Dual-core Intels by 2005

http://story.news.yahoo.com/news?tmp...el_otellini_dc

http://tinyurl.com/nmym

At the Intel Developer Forum, Intel touts that it will bring out dual-core
processors by 2005 or 2006. Tulsa will be a successor to 32-bit Xeon,
expected around 2-3 years time. Tanglewood will be a successor to 64-bit
Itanium.

Another interesting quote from Intel's Otellini when asked about when Intel
will bring out a desktop 64-bit processor:

"The production operating systems are not there yet" for 64-bit desktop
computers, he said. "The mainstream applications won't exist until next
year."

Which would tend to imply that Intel is waiting for AMD's processors to
attain critical mass before they introduce their own version!

Yousuf Khan

On Wed, 17 Sep 2003 04:29:19 GMT, "Yousuf Khan"
wrote:
At the Intel Developer Forum, Intel touts that it will bring out dual-core
processors by 2005 or 2006. Tulsa will be a successor to 32-bit Xeon,
expected around 2-3 years time. Tanglewood will be a successor to 64-bit
Itanium.

A friend of mine made some spoof about Intel will also be going into
the liquid nitrogen cooling business with this... given the max spec
for the current P4 is already into the 100W range :PpPp
--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

In article , a?n?g?e?
says...
On Wed, 17 Sep 2003 04:29:19 GMT, "Yousuf Khan"
wrote:
At the Intel Developer Forum, Intel touts that it will bring out dual-core
processors by 2005 or 2006. Tulsa will be a successor to 32-bit Xeon,
expected around 2-3 years time. Tanglewood will be a successor to 64-bit
Itanium.

A friend of mine made some spoof about Intel will also be going into
the liquid nitrogen cooling business with this... given the max spec
for the current P4 is already into the 100W range :PpPp

Don't laugh too much. Intel has been seriously talking about
liquid cooling for a couple of years. I see it coming.

--
Keith

Keith R. Williams wrote:
In article , a?n?g?e?
says...

On Wed, 17 Sep 2003 04:29:19 GMT, "Yousuf Khan"
wrote:

At the Intel Developer Forum, Intel touts that it will bring out dual-core
processors by 2005 or 2006. Tulsa will be a successor to 32-bit Xeon,
expected around 2-3 years time. Tanglewood will be a successor to 64-bit
Itanium.

A friend of mine made some spoof about Intel will also be going into
the liquid nitrogen cooling business with this... given the max spec
for the current P4 is already into the 100W range :PpPp


Don't laugh too much. Intel has been seriously talking about
liquid cooling for a couple of years. I see it coming.

They even did it for most of the P4 IDF demos the last few years.
Remember the 3GHz and 4GHz P4 demos two years back? And we are still at
3GHz.

BTW, if the 104W Prescott numbers at 3GHz that were going around the net
are true, I can't wait to see how they intend to reach 4 or 5GHz. At
5GHz that would be...173W? Assuming they don't have to increase the
voltage level, which they would.

Alex
--
My words are my own. They represent no other; they belong to no other.
Don't read anything into them or you may be required to compensate me
for violation of copyright. (I do not speak for my employer.)

On Thu, 18 Sep 2003 08:38:15 -0400, Alex Johnson
wrote:
BTW, if the 104W Prescott numbers at 3GHz that were going around the net
are true, I can't wait to see how they intend to reach 4 or 5GHz. At
5GHz that would be...173W? Assuming they don't have to increase the
voltage level, which they would.

The current top end P4 will already be in 100W range if you calculate
their actual max from the TDP figure. So not surprising if Prescott
hits that. I suppose they could go for a die shrink when they hit
130W?

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

Couldn't find much at Intel on liquid cooling, I two other methods first:
o Heatpipes
---- cheap bolt-on
o Ducting with larger case mounted fans (Dual G5 shows it best)
---- moves the case into the upgrade cycle once again

Ducting is the best solution - 25cfm will cool 125W of CPU, and
36cfm will cool 180W of CPU - unremarkable for DC Tube-Axial fans.

Liquid cooling I think unlikely - except for "what can it do".
Most of the world has been ripping out chilled-water systems, and
are not likely to want to put them back into racks unless there is a
truly huge benefit in terms of complete rethink of PC architecture.

That said, heatpipes are a self-contained liquid cooling system,
which often uses water in a copper tube - which I find interesting.
IBM gave up on using a heatpipe style cooling on their processor
modules on an early mainframe - due to deposition of impurities on
the heat-inputting surface. It was not possible to get the interior of
the device clean enough to prevent progressive impurities creating
a thermal barrier degrading the device's thermal capability over time.

Often wondered why Heatpipes don't suffer the same problem,
perhaps they are clean enough or use something to prevent them
binding to the thermal surface keeping them enough in suspension.
--
Dorothy Bradbury
http://homepage.ntlworld.com/dorothy...ry/panaflo.htm (Direct)

The little lost angel wrote:
On Thu, 18 Sep 2003 08:38:15 -0400, Alex Johnson
wrote:

BTW, if the 104W Prescott numbers at 3GHz that were going around the net
are true, I can't wait to see how they intend to reach 4 or 5GHz. At
5GHz that would be...173W? Assuming they don't have to increase the
voltage level, which they would.


The current top end P4 will already be in 100W range if you calculate
their actual max from the TDP figure. So not surprising if Prescott
hits that. I suppose they could go for a die shrink when they hit
130W?

Not likely. Prescott will be the first 90nm processor. It will be
about two years before the next process is ready and a fab is equipped.
Prescott will exceed 130W long before that...almost immediately unless
there is some serious power leak that can be plugged, cutting power
usage down dramatically.

Alex
--
My words are my own. They represent no other; they belong to no other.
Don't read anything into them or you may be required to compensate me
for violation of copyright. (I do not speak for my employer.)

"dorothy.bradbury" wrote in message
...
That said, heatpipes are a self-contained liquid cooling system,
which often uses water in a copper tube - which I find interesting.
IBM gave up on using a heatpipe style cooling on their processor
modules on an early mainframe - due to deposition of impurities on
the heat-inputting surface. It was not possible to get the interior of
the device clean enough to prevent progressive impurities creating
a thermal barrier degrading the device's thermal capability over time.

Sounds like the same problems inherent in automotive radiators. The solution
there (no pun intended) is to mix the water with anti-freeze which also
breaks up the impurities in the water and prevents it from depositing. Maybe
we'll be putting anti-freeze into our computers soon too? :-)

Yousuf Khan

"The little lost angel" wrote in
message ...
On Wed, 17 Sep 2003 04:29:19 GMT, "Yousuf Khan"
wrote:
At the Intel Developer Forum, Intel touts that it will bring out
dual-core
processors by 2005 or 2006. Tulsa will be a successor to 32-bit Xeon,
expected around 2-3 years time. Tanglewood will be a successor to 64-bit
Itanium.

A friend of mine made some spoof about Intel will also be going into
the liquid nitrogen cooling business with this... given the max spec
for the current P4 is already into the 100W range :PpPp

You know, the fact that their announcing stuff that they expect for 2005 or
2006, now, would mean they'll have it in time to meet their competitors own
dual-cores.

Yousuf Khan

"Alex Johnson" wrote in message
...
The little lost angel wrote:
The current top end P4 will already be in 100W range if you calculate
their actual max from the TDP figure. So not surprising if Prescott
hits that. I suppose they could go for a die shrink when they hit
130W?

Not likely. Prescott will be the first 90nm processor. It will be
about two years before the next process is ready and a fab is equipped.
Prescott will exceed 130W long before that...almost immediately unless
there is some serious power leak that can be plugged, cutting power
usage down dramatically.

Doesn't look like the die-shrinks are reducing power requirements all that
much any more. At best, they are keeping them at the same level as before.

Yousuf Khan