Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.

Jon

On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:

Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.


Cold fluorinert immersion! Expensive but VERY effective.

Jon Elson wrote:
Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.

That's the best thing about IBM, though. They only make
stuff that works in the extreme stress loads of Supermarkets.

So they're ALWAYS Profit+ competive with 1892,
rather than just the standard Einstone 1492 Jesuit Profits.










Jon

Roy L. Fuchs wrote:

On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:



Cliff wrote:


http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]


Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.




Cold fluorinert immersion! Expensive but VERY effective.


No, this has nothing to do with cooling. It uses certain metals or alloys
to impart mechanical stresses in the silicon that form the conductive
channels, I think, of the transistors. They need compressive stress in
the N channel and tensile stress in the P channel (or maybe I have that
reversed). it gets around the problem of leakage current in the transistors
when the lithographic process is shrunk, with has been the big stumbling
block in continuing the progress in performance of digital chips for the
next generation beyond 90 nm feature size.

Jon

On Wed, 08 Feb 2006 16:54:09 -0600, Jon Elson
Gave us:



Roy L. Fuchs wrote:

On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:



Cliff wrote:


http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]


Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.




Cold fluorinert immersion! Expensive but VERY effective.


No, this has nothing to do with cooling. It uses certain metals or alloys
to impart mechanical stresses in the silicon that form the conductive
channels, I think, of the transistors. They need compressive stress in
the N channel and tensile stress in the P channel (or maybe I have that
reversed). it gets around the problem of leakage current in the transistors
when the lithographic process is shrunk, with has been the big stumbling
block in continuing the progress in performance of digital chips for the
next generation beyond 90 nm feature size.


I thought that feature size limit was just a photo-lithographic
limit.

Have you read anything about Cymer's UV immersion bath phot-lith
process? They can focus even tighter in a thin bath of liquid that
the UV beam shoots through.

On Wed, 08 Feb 2006 16:54:09 -0600, Jon Elson wrote:

No, this has nothing to do with cooling. It uses certain metals or alloys
to impart mechanical stresses in the silicon that form the conductive
channels, I think, of the transistors. They need compressive stress in
the N channel and tensile stress in the P channel (or maybe I have that
reversed).

Do I recall something about quantum effects in very thin
conductors having an effect on resistance?

it gets around the problem of leakage current in the transistors
when the lithographic process is shrunk, with has been the big stumbling
block in continuing the progress in performance of digital chips for the
next generation beyond 90 nm feature size.

That would be trace to trace, right?
--
Cliff

In article ,
says...
On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:

Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.


Cold fluorinert immersion! Expensive but VERY effective.

IBM tried that in the '70s. "Black plague" left by distillation
convinced them to abandon the idea in favor of the He filled
aluminum "Thermal Conduction Module". If you think today's PCs are
power hungry, these things had a power dissipation of up to 2kW.

http://www-03.ibm.com/ibm/history/ex...ntage/vintage_
4506VV2137.html

Expensive indeed!

--
Keith

"Jon Elson" wrote in message ...


Roy L. Fuchs wrote:

On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:


Cliff wrote:

http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]

Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.



Cold fluorinert immersion! Expensive but VERY effective.

No, this has nothing to do with cooling. It uses certain metals or alloys
to impart mechanical stresses in the silicon that form the conductive
channels, I think, of the transistors. They need compressive stress in
the N channel and tensile stress in the P channel (or maybe I have that
reversed). it gets around the problem of leakage current in the transistors
when the lithographic process is shrunk, with has been the big stumbling
block in continuing the progress in performance of digital chips for the
next generation beyond 90 nm feature size.


Hi Jon,

I wonder if the new chip has something to do with this ?

http://www.sciencedaily.com/releases...1002081121.htm

"Adding the diode to silicon circuits will increase circuit performance and
could extend the life of existing production lines--a considerable savings,
as a new fabrication facility can cost over $1 billion."

Light At The End Of The Tunnel
A tunnel diode is a "versatile, high-speed, semiconductor switch," Seabaugh explains.
"You can switch the electrical current back and forth, or you can store information
with it." That's because electrons zipping through a tunnel diode don't follow the
rules of classical physics and remain within designated pathways. Instead, they tunnel
through barrier regions. As more voltage is applied to the diode, he says, more electrons
flow, then less, then more again at a critical voltage level, a consequence of the quirky,
quantum mechanical behavior of tunneling electrons known as "negative differential resistance"
-an up-and-down pattern of current flow.

http://news.com.com/IBM,+Intel+to+ti..._3-828361.html

John

In article ,
says...


Roy L. Fuchs wrote:

On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:



Cliff wrote:


http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]


Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.




Cold fluorinert immersion! Expensive but VERY effective.


No, this has nothing to do with cooling. It uses certain metals or alloys
to impart mechanical stresses in the silicon that form the conductive
channels, I think, of the transistors. They need compressive stress in
the N channel and tensile stress in the P channel (or maybe I have that
reversed). it gets around the problem of leakage current in the transistors
when the lithographic process is shrunk, with has been the big stumbling
block in continuing the progress in performance of digital chips for the
next generation beyond 90 nm feature size.

Not so much leakage as higher carrier mobility for both the 'P' and
'N' devices and higher drive strength (which may improve leakage).

http://www.electrochem.org/publicati...ring2005/IF03-
05-Pg36-37.pdf

--
Keith

On Thu, 9 Feb 2006 09:59:05 -0500, Keith Williams
Gave us:

In article ,
says...
On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:

Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.


Cold fluorinert immersion! Expensive but VERY effective.

IBM tried that in the '70s. "Black plague" left by distillation
convinced them to abandon the idea in favor of the He filled
aluminum "Thermal Conduction Module". If you think today's PCs are
power hungry, these things had a power dissipation of up to 2kW.

http://www-03.ibm.com/ibm/history/ex...ntage/vintage_
4506VV2137.html

Expensive indeed!

As far as I know, fluorinert is NOT reactive with any metals at all.

It never affected any of our circuits, and we used it all the time
up to 50kV.

In article ,
says...
On Thu, 9 Feb 2006 09:59:05 -0500, Keith Williams
Gave us:

In article ,
says...
On Wed, 08 Feb 2006 12:33:34 -0600, Jon Elson
Gave us:

Cliff wrote:
http://www.theregister.co.uk/2006/02...m_power6_show/

[
The Power6 chip will run between 4GHz and 5GHz, and has been shown to hum away
at 6GHz in the lab. IBM reckons that some process technology breakthroughs have
allowed it to kick the frequency higher while still keeping heat and power
consumption issues under control. All told, IBM claims that Power6 will be twice
as fast as competing server processors from Intel, AMD and Sun Microsystems.
]

[
IBM claims to have made major performance gains by stretching and squeezing
silicon and using insulation techniques.
]
Oh, stressed silicon! Yes, this stuff really works. Unless
IBM is WAY ahead of the competition, it will be quite a few
years before you can buy this at the local computer store.


Cold fluorinert immersion! Expensive but VERY effective.

IBM tried that in the '70s. "Black plague" left by distillation
convinced them to abandon the idea in favor of the He filled
aluminum "Thermal Conduction Module". If you think today's PCs are
power hungry, these things had a power dissipation of up to 2kW.

http://www-03.ibm.com/ibm/history/ex...ntage/vintage_
4506VV2137.html

Expensive indeed!

As far as I know, fluorinert is NOT reactive with any metals at all.

Not reaction. Residue from distillation. The gunk was an impact
on circuit reliability. Multi-MegaBux were spent trying to fix
this and the TCM was the outcome.

It never affected any of our circuits, and we used it all the time
up to 50kV.

Perhaps your folks weren't as good as IBM's.

--
Keith