[P2B] dual celeron tualatin -- is it possible?

hello,

looking for the answer to that question I googled a little and saw some
posts/pages which stated that "intel does not support dual celeron
tualatin configuration" but they didn't state this configuration was
impossible. however some other pages state that it is not possible. what
is the truth? is is possible to run dual celeron tualatin on P2B-D
motherboard as it is possible to run dual PIII tualatin with it?

cheers,
geos

geos wrote:
hello,

looking for the answer to that question I googled a little and saw some
posts/pages which stated that "intel does not support dual celeron
tualatin configuration" but they didn't state this configuration was
impossible. however some other pages state that it is not possible. what
is the truth? is is possible to run dual celeron tualatin on P2B-D
motherboard as it is possible to run dual PIII tualatin with it?

Impossible. intel does not connect the pins necessary for dual
processing to the cpu die.
Dual celerons only worked with mendocinos, neither coppermine nor
tualatins will work.
Oh and PIII tualatins are strictly uniprocessor too, you'd need PIII-S
(the 512KB cache version) for dual.

Roland

"Leythos" wrote in message
...
In article , says...
hello,

looking for the answer to that question I googled a little and saw some
posts/pages which stated that "intel does not support dual celeron
tualatin configuration" but they didn't state this configuration was
impossible. however some other pages state that it is not possible. what
is the truth? is is possible to run dual celeron tualatin on P2B-D
motherboard as it is possible to run dual PIII tualatin with it?

I have Dual P3/1Ghz running in a Compaq Server and it works fine. I also
have Dual Celeron 500Mhz running on a BP6. It was my understanding that
after the P3 is when things changed on Duals. Before the P4 if your
board had Dual sockets, as long as it supported the CPU that you could
run Duals - you might have to make sure they are the same STEP.


Same stepping works better, but not always necessary. I bought a dual and
found out later:

System Information report written at: 10/29/2004 08:33:37 AM
[System Summary]

Item Value
OS Name Microsoft Windows 2000 Server
Version 5.0.2195 Service Pack 4 Build 2195
OS Manufacturer Microsoft Corporation
System Name DARKSTAR
System Manufacturer GBT___
System Model AWRDACPI
System Type X86-based PC
Processor x86 Family 6 Model 7 Stepping 2 GenuineIntel ~498 Mhz
Processor x86 Family 6 Model 7 Stepping 3 GenuineIntel ~498 Mhz
BIOS Version Award Modular BIOS v4.51PG
Windows Directory C:\WINNT
System Directory C:\WINNT\system32
Boot Device \Device\Harddisk0\Partition1

I get this error, but all seems to work otherwise:

The CPUs in this multiprocessor system are not all the same revision level.
To use all processors the operating system restricts itself to the features
of the least capable processor in the system. Should problems occur with
this system, contact the CPU manufacturer to see if this mix of processors
is supported.

Roland Scheidegger wrote:
geos wrote:

hello,

looking for the answer to that question I googled a little and saw
some posts/pages which stated that "intel does not support dual
celeron tualatin configuration" but they didn't state this
configuration was impossible. however some other pages state that it
is not possible. what is the truth? is is possible to run dual celeron
tualatin on P2B-D motherboard as it is possible to run dual PIII
tualatin with it?


Impossible. intel does not connect the pins necessary for dual
processing to the cpu die.

I've heard that, and even repeated it myself - but do we *know* that's
the case?

There are visible traces leading from both BR1# pins (x2 and n33)
towards the core on a Tualatin Celeron...

Dual celerons only worked with mendocinos, neither coppermine nor
tualatins will work.
Oh and PIII tualatins are strictly uniprocessor too, you'd need PIII-S
(the 512KB cache version) for dual.

Roland

P2B wrote:

Impossible. intel does not connect the pins necessary for dual
processing to the cpu die.


I've heard that, and even repeated it myself - but do we *know* that's
the case?

There are visible traces leading from both BR1# pins (x2 and n33)
towards the core on a Tualatin Celeron...

It is possible intel disables this functionality the same way as they
lock the multiplier - by burning on-die fuses.
Nobody *knows* this (I'm sure some people at intel know it :-)) - but
experiments showed the cpus simply do not work in a dual configuration
(I'm pretty sure some people tried that, didn't you do that yourself?
I've only got a lone single-cpu P2B...). And this sounds like the most
probable reason.
(though I'm not sure for tualatins, afaik coppermine celerons used the
same die as coppermine PIII - but the tualatin PIII-S uses a different
die (?) than tualatin PIII/celeron, so there would be certainly other
ways do disable multiprocessing capabilities.)

Roland

thank you guys for all the answers

cheers,
geos