x2 12V connector, after doing that
and adding the memory, it's now back to the vocal POST message
"System failed due to CPU overclocking."

Any idea what triggers the overclocking POST message?

(The processor is a 2.8GHz P4; came in a retail box.)

Thanks,
Steve

Is the 2x2 ATX12V power connector plugged in ?

That might be enough to stop it.

"System failed CPU test" is based on a timer used
by the Winbond Voice POST chip. The Voice POST chip
has its own crude internal processor and executes a
program stored in the same eight pin DIP chip, as the
voice samples. The timer in the Voice POST chip can be
set up by that code, and if it times out before the CPU
wakes up and clears the timer, you get the System
Failed CPU Test message. So anything which can delay
the execution of BIOS code, would be enough to prevent
the board from getting past that voice message.

Another example of a possible problem, would be a
stuck reset switch. The board does not need a
reset switch to work, so you can disconnect it. If
the reset switch is stuck in the depressed ON position,
it leaves the board in reset, and that might be enough
to do it. (But, you would think a stuck reset would
also stop the Winbond chip to, so maybe this is
a dumb idea.)

I don't know of an easy way to be able to tell whether
the processor is doing anything or not. If you owned
a BIOS POST card (a PCI/ISA card with two digit hex
display on it), you could check to see if the processor
is writing to port 80. Other than that, you would need
to see visible signs of life, like drives being
accessed, or the video card waking up etc.

You don't mention what kind of processor you've got, but
if the board were to have an older BIOS in it, and you
had a Prescott or a Celeron D, it could be the BIOS
doesn't support the processor. You can find out what
combinations work on the cpusupport web page:

http://www.asus.com.tw/support/cpusupport/cpusupport.aspx

If the slip in the box said the BIOS was updated, you
have no way of knowing whether that was done or not. Since
it sounds like you didn't send back the processor with the
motherboard, how do they know what they are fixing ?
Maybe you can assume that if there is a paper label stuck
on the BIOS chip, then that is the version of BIOS you've
got. Consult the cpusupport web page based on the paper
label, and see if the combo of processor and BIOS will
work.

If you have a spare P4 processor, test the motherboard with
that. If the board will run with a spare P4, then it means
your Zipzoomfly processor hasn't killed the board. And,
that means you could take the Zipzoomfly processor to a
friend's house, and test it on another motherboard, assuming
you don't have a mobo to test the processor with.

Unless you've zapped the new processor with static, or broken
a pin off, it is highly unlikely the new processor is busted.
I assume if the processor is a retail boxed unit, that the
plastic packaging was intact, and the seals were still on it.
If so, then the odds of the processor being dead are like
10000:1 . This is more likely to still be a BIOS problem,
and the slip in the box could be a lot of BS. If you can get
the board to run with another P4 (a Northwood, for example),
then you can flash the board up to the latest BIOS yourself.

HTH,
Paul

Delete the ZZZ.
It keeps spam
off the server.

 

[Paul]

Paul,

Thanks a lot for your help. Although I did discover <blush> that
I had failed to plug in the 2x2 12V connector, after doing that
and adding the memory, it's now back to the vocal POST message
"System failed due to CPU overclocking."

Any idea what triggers the overclocking POST message?

(The processor is a 2.8GHz P4; came in a retail box.)

Thanks,
Steve

I don't really know for sure. It could be something simple,
like a CMOS bit that is cleared at the beginning of the POST,
and is set just as the computer starts to boot. If the
BIOS code doesn't run to completion, then the value of the
bit can be checked at the beginning of the next POST, at
which time the overclocking message will be given. Many
of the Voice POST messages are commanded by the processor,
and only a few run via the timer, or by an actual hardware
signal to the Winbond chip.

Something is preventing the BIOS from completing properly,
that is about all that can be concluded from the message.
Stripping the computer down to the minimum amount of
hardware, is as close as you'll get to isolating what
is causing it. And, swapping processor and RAM, is an
expensive option. It is times like this that you need
a well stocked junk parts cabinet. (I had one build
where I bought duplicates of everything except the
motherboard :-(

If you had a PCI/ISA POST card (the one with a two
digit hex display), you can watch the routine numbers
flash by, and there are tables available, that map the
Port 80 codes sent to the POST card, to routines
in the BIOS. A POST card can be had pretty cheaply
from Ebay now, as some traders sell batches of them
new on Ebay. At one time, POST cards were expensive,
considering the low tech chips on the card. The POST
card is useful if the BIOS runs for a few seconds and
then stops dead - if the board is bootable, then the
info is useless. (POST cards are good for faults where
the machine gets stuck.)

The "system failed" message is reporting what the board
thinks happened on the _last_ POST, and not the
current one. The presumption is the board crashed on the
last BIOS POST, and the BIOS is using safe values on
the current boot, to allow you to enter the BIOS and
"fix' the overclocking problem. I have an A7N8X-E that
will report this from boot cycle to boot cycle, and the
board still works. On that board, for me, the presence
of the message indicates something is still not right
with the board, and if I am overclocking, it means I
am probably pushing the board too far - note that in
my case, it doesn't prevent me from using the board.
I think some people's Intel boards are the same way,
you can be getting that message and still use the
board with no problems. The message doesn't guarantee
you've got a problem, and I would treat it as a gentle
hint that something isn't right.

If you can successfully boot the board, run memtest86 from
memtest.org to test for solid faults in the memory, then
boot into Windows and run Prime95 (mersenne.org) to test
for pattern sensitivity in the memory, or problems with
the processor and/or Northbridge on the motherboard. A
benchmark like 3DMark is good for accelerating video card
or AGP slot problems.

Those kinds of tests are the things you can run, to determine
whether the "System failed..." message indicates a real
problem or not.

HTH,
Paul

 

[Steve Conover]

Problem solved. It was a bad motherboard; I bought another
identical motherboard today at Fry's, took it one step at a time
with no problems, and I just made it through the BIOS setup.
Tomorrow I'll load the OS (XP).

Thanks to all who helped me diagnose this situation over the last
month or so, especially Paul.

Kinda ticks me off that ASUS' RMA process not only failed to
discover that the first mb was bad, but also that they reported
the problem "solved" after flashing the latest BIOS.

ASUS will be hearing from me on this fiasco soon, and so will
ZipZoomFly. (I paid ZZF $174 in return for a piece of junk and
thirty-plus days of frustration.) Buying two motherboards to get
one that works was not on my agenda or in my budget this year; I
guess I should be thankful it wasn't the processor.

Thanks again, folks.
Steve


Delete the ZZZ.
It keeps spam
off the server.

 

[mapmaker]

Problem solved. It was a bad motherboard; I bought another
identical motherboard today at Fry's, took it one step at a time
with no problems, and I just made it through the BIOS setup.
Tomorrow I'll load the OS (XP).

Thanks to all who helped me diagnose this situation over the last
month or so, especially Paul.

Kinda ticks me off that ASUS' RMA process not only failed to
discover that the first mb was bad, but also that they reported
the problem "solved" after flashing the latest BIOS.

ASUS will be hearing from me on this fiasco soon, and so will
ZipZoomFly. (I paid ZZF $174 in return for a piece of junk and
thirty-plus days of frustration.) Buying two motherboards to get
one that works was not on my agenda or in my budget this year; I
guess I should be thankful it wasn't the processor.

Thanks again, folks.
Steve


Delete the ZZZ.
It keeps spam
off the server.

Thank You Steve,
....and Paul,
For posting this complete, clear, running commentary on this problem
and it's solution. It is so pleasant to read the logical progression
and solution of the issue even though I have not had this problem
myself. This thread will remain with me for some time to come.
Thanks again.
MapMaker