Wrong CPU being shown by the BIOS

[ipleb]

Hi all

I have an Asus P5N motherboard with an Intel Core 2 Duo E4300 1.86ghz
Socket 775 FSB800 CPU. All was working fine ... until recently when I
installed an extra gig of RAM which put the system total at 3gig - the
PC runs Vista home pre.

Now I am not sure if installing the RAM did this or messing about in
the OC settings of the bios, but everytime the PC boots it halts at
the BIOS screen with a message sayinh.

"Unknown CPU detected, install latest BIOS to unleash Power"

Now I have the latest BIOS installed but the problem seems to be that
the CPU is coming up as an Intel Xeon [email protected] ghz, 2 CPUs detected.

Vista also shows that an XEON is installed

Can anyone think why this has happened?

Thanks

iPleb

 

[Jan Alter]

Hi all

I have an Asus P5N motherboard with an Intel Core 2 Duo E4300 1.86ghz
Socket 775 FSB800 CPU. All was working fine ... until recently when I
installed an extra gig of RAM which put the system total at 3gig - the
PC runs Vista home pre.

Now I am not sure if installing the RAM did this or messing about in
the OC settings of the bios, but everytime the PC boots it halts at
the BIOS screen with a message sayinh.

"Unknown CPU detected, install latest BIOS to unleash Power"

Now I have the latest BIOS installed but the problem seems to be that
the CPU is coming up as an Intel Xeon [email protected] ghz, 2 CPUs detected.

Vista also shows that an XEON is installed

Can anyone think why this has happened?

Thanks

iPleb

Try changing the bios settings back the way they were.

 

[Dave]

Hi all

I have an Asus P5N motherboard with an Intel Core 2 Duo E4300 1.86ghz
Socket 775 FSB800 CPU. All was working fine ... until recently when I
installed an extra gig of RAM which put the system total at 3gig - the
PC runs Vista home pre.

Now I am not sure if installing the RAM did this or messing about in
the OC settings of the bios, but everytime the PC boots it halts at
the BIOS screen with a message sayinh.

"Unknown CPU detected, install latest BIOS to unleash Power"

Now I have the latest BIOS installed but the problem seems to be that
the CPU is coming up as an Intel Xeon [email protected] ghz, 2 CPUs detected.

Vista also shows that an XEON is installed

Can anyone think why this has happened?

Thanks

Yes, the BIOS makes a guess as to what CPU is installed based in part on
clock frequency and multiplier, along with other settings you might "tweak"
when you are overclocking. Apparently, in messing around with the OC
settings, you have stumbled upon settings that closely match the specs of a
Xeon. -Dave

 

[ipleb]

Yes, the BIOS makes a guess as to what CPU is installed based in part on
clock frequency and multiplier, along with other settings you might "tweak"
when you are overclocking.  Apparently, in messing around with the OC
settings, you have stumbled upon settings that closely match the specs ofa
Xeon.   -Dave

Hmmmmm

Do you know how I could work out what the original settings the actual
CPU would be?

 

[Jan Alter]

Yes, the BIOS makes a guess as to what CPU is installed based in part on
clock frequency and multiplier, along with other settings you might
"tweak"
when you are overclocking. Apparently, in messing around with the OC
settings, you have stumbled upon settings that closely match the specs of
a
Xeon. -Dave

Hmmmmm

Do you know how I could work out what the original settings the actual
CPU would be?

Try going into the bios and use 'default settings'
Then after you've got a confirmed startup of the system go back into the
bios and look at what the cpu settings have been adjusted by the bios
itself. That is usually a safe starting point.

 

[Paul]

Hmmmmm

Do you know how I could work out what the original settings the actual
CPU would be?

Just for chuckles, (assuming you get it running again), why not run the
Intel Processor Identification Utility.

http://www.intel.com/support/processors/tools/piu/sb/CS-014921.htm

(Supported processors list)
http://www.intel.com/support/processors/tools/piu/sb/cs-015472.htm

That utility should be using the ideas in this document. Intel
provides assembler and C code identification sequences, for their
processors. BIOS designers are supposed to incorporate these
ideas, as part of the BIOS design.

"AP-485 Intel Processor Identification and the CPUID Instruction"
http://www.intel.com/assets/pdf/appnote/241618.pdf

Processor identification is also addressed here, in the thousands
of pages of docs.

http://support.intel.com/products/processor/manuals/index.htm

*******
As an example, the Xeon X3220 and the Q6600, are suspected to be
very similar. They're both LGA775, and overclockers have tried
both of them in desktop boards.

Here is an X3220 -

http://www.ocforums.com/showthread.php?p=5310073#post5310073

And here, a Q6600 -

http://i22.tinypic.com/2qn8dn9.png

So while the basic CPUID info looks the same, the Intel utility
can still tell the difference. There must be a secondary indicator
somewhere, that is giving it away. And it won't be solely frequency
- the multiplier and multiplier range prevent overclocking from
hiding the identity completely. (I'm not sure if BSEL is
protected and available inside the processor or not - BSEL
are the pins on the processor, that announce the nominal
clock required.)

So I'd be curious what the Intel Utility (and perhaps CPUZ) think.

Paul

 

[david]

Hmmmmm

Do you know how I could work out what the original settings the actual
CPU would be?

Check your motherboard documentation. Most MBs have a jumper that can be
set,and removing the battery for a minute or so will clear the settings
to the factory default. Don't forget to put the jumper back to where it
was.

 

[ipleb]

Check your motherboard documentation.  Most MBs have a jumper that can be
set,and removing the battery for a minute or so will clear the settings
to the factory default.  Don't forget to put the jumper back to where it
was.

Hi there ...

I have done the removing the battery and jumper combo but to no avail

I plan to remove the RAM next to see what happend

Thanks

 

[ipleb]

Just for chuckles, (assuming you get it running again), why not run the
Intel Processor Identification Utility.

http://www.intel.com/support/processors/tools/piu/sb/CS-014921.htm

(Supported processors list)http://www.intel.com/support/processors/tools/piu/sb/cs-015472.htm

That utility should be using the ideas in this document. Intel
provides assembler and C code identification sequences, for their
processors. BIOS designers are supposed to incorporate these
ideas, as part of the BIOS design.

"AP-485 Intel Processor Identification and the CPUID Instruction"http://www.intel.com/assets/pdf/appnote/241618.pdf

Processor identification is also addressed here, in the thousands
of pages of docs.

http://support.intel.com/products/processor/manuals/index.htm

*******
As an example, the Xeon X3220 and the Q6600, are suspected to be
very similar. They're both LGA775, and overclockers have tried
both of them in desktop boards.

Here is an X3220 -

http://www.ocforums.com/showthread.php?p=5310073#post5310073

And here, a Q6600 -

http://i22.tinypic.com/2qn8dn9.png

So while the basic CPUID info looks the same, the Intel utility
can still tell the difference. There must be a secondary indicator
somewhere, that is giving it away. And it won't be solely frequency
- the multiplier and multiplier range prevent overclocking from
hiding the identity completely. (I'm not sure if BSEL is
protected and available inside the processor or not - BSEL
are the pins on the processor, that announce the nominal
clock required.)

So I'd be curious what the Intel Utility (and perhaps CPUZ) think.

    Paul

Cheers - I will give this a go and see what happens

 

[ipleb]

Just for chuckles, (assuming you get it running again), why not run the
Intel Processor Identification Utility.

http://www.intel.com/support/processors/tools/piu/sb/CS-014921.htm

(Supported processors list)http://www.intel.com/support/processors/tools/piu/sb/cs-015472.htm

That utility should be using the ideas in this document. Intel
provides assembler and C code identification sequences, for their
processors. BIOS designers are supposed to incorporate these
ideas, as part of the BIOS design.

"AP-485 Intel Processor Identification and the CPUID Instruction"http://www.intel.com/assets/pdf/appnote/241618.pdf

Processor identification is also addressed here, in the thousands
of pages of docs.

http://support.intel.com/products/processor/manuals/index.htm

*******
As an example, the Xeon X3220 and the Q6600, are suspected to be
very similar. They're both LGA775, and overclockers have tried
both of them in desktop boards.

Here is an X3220 -

http://www.ocforums.com/showthread.php?p=5310073#post5310073

And here, a Q6600 -

http://i22.tinypic.com/2qn8dn9.png

So while the basic CPUID info looks the same, the Intel utility
can still tell the difference. There must be a secondary indicator
somewhere, that is giving it away. And it won't be solely frequency
- the multiplier and multiplier range prevent overclocking from
hiding the identity completely. (I'm not sure if BSEL is
protected and available inside the processor or not - BSEL
are the pins on the processor, that announce the nominal
clock required.)

So I'd be curious what the Intel Utility (and perhaps CPUZ) think.

    Paul

Hi there Paul

I have run that app and it is still reporting an Intel XEON
processor ......

CUT/PASTE ========================

Intel(R) Processor Identification Utility
Version: 3.9.20080910
Time Stamp: 2008/09/30 14:30:26
Number of processors in system: 1
Current processor: #1
Cores per processor: 2
Disabled cores per processor: 0
Processor Name: Intel(R) Xeon(R) CPU 4812 @ 1.60GHz
Type: 0
Family: 6
Model: F
Stepping: 2
Revision: 0
Maximum CPUID Level: A
L1 Instruction Cache: 2 x 32 KB
L1 Data Cache: 2 x 32 KB
L2 Cache: 2 MB
Packaging: LGA771
EIST: Yes
MMX(TM): Yes
SSE: Yes
SSE2: Yes
SSE3: Yes
SSE4: No
Enhanced Halt State: No
Execute Disable Bit: Yes
Hyper-Threading Technology: No
Intel(R) 64 Architecture: Yes
Intel(R) Virtualization Technology: Yes
Expected Processor Frequency: 1.60 GHz
Reported Processor Frequency: 1.60 GHz
Expected System Bus Frequency: 800 MHz
Reported System Bus Frequency: 800 MHz

CUT/PASTE ========================

This data doesn't help me out much :-(

Simon

 

[jaster]

Hi there ...

I have done the removing the battery and jumper combo but to no avail

I plan to remove the RAM next to see what happend

Thanks

Are you sure you installed the right bios upgrade?

Is the new memory module compatible with the original modules?

Does your motherboard support 3 memory modules or requires paired of
memory? Maybe it works fine with 2 or 4 DDR but doesn't work with 3?

 

[Paul]

Hi there Paul

I have run that app and it is still reporting an Intel XEON
processor ......

CUT/PASTE ========================

Intel(R) Processor Identification Utility
Version: 3.9.20080910
Time Stamp: 2008/09/30 14:30:26
Number of processors in system: 1
Current processor: #1
Cores per processor: 2
Disabled cores per processor: 0
Processor Name: Intel(R) Xeon(R) CPU 4812 @ 1.60GHz
Type: 0
Family: 6
Model: F
Stepping: 2
Revision: 0
Maximum CPUID Level: A
L1 Instruction Cache: 2 x 32 KB
L1 Data Cache: 2 x 32 KB
L2 Cache: 2 MB
Packaging: LGA771
EIST: Yes
MMX(TM): Yes
SSE: Yes
SSE2: Yes
SSE3: Yes
SSE4: No
Enhanced Halt State: No
Execute Disable Bit: Yes
Hyper-Threading Technology: No
Intel(R) 64 Architecture: Yes
Intel(R) Virtualization Technology: Yes
Expected Processor Frequency: 1.60 GHz
Reported Processor Frequency: 1.60 GHz
Expected System Bus Frequency: 800 MHz
Reported System Bus Frequency: 800 MHz

CUT/PASTE ========================

This data doesn't help me out much :-(

Simon

There is a field in the "Branding String" with
provision for indicating a Xeon. The above doesn't provide
the raw data from the processor, to see whether that is
what is being returned.

One thing that concerns me, is "Revision: 0". That means
the processor is not getting a microcode patch installed
by the BIOS. It implies that the BIOS is not fully set up
to support E4300. The BIOS really doesn't know what processor
it is.

I would check the BIOS version installed. If you go here, you
can see what version is needed for your board. Enter the
model number, like P5N-D, and see what minimum BIOS version
is needed.

http://support.asus.com/cpusupport/cpusupport.aspx?SLanguage=en-us

The other possibility is an internal processor failure, but
only a failure in the branding information seems highly
unlikely.

The Intel utility identification of Intel(R) Xeon(R) CPU 4812 @ 1.60GHz
is also bogus, as I tried looking on processorfinder.intel.com
and there is no X4812 that I could find. So even the Intel
utility is being fooled by something. But exactly
what, I cannot tell you.

I checked the pinout on Xeon LGA775 and Core2 LGA775 socket,
and they seem to be identical. I was thinking that perhaps
the processor wasn't making good contact with the socket,
but that doesn't appear to be a possible cause, as the pinouts
are identical. So either something is screwed up inside the
processor itself, or the BIOS has a hand in this. I would not
expect a clear CMOS to make a difference, as I don't recall
a place in the CMOS RAM structure, to store processor info.
That normally gets stored (somewhere) in DMI/ESCD. It has
to be stored, so the BIOS can detect a processor change, if
one occurs.

Your Intel processor should be provided with a warranty from
Intel. Perhaps an Intel tech support person could comment
on whether they've heard of this before or not - a change
in ID info. You would think, that at least the Intel PIU
utility would get the ID right, unless the info coming
from the processor, really is screwed up.

Paul

 

[ipleb]

There is a field in the "Branding String" with
provision for indicating a Xeon. The above doesn't provide
the raw data from the processor, to see whether that is
what is being returned.

One thing that concerns me, is "Revision: 0". That means
the processor is not getting a microcode patch installed
by the BIOS. It implies that the BIOS is not fully set up
to support E4300. The BIOS really doesn't know what processor
it is.

I would check the BIOS version installed. If you go here, you
can see what version is needed for your board. Enter the
model number, like P5N-D, and see what minimum BIOS version
is needed.

http://support.asus.com/cpusupport/cpusupport.aspx?SLanguage=en-us

The other possibility is an internal processor failure, but
only a failure in the branding information seems highly
unlikely.

The Intel utility identification of Intel(R) Xeon(R) CPU 4812 @ 1.60GHz
is also bogus, as I tried looking on processorfinder.intel.com
and there is no X4812 that I could find. So even the Intel
utility is being fooled by something. But exactly
what, I cannot tell you.

I checked the pinout on Xeon LGA775 and Core2 LGA775 socket,
and they seem to be identical. I was thinking that perhaps
the processor wasn't making good contact with the socket,
but that doesn't appear to be a possible cause, as the pinouts
are identical. So either something is screwed up inside the
processor itself, or the BIOS has a hand in this. I would not
expect a clear CMOS to make a difference, as I don't recall
a place in the CMOS RAM structure, to store processor info.
That normally gets stored (somewhere) in DMI/ESCD. It has
to be stored, so the BIOS can detect a processor change, if
one occurs.

Your Intel processor should be provided with a warranty from
Intel. Perhaps an Intel tech support person could comment
on whether they've heard of this before or not - a change
in ID info. You would think, that at least the Intel PIU
utility would get the ID right, unless the info coming
from the processor, really is screwed up.

    Paul

Hi Paul - thanks for your comments - I have also tried a small app
called CPUID - this gave the following data:

it tells me I am running an Intel Xeon 5110 - a socket 771 lga package
- weird as my mobo is obviously a socket 775

The core speed is at 1600 with a multiplyer of x8.0 bus speed at
200mhz and a rated FSB of 800mhz

I have two ddr2 533 pc2 4300 1gig chips and two ddr2 533 pc2 4300
512mb chips to give a total of 3gigs
The DRAM freq is at 266/7 and the max bandwidth of each chip is
pc2-4300 (266Mhz)

Does that shed any light?

It is definately the right bios update, and I can't remember whether
the problem started post bios update or post ram install

Finally - if the PC boots ok and run ok and all seems good - in real
terms does it actually make a difference that it is showing as an XEON
cpu and not a Core 2Duo ???

Thanks
Simon

 

[Paul]

Hi Paul - thanks for your comments - I have also tried a small app
called CPUID - this gave the following data:

it tells me I am running an Intel Xeon 5110 - a socket 771 lga package
- weird as my mobo is obviously a socket 775

The core speed is at 1600 with a multiplyer of x8.0 bus speed at
200mhz and a rated FSB of 800mhz

I have two ddr2 533 pc2 4300 1gig chips and two ddr2 533 pc2 4300
512mb chips to give a total of 3gigs
The DRAM freq is at 266/7 and the max bandwidth of each chip is
pc2-4300 (266Mhz)

Does that shed any light?

It is definately the right bios update, and I can't remember whether
the problem started post bios update or post ram install

Finally - if the PC boots ok and run ok and all seems good - in real
terms does it actually make a difference that it is showing as an XEON
cpu and not a Core 2Duo ???

Thanks
Simon

I agree with your basic idea, that this is irrelevant. For the most
part it is. But there is still the issue of the "Revision" listed for
the processor.

Microcode patches are issued by Intel, to correct errors in the processor.
Every processor has a list of around a hundred minor to serious problems.
Usually they don't affect the processor that much, and maybe you'd survive
the boot sequence even if the microcode patch was not installed. When
the processor is released, Intel should have a microcode patch from
day one, for the problems they've discovered since the masks were generated.

There are two opportunities to patch the microcode. The BIOS has a file inside
it, with a name like CPUCODE.EXE, that contains eight or more microcode
patches. Each patch has a checksum that the processor can verify, and the
BIOS code can tell when a "patch" "takes". The patch is stored in a small
RAM inside the processor. One of the fields in the patch, contains the
revision number of the patch, which is how people keep track of what
is installed.

When an OS like WinXP or Vista boots, there is a "microcode loader" available
as a driver for the OS. The "microcode loader" can check to see if a patch
has been loaded, and load a later one if available. (How Microsoft gets the
patches to the end user is unclear. Windows Update?)

Now, if we go back to the Intel Processor Identification Utility again,
it has a "revision" field. That field is extracted from the microcode
patch that was loaded into the processor. Either the BIOS loads one
or the OS loads one.

Now, what happens if neither the BIOS nor the OS, are savvy about
your processor ? If that happens, the "Revision" field will be 00.
That is how you'd know the processor was unpatched.

If microcode is not loading, it would be because the processor
is a "stranger". Now, why it is a stranger isn't clear to me.
A "busted" processor seems a little inconvenient, because
the odds of just a few gates being broken in the processor
are pretty slim. Especially when the gates are "visible".
When Intel was testing, the tester machine would have tested
and noticed if the identity info was wrong. (There can be
broken gates which are "hidden" from a test pattern, but in
this case, we know your problem is perfectly visible to
the tester machine. It should have been caught.)

So check the revision field. If it is a value other than "00"
or "0", you're safe and can carry on. If the value is zero,
then it means your processor is unpatched. Whether the machine
could crash because of that, only Intel knows for sure.

This is the info on my Intel machine.

Processor Name: Intel(R) Pentium(R) 4 CPU 2.80C GHz
Type: 0
Family: F
Model: 2
Stepping: 9
Revision: 17 <---- Non zero, so has been patched OK.

Other than that detail, it wouldn't matter.

Could you verify again, what board you're using ? Is it P5N-D ?

Paul

 

[ipleb]

I agree with your basic idea, that this is irrelevant. For the most
part it is. But there is still the issue of the "Revision" listed for
the processor.

Microcode patches are issued by Intel, to correct errors in the processor..
Every processor has a list of around a hundred minor to serious problems.
Usually they don't affect the processor that much, and maybe you'd survive
the boot sequence even if the microcode patch was not installed. When
the processor is released, Intel should have a microcode patch from
day one, for the problems they've discovered since the masks were generated.

There are two opportunities to patch the microcode. The BIOS has a file inside
it, with a name like CPUCODE.EXE, that contains eight or more microcode
patches. Each patch has a checksum that the processor can verify, and the
BIOS code can tell when a "patch" "takes". The patch is stored in a small
RAM inside the processor. One of the fields in the patch, contains the
revision number of the patch, which is how people keep track of what
is installed.

When an OS like WinXP or Vista boots, there is a "microcode loader" available
as a driver for the OS. The "microcode loader" can check to see if a patch
has been loaded, and load a later one if available. (How Microsoft gets the
patches to the end user is unclear. Windows Update?)

Now, if we go back to the Intel Processor Identification Utility again,
it has a "revision" field. That field is extracted from the microcode
patch that was loaded into the processor. Either the BIOS loads one
or the OS loads one.

Now, what happens if neither the BIOS nor the OS, are savvy about
your processor ? If that happens, the "Revision" field will be 00.
That is how you'd know the processor was unpatched.

If microcode is not loading, it would be because the processor
is a "stranger". Now, why it is a stranger isn't clear to me.
A "busted" processor seems a little inconvenient, because
the odds of just a few gates being broken in the processor
are pretty slim. Especially when the gates are "visible".
When Intel was testing, the tester machine would have tested
and noticed if the identity info was wrong. (There can be
broken gates which are "hidden" from a test pattern, but in
this case, we know your problem is perfectly visible to
the tester machine. It should have been caught.)

So check the revision field. If it is a value other than "00"
or "0", you're safe and can carry on. If the value is zero,
then it means your processor is unpatched. Whether the machine
could crash because of that, only Intel knows for sure.

This is the info on my Intel machine.

    Processor Name: Intel(R) Pentium(R) 4 CPU 2.80C GHz
    Type: 0
    Family: F
    Model: 2
    Stepping: 9
    Revision: 17 <---- Non zero, so has been patched OK.

Other than that detail, it wouldn't matter.

Could you verify again, what board you're using ? Is it P5N-D ?

    Paul

Hi Paul

I am using a P5N-E SLI

All of the RAM is DDR2 533 PC2-4300 - Is there is blip when the CPUID
software is telling me that the DRAM frequenc is 266Mhz?

:S

Simon

 

[jaster]

Hi Paul - thanks for your comments - I have also tried a small app
called CPUID - this gave the following data:

it tells me I am running an Intel Xeon 5110 - a socket 771 lga package -
weird as my mobo is obviously a socket 775

The core speed is at 1600 with a multiplyer of x8.0 bus speed at 200mhz
and a rated FSB of 800mhz

I have two ddr2 533 pc2 4300 1gig chips and two ddr2 533 pc2 4300 512mb
chips to give a total of 3gigs
The DRAM freq is at 266/7 and the max bandwidth of each chip is pc2-4300
(266Mhz)

Does that shed any light?

It is definately the right bios update, and I can't remember whether the
problem started post bios update or post ram install

Finally - if the PC boots ok and run ok and all seems good - in real
terms does it actually make a difference that it is showing as an XEON
cpu and not a Core 2Duo ???

Thanks
Simon

I don't think its a problem with the cpu but a problem with the bios.
I'm pretty sure you installed an incorrect bios update. There seems like
a there's a bunch of P5N boards from Asus and IMO you've probably
installed the wrong bios or didn't follow the bios update procedure
properly.

I checked the Asus site and they only list 1 socket 771 board so
something is amiss with the board. BTW I would download a Ultimate
Boot CD so you can run memtest86, cpuid etc without booting Windows.
Windows only reports what the bios says is there.

 

[jaster]

Hi Paul

I am using a P5N-E SLI

All of the RAM is DDR2 533 PC2-4300 - Is there is blip when the CPUID
software is telling me that the DRAM frequenc is 266Mhz?

:S

Simon

Seems like a nice m/b. I still think its a bios update error but if
you're 100% sure you correctly updated the correct bios for your
motherboard then I suggest :

#1 reset the bios to factory default (use RTC jumper) to see if that
fixes
#2 remove the extra memory to see if it fixes
#3 restore the original bios to see if it fixes

I've a few boards fail, a few video cards fail, a few hard drives fail,
some incompatible memory or failures, but I can only recall 1 cpu (a
Cyrus 200mhz) failing back in 1998.

 

[Paul]

I don't think its a problem with the cpu but a problem with the bios.
I'm pretty sure you installed an incorrect bios update. There seems like
a there's a bunch of P5N boards from Asus and IMO you've probably
installed the wrong bios or didn't follow the bios update procedure
properly.

I checked the Asus site and they only list 1 socket 771 board so
something is amiss with the board. BTW I would download a Ultimate
Boot CD so you can run memtest86, cpuid etc without booting Windows.
Windows only reports what the bios says is there.

It probably is not identifying the "LGA771" part with CPUID or equivalent
instruction. It is going by the Xeon label for that. There are some
Xeons that are LGA775 (like an X3220), so they do exist for that motherboard.
I checked the Asus lists, and cannot see any active support for things
like X3220 (like on other motherboards). But overclockers do purchase them
and install them in desktop boards.

Man, this is hilarious. Now that I enter the correct motherboard model of
P5N-E SLI, instead of P5N-D, the E4300 is not listed! It looks like an
oversight, but who knows, it may have some significance.

http://support.asus.com/cpusupport/cpusupport.aspx?SLanguage=en-us&model=P5N-e SLI

Paul

 

[ipleb]

Seems like a nice m/b.   I still think its a bios update error but if
you're 100% sure you correctly updated the correct bios for your
motherboard then I suggest :

#1 reset the bios to factory default (use RTC jumper) to see if that
fixes
#2 remove the extra memory to see if it fixes
#3 restore the original bios to see if it fixes

I've a few boards fail, a few video cards fail, a few hard drives fail,
some incompatible memory or failures,  but I can only recall 1 cpu (a  
Cyrus 200mhz) failing back in 1998.

I used the supplied ASUS biod updater software to update the bios

Im on nights at the mo but will keep looking into it

Simon

 

[Franc Zabkar]

"AP-485 Intel Processor Identification and the CPUID Instruction"
http://www.intel.com/assets/pdf/appnote/241618.pdf

The above URL fails, but this one works ... after apparently
redirecting to the one above:
http://download.intel.com/design/processor/applnots/24161832.pdf

- Franc Zabkar