The New P3 Tualatin Processor

Tualatin, as might be expected, is yet another river in Oregon. It is also Intel's internal code name for its P3 die shrink to 0.13 micron. Though bigger caches are a possibility, the primary benefits of this die shrink will be faster clock speeds, lower cost and the option to move to a faster front side bus (200MHz).

Tualatin will be ready for production sometime in Q3'01 or perhaps late Q2'01. The Almador chip set may be ready to ship before Tualatin, but OEMs will be able to verify Tualatin compatibility prior to shipping Almador platforms in volume. The Almador platform will also ship with Coppermine processors.

Already, the industry seems to agree that Intel has fallen behind AMD in terms of its ability to ship high-speed P3 processors. From Q3'00 until Q3'01, this situation is expected to worsen. Willamette at 1.5GHz will allow Intel to claim the highest clock speeds, but in terms of computational throughput, balanced against system cost, Intel will still have a huge 12-month long 'dead-zone' in its roadmap.

Intel's 'dead-zone' is clearly indicated by the vacancy in the 'Mainstream 3' category of the roadmap from Q3'00 to Q3'01. This is where AMD will really be able to threaten Intel. Into this cost/performance segment, AMD will ship its 1GHz - 1.2GHz single processor Tbird-Athlon systems with low cost PC266 DDR SDRAM. AMD will extend its reach into the performance segment with a dual processor version of the same DDR platform.

Intel's only dual processor solution will be the 840/Coppermine, slightly disadvantaged in clock speed and burdened with RDRAM cost/availability problems. Willamette is uniprocessor only, and carries the same RDRAM cost burden. Finally, the 815 is a decidedly midrange product. Intel seems to have all the corners covered, but has a big hole in the space between these platforms.

Intel's difficulties in this vital product range will be eased somewhat when Tualatin and a fast DDR platform is ready for production. Beginning in late Q2'01 Tualatin can fill in from below, followed by Northwood filling in from above in Q3'01. This will help to close Intel's very messy cost/performance gap, but AMD will still have a significant opportunity to entrench itself in this market space. Still, Intel must figure out how to respond to the Athlon dual processor threat in the high end PC and workstation space.

This 'dead-zone' will be the acid test for OEMs such as Dell, which has maintained an Intel-only product mix. If Dell persists in its stance during this period, it could stand to lose market share in this most lucrative segment - one where Dell is currently seen as a leader.

Prev | Next