At the opening keynote at Intel's autumn 2005 developer forum in San Francisco on Tuesday, chief executive Paul Otellini announced a new chip architecture and a new class of computer.
The next generation power-optimised micro-architecture, whose name was not revealed, combines elements of the Netburst design behind the Pentium 4 with aspects of the low-power Pentium M at the heart of the Centrino platform. It will be the basis for three new 65nm dual-core products to be launched in the second half of 2006 — Woodcrest for servers, Conroe for desktops and Merom for mobile use. All three products will share common features such as 64-bit compatibility, virtualisation, trusted platform support and management features.
Otellini showed off systems based on prototypes of the new chips, running Windows and Linux.
Otellini also said that by the end of the decade Intel would have another new architecture running x86 code at ten times less power than the lowest-power version currently available — around 0.5W. This could either be used for portable systems running at very low power, or standard systems running at 10 times the performance-per-watt of current designs, Otellini suggested.
Otellini also showed off prototypes of a new style of portable computer, the handtop. These combined the performance of PCs with the portability of handsets, he said, with all-day batteries and always-connected wireless. The devices could be configured in PDA, Blackberry or laptop mode, and would run fully featured operating systems.
Talkback
Intel chips using low power? They process at a snails pace with high power. Considering their "unofficial" partners at Microsoft keep building larger OS software which requires more power, is this a sign that these lovers are falling out? I hope they screw it up & AMD take the larger share of the processor market.
Speed and power create heat. This is based on the laws of physics, and even Intel can not circumvent these laws. Sure there are small changes that can be made to improve efficiency, like reducing the voltage ( Now at a little more than 1 volt in current Intel processors ), that being said how close to zero voltage can you go: not much closer. I do not have any confidence that Intel has broken any barriers in physics. If they had it would be all over scientific journals. Low power = low performance, simple rules of the Physical universe.