jmke

Intel will reportedly announce that it will discontinue the development process of the next-generation NetBurst microprocessors code-named Tejas and Jayhawk. The chips are apparently too hot to go the market.

According to reports over Reuters, The Wall Street Journal and some other sources, Intel will announce on Friday its future plans that do not include Tejas and Jayhawk processors, but that are formed around the architecture used in the Pentium M chips.

The move is believed to represent a significant shift in the development plans of the world’s largest chipmaker and stems from its desire to build chips that are powerful without generating excessive amounts of heat, Reuters reported.

Originally planned for the launch in the second half of 2004, Xeon processors “Jayhawk” – successors of Intel Xeon “Nocona” CPUs, were then said to release in the second quarter of 2005. The Jayhawk core was expected to boast the same micro-architecture as a desktop chip code-named Tejas that would continue quantitative and qualitative boosts of NetBurst specifications. The chip were projected to have 24KB L1 cache, 16K uOps Trace Cache, 1MB L2 cache, a more efficient branch prediction mechanism, a new set of instructions known as “Tejas New Instructions” as well as improved Hyper-Threading organization and so on.

The key performance driver of all NetBurst microprocessors is their clock-speed. However, extremely high frequencies also drive power consumption and heat generation dramatically upwards. Additionally, the Santa Clara, California-based chipmaker ran into problems with its 90nm fabrication process that caused chips to dissipate even more heat because of the power leakage process.

In contrast, Intel Pentium M processors designed for notebooks consume considerably less energy compared to the Pentium 4, but performs nearly as fast. There are indications of dual-core Pentium M-like chips as well as 64-bit capable Pentium M chips to enter the market next year or in 2006.

http://www.xbitlabs.com/news/cpu/dis...507000306.html

__________________

The Senile Doctor

http://www.reuters.com/newsArticle.j...ryID=5069 106

canned...
unbelievable...

xtreme wattage must be the problem...
I think speedincreases will slow down extremely over the next few years...
Which could be a good thing too... my socket 939 amd at 3.2 ghz under a mach2gt could well be the fastest processor for 18-24 months...

__________________
OC-2-the-death
Where the Reverend is doing his Magick, all mortals be silent
Doom over the world

jmke

keep on dreaming. Dual Core CPU's in 2006 will blow that overclocked S939 away


edit, Dual Operton system.. possible
http://www.theregister.co.uk/2004/05...on_price_cuts/

__________________

The Senile Doctor

thank you for confirming my point, JMke...
beginning 2006 is 18 months away, end 2006 even 29...
so, as I said, in two months a socket 939 system heavily overclocked could well be the thing to keep us highspeed until 18-24(or should I say 29) months from now.


dual opteron systems... for gaming too? mt main rig is exclusively a gaming rig

__________________
OC-2-the-death
Where the Reverend is doing his Magick, all mortals be silent
Doom over the world

jmke

If a Dual Core in 2006 will, I quote "blow away" your S939, then you can be sure that in less then 18-24 months there will be faster setups


Dual opteron for servers and workstation

__________________

The Senile Doctor

hmm.

__________________
OC-2-the-death
Where the Reverend is doing his Magick, all mortals be silent
Doom over the world

jmke

indeed my friend
if you buy your S939 setup in June, you probably end with an AGP DDR1 AMD setup

at the end of the year however you might have DDR-II and PCI-Express motherboards available, it would be very odd if these show no improvement of performance over the current DDRI/AGP design structure.

__________________