Editor's Page

Talking it up at Semicon

What's the favorite pastime at Semicon West? Talking, of course! Sales pitches, technobabble, egotistical opinions, or just plain gossip provide the quantum energy source on the trade-show floor and in the bars and restaurants after hours. In the interests of fostering good chatter, here are some suggested Semicon conversation topics.

Have you noticed that the terms "front end" and "back end" have ceased to have any specific meaning and that their definitions vary depending on the context in which they are being used? People often refer to the front and back ends of a particular process cluster. The most common misuses of front end and back end are the now-ubiquitous terms, front end of line (FEOL) and back end of line (BEOL), which are really the front and back of the "traditional" front end, also known as the wafer processing sequence. Then there's a relatively new term muddying up the waters —postfab — a murky patch at the very end of metallization yet before the wafers go on to the assembly and packaging phases. Anyone who throws out these terms better be prepared to define them before becoming bogged down in a morass of misunderstanding.

Ask yourself where you've heard this before: "This technology will change the face of the industry and allow for greater profitability, so you better get your act together or be left at the starting gate." Am I the only one who thinks that the hype about copper-interconnect sounds much like that surrounding the dawn of the 300-mm era? Will equipment and materials suppliers again be left holding the bag, having to develop prototypes and putting further pressure on already-strapped R&D budgets, while their chipmaker customers work out their dual-damascene process kinks? Once the lines are in place, how fast do you think the yields can be ramped up when a whole new category of defects spams the works?

No big show would be complete without the latest from the mergers-and-acquisition rumor mill. Expect the usual comments about Applied Materials gobbling up someone from the photolithography segment. But then, acquisition isn't the only way to get into that business. When is the equipment industry's big dog going to start barking about its mysterious PRP division?

Want to hear the basis of a very interesting story that will likely never be told? Buttonhole an ex-Digital Semiconductor employee or Intel minion involved in the transition of what was Digital's Fab 6 to what is now Intel's Fab 17. Some Digital folks expressed concerns that the new owners would impose the legendary "copy exactly" approach on their pride and joy. It turned out that the "best-known method" alternative won out, and the Alpha fab has been left much as it was. The transfer of a state-of-the-art semiconductor facility from one industry leader to another is almost without precedent, but don't expect to read many details in the trade press, since the new parent company sees no value (for itself) in recounting the story.

One subject that should garner increasing airtime as we stumble toward the new century is the Y2K problem, commonly called the Millennium bug. Check out our Viewpoint piece written by Rick Scott of SEMI/Sematech in next month's show issue of MICRO for talking points on this looming problem.

As for this pesky downturn, here's a good one heard recently from the vp of a major materials manufacturer: "How long do we have to hold our breath before we can start swimming again?" We'd better take a deep breath, because it looks like we all have to turn a deeper shade of blue before we can come up for air.

Tom Cheyney
Editor

tom.cheyney@cancom.com

MicroHome | Search | Current Issue | MicroArchives
Buyers Guide | Media Kit