Opteron versus Xeon

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Opteron versus Xeon

Post by bob » Thu Jul 19, 2007 9:24 pm

Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
CRoberson wrote:It would not be a nanoprocessor. A 16 nanometer "process"
(not processor) doesn't mean the cpu is 16 nanometers in size.

Think about it; you can easily see the cpu in your computer without
a microscope.
You didn't read what I wrote. I said a true nanoprocessor would use a 1 nanometer process, not 16.

Try and view the circuits without a microscope, you can't! We're not talking about the size of the die, or the overall size of the chip, but rather the size of a single logic gate.

By your reasoning microprocessors are not microprocessors as you can view the chip!

Think About It!
None of this matters as you can't make a gate with 2 atoms, so it's a moot point...

feature-size is not going _that_ small.
AFAIK it's already been done with experimental logic gates, some the size of a few molecules, and some the size of a single atom.

Do you think microchip technology is the end of computing? We haven't yet begun!
I've not seen _anything_ about a gate the size of an atom. The problem is that to build a transistor, you need two junctions between dissimilar materials. NPN or PNP for example. That has to be at least 3 atoms and nobody has put that together that I am aware of. Physics suggests it is not possible in fact.

Terry McCracken

Re: Opteron versus Xeon

Post by Terry McCracken » Thu Jul 19, 2007 9:29 pm

bob wrote:
Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
CRoberson wrote:It would not be a nanoprocessor. A 16 nanometer "process"
(not processor) doesn't mean the cpu is 16 nanometers in size.

Think about it; you can easily see the cpu in your computer without
a microscope.
You didn't read what I wrote. I said a true nanoprocessor would use a 1 nanometer process, not 16.

Try and view the circuits without a microscope, you can't! We're not talking about the size of the die, or the overall size of the chip, but rather the size of a single logic gate.

By your reasoning microprocessors are not microprocessors as you can view the chip!

Think About It!
None of this matters as you can't make a gate with 2 atoms, so it's a moot point...

feature-size is not going _that_ small.
AFAIK it's already been done with experimental logic gates, some the size of a few molecules, and some the size of a single atom.

Do you think microchip technology is the end of computing? We haven't yet begun!
I've not seen _anything_ about a gate the size of an atom. The problem is that to build a transistor, you need two junctions between dissimilar materials. NPN or PNP for example. That has to be at least 3 atoms and nobody has put that together that I am aware of. Physics suggests it is not possible in fact.
There are other methods, I'm not talking about breaking the laws of physics, but that might be possible too, but that's another matter. :wink:

As for your 3 atoms, I believe experimentally it has been done. Although it's not yet practicle for everyday use.

Wait another 5-10 years. I bet you'll be surprised :o

Best,
Terry

bob
Posts: 20923
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: Opteron versus Xeon

Post by bob » Fri Jul 20, 2007 2:02 am

Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
CRoberson wrote:It would not be a nanoprocessor. A 16 nanometer "process"
(not processor) doesn't mean the cpu is 16 nanometers in size.

Think about it; you can easily see the cpu in your computer without
a microscope.
You didn't read what I wrote. I said a true nanoprocessor would use a 1 nanometer process, not 16.

Try and view the circuits without a microscope, you can't! We're not talking about the size of the die, or the overall size of the chip, but rather the size of a single logic gate.

By your reasoning microprocessors are not microprocessors as you can view the chip!

Think About It!
None of this matters as you can't make a gate with 2 atoms, so it's a moot point...

feature-size is not going _that_ small.
AFAIK it's already been done with experimental logic gates, some the size of a few molecules, and some the size of a single atom.

Do you think microchip technology is the end of computing? We haven't yet begun!
I've not seen _anything_ about a gate the size of an atom. The problem is that to build a transistor, you need two junctions between dissimilar materials. NPN or PNP for example. That has to be at least 3 atoms and nobody has put that together that I am aware of. Physics suggests it is not possible in fact.
There are other methods, I'm not talking about breaking the laws of physics, but that might be possible too, but that's another matter. :wink:

As for your 3 atoms, I believe experimentally it has been done. Although it's not yet practicle for everyday use.

Wait another 5-10 years. I bet you'll be surprised :o

Best,
Terry
the problem is that those three atoms are wider than a nanometer. :)

IBM has dragged individual atoms around with a variant of a tunneling electron microscope. But that was done atom by atom. Which isn't going to cut it when making a chip unless you want to make one per 100 years. :)

Terry McCracken

Re: Opteron versus Xeon

Post by Terry McCracken » Fri Jul 20, 2007 2:36 am

bob wrote:
Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
bob wrote:
Terry McCracken wrote:
CRoberson wrote:It would not be a nanoprocessor. A 16 nanometer "process"
(not processor) doesn't mean the cpu is 16 nanometers in size.

Think about it; you can easily see the cpu in your computer without
a microscope.
You didn't read what I wrote. I said a true nanoprocessor would use a 1 nanometer process, not 16.

Try and view the circuits without a microscope, you can't! We're not talking about the size of the die, or the overall size of the chip, but rather the size of a single logic gate.

By your reasoning microprocessors are not microprocessors as you can view the chip!

Think About It!
None of this matters as you can't make a gate with 2 atoms, so it's a moot point...

feature-size is not going _that_ small.
AFAIK it's already been done with experimental logic gates, some the size of a few molecules, and some the size of a single atom.

Do you think microchip technology is the end of computing? We haven't yet begun!
I've not seen _anything_ about a gate the size of an atom. The problem is that to build a transistor, you need two junctions between dissimilar materials. NPN or PNP for example. That has to be at least 3 atoms and nobody has put that together that I am aware of. Physics suggests it is not possible in fact.
There are other methods, I'm not talking about breaking the laws of physics, but that might be possible too, but that's another matter. :wink:

As for your 3 atoms, I believe experimentally it has been done. Although it's not yet practicle for everyday use.

Wait another 5-10 years. I bet you'll be surprised :o

Best,
Terry
the problem is that those three atoms are wider than a nanometer. :)

IBM has dragged individual atoms around with a variant of a tunneling electron microscope. But that was done atom by atom. Which isn't going to cut it when making a chip unless you want to make one per 100 years. :)
Yes it is bigger than a nanometer, It would be ~1.5 nm.

Maybe chips aren't the key? :wink:

A new technology, perhaps? Maybe we can shrink atoms? Maybe something else, or reduce dimensions? Or something mind blowing that hasn't even been thought of, let alone considered? Transistors are getting a little old :wink:

Looking for Ideas Regards,
Terry

Post Reply