From: Dion Macale on 7 Feb 2006 17:40
this error appears when i try to boot up windows XP home, SP2
"New CPU installed! Please enter setup
to configure your system."
"Press F1 to enter setup"
"Press F2 to load default values and continue"
the mobo is an Asus P4P800SE, and the cpu is a s478 2.4ghz northwood.
it only happens with this particular CPU.
i have all the correct jumper settings and the CMOS battery is fine.
any help would be appreciated
From: Custom Computers on 7 Feb 2006 18:12
Did you press F-1 to enter setup and make sure your CPU Operating
Frequency is set correctly?
From: Paul on 8 Feb 2006 02:32
In article <1139352014.178055.163400(a)g47g2000cwa.googlegroups.com>, "Dion
Macale" <dionmacale(a)gmail.com> wrote:
> Gday all,
> this error appears when i try to boot up windows XP home, SP2
> "New CPU installed! Please enter setup
> to configure your system."
> "Press F1 to enter setup"
> "Press F2 to load default values and continue"
> the mobo is an Asus P4P800SE, and the cpu is a s478 2.4ghz northwood.
> it only happens with this particular CPU.
> i have all the correct jumper settings and the CMOS battery is fine.
> any help would be appreciated
I have heard of this happening before, but usually it happens with
some obscure model of high end processor. I also haven't seen a
post back from anyone, claiming that they fixed it. I can give you
a little background info, which may suggest some things to try,
but don't hold your breath in terms of a cure.
There are several "processor recognition" issues that the
BIOS wants to solve. The first one is the obvious one - identify
the processor type, so that hardware settings in the processor
and/or chipset can be applied, such that the processor will operate
properly. Perhaps that might include enabling the L1 and L2
cache, setting necessary frequencies (if the BSEL pins didn't do
their thing properly) and so on.
A second flavor of "processor recognition", is the loading of
microcode patches. A microcode patch is a series of corrections
to the operation of the processor, which is loaded into some
RAM on the processor. If the processor likes the patch (patch
has a good checksum and appears to be for that model of processor),
it will accept an attempt to load a microcode. If no microcode
patch works, you may see a warning on the screen, but I don't
think there are any other consequences (except for the obvious
one, of the processor not computing correctly). An OS like
WinXP also has a microcode loader, and it is possible for the
OS to load a later microcode patch, if one is stored in the OS.
The third issue, is a bookkeeping type issue. There are areas of
the flash, called DMI/ESCD, that the BIOS writes the current
hardware details. A tool like Asus Probe or perhaps some earlier
utility, would have a "DMI Explorer" tab, that can read the
contents of what is stored in DMI/ESCD. I suspect the code that
handles DMI/ESCD is responsible for what you are seeing.
The way the processor might work, is when the BIOS starts to
run, it looks to the DMI/ESCD. Say that the word "Celeron" is
written in DMI/ESCD, and the BIOS finds "Pentium 4" in the
motherboard socket. The BIOS will rewrite the DMI/ESCD area
of the BIOS flash chip, with the new information, so that the
next time the computer starts, the word "Pentium 4" is recorded
in that bookkeeping area.
Now, say the DMI/ESCD code recognizes that the old info is
not equal to the new info, and yet the update of DMI/ESCD
fails. Maybe it fails because the code cannot properly handle
the information it is getting from the processor. Perhaps the
attempt to flash the DMI/ESCD segment is failing altogether.
So that suggests, the flashing the BIOS will attempt to overwrite
the DMI/ESCD. Perhaps flashing the BIOS will be enough to
get the BIOS to compute the DMI/ESCD properly. I cannot see
a Northwood failing to be recognized, either from a config or
a microcode point of view. I suppose you could use BIOS tools
to extract the microcode and examine what patches are loaded,
but I wouldn't really expect to see support for a 2.4GHz
Northwood to be missing.
I don't think clearing the CMOS will help here, as I cannot
imagine there being enough room in the CMOS to hold anything
that would screw up the process. But if you run out of things
to try, I suppose it might be worth a shot. (Unplug the
computer before following the procedure in the manual, so
that +5VSB cannot damage anything. Some motherboard designs
are not "+5VSB safe", and always unplugging removes the
uncertainty about the potential for damage.)
If you do find a solution, please post back what you find.
From: Dion Macale on 9 Feb 2006 17:19
i just flashed the bios up to version 1006, as i did not want to
install a more recent beta version, however the problem persists.
i am not really sure where to go from here
From: sdlomi2 on 10 Feb 2006 22:22
"Dion Macale" <dionmacale(a)gmail.com> wrote in message
> i just flashed the bios up to version 1006, as i did not want to
> install a more recent beta version, however the problem persists.
> i am not really sure where to go from here
Have you "pressed F2 to continue" to see if it will then boot? If so,
it will be running at minimal output due to its automatically setting
parameters to minimal output in cmos in order to 'ensure' a boot; but it
would at least allow you to enter cmos and attempt tweaking from that point.
HTH & good luck, s