Can Intel pump up Tualatin volume?
On the horns of a fab dilemma
Intel Australia's representatives are obviously a little less tight lipped than their European counterparts, judging from the fact that one of them let Insane Hardware take a snap of a sample of one of the .13 micron chips the other day.
Our representatives here in Europe won't even comment about Intel codenames, never mind opening their kimonos to reveal their chip knickers.
While it's evident from the photograph that early samples are beginning to roll, there are, however, serious questions about how many of these babies Intel factories can produce by the end of the year.
Tualatin processors use a .13 micron process, and this, in turn, depends on Intel either converting its existing factories or building new factories to churn out the chips.
This time last year, Intel faced a similar problem as it struggled to shift capacity from its older .25 micron process to .18 micron, Coppermine technology. It was struggling to do so, and this explains why Intel found itself with serious shortages for much of the first six months of 2000.
Switching from one process technology to another demands capital investment in expensive equipment, as well as extensive re-jigging of existing plants to take advantage of the new technology.
No wonder executives at semiconductor companies are tough. We can only begin to imagine the planning and execution involved in such a switch. The chemicals are dangerous, the process is complex, and then there's selling the pesky things once you've made them.
Once more, Intel will have to perform one of its complex juggling acts, as it pushes the market towards the Pentium 4, slashes prices on its existing Pentium III microprocessors, and switches production cycles - depending how the market goes, of course.
Tualatin is the rabbit conjurors at Intel need to pull out of the hat around the middle of the year, but whether it can do so in sufficient quantities by then is hard, right now, to judge. ®