Can i upgrade my processor?

[elton77]

On one of my PC's I have a Intel Pentium 4, Can I go to a duel core o
would I have to buy a new mother board

 

[Mark]

On one of my PC's I have a Intel Pentium 4, Can I go to a duel core or
would I have to buy a new mother board?

Yes.

 

[johannes]

Yes.

Yes and No.

 

[Bruce]

Yes and No.

Depends

(won't always work as P4s came in a few different socket types, and some
motherboard of the correct socket type (775) still will not support the
duel core)

 

[Jesco Lincke]

On one of my PC's I have a Intel Pentium 4, Can I go to a duel core or
would I have to buy a new mother board?

You'll have to supply a bit more information before expecting a truly
meaningful answer.
What mainboard are you using? Does the manufacturer have a CPU
compatibility list on his homepage?
Since you're upgrading from a P4-system a mainboard upgrade sounds quite
likely if you want to change to a dual core.
Bear in mind that changing CPU and mainboard quite often results in a
change of RAM and possibly also other parts.

 

[johannes]

Depends

(won't always work as P4s came in a few different socket types, and some
motherboard of the correct socket type (775) still will not support the
duel core)

I am actually surprised to see that the mobile Core 2 Duo processors still
uses socket 479 like the Pentium M. So what is the reason for using as many
as 775 pins, balls or whatever you call them for the other Intel processors?

 

[Bruce]

I am actually surprised to see that the mobile Core 2 Duo processors still
uses socket 479 like the Pentium M. So what is the reason for using as many
as 775 pins, balls or whatever you call them for the other Intel processors?

http://en.wikipedia.org/wiki/Socket_775
"Intel changed from Socket 478 to LGA775 because the new pin type offers
better power distribution to the processor, allowing the front side bus
to be raised to 1066 MT/s."

 

[johannes]

http://en.wikipedia.org/wiki/Socket_775
"Intel changed from Socket 478 to LGA775 because the new pin type offers
better power distribution to the processor, allowing the front side bus
to be raised to 1066 MT/s."

But doesn't quite explain the higher number of pins.

 

[Bruce]

But doesn't quite explain the higher number of pins.

Agreed, but I interpreted it to mean that more pins allowed better power
distribution and less load on each thus allowing higher clocks.

 

[Robert Redelmeier]

But doesn't quite explain the higher number of pins.

It does, but you need to either trust the explantion given
or seek out more specialized knowledge.

Consider: a modern CPU drawing 60-100W power at 1.5-2V needs
to move 40-70 Amperes in from power and out to ground with
very little budget for resistance.

In the relatively tolerant electrical power world, this is
moved by cm2 of contact area and #6 or #8 wire.

In CPUs it has to be pushed through teeny-tiny pins (gold plated
out of necessity) that don't look like they could even handle
the 200 mA put through each. On carrier it just gets worse.
5-10 micron wires moving those mA to chip pads. On-chip it gets
even worse, as fractional micron metalic layers have to move mA.
The further they go and higher the load, the worse the voltage
loss and signalling margin.

Pins cost serious money. CPU mfrs don't add them for fun,
but out of necessity and compromise. I've sometimes wondered
whether CPUs should have some alternate power/ground connection
other than std signalling pins. Maybe long bars around the
outside of the carrier bonded to pwr/gnd layers.


-- Robert