Paul, I do not know who is the smart money pouring $120 MM into this pipe dream, but the last sentence of the article indicates they know little about the problems facing them, the current cost of Si wafers is less than 5% of the chip cost, big saving using less of it.
This remind me of the $200 MM that went into "wafer scale integration" in Trilogy, another pipe dream beaten down by the realities of heat dissipation. Air insulated balls will have to depend for its heat dissipation on radiation rather than conductivity. That is not an easy problem to solve. A 1 mm ball will have how many overlaying layers?
Last, for that approach why even use silicon as a substrate, unless of course it is not a ball but a microcube, in that even good luck on turning sharp corners (field amplification). Someone, please put me in contact with the financial backers of this outfit, I have a real nice bridge...
Zeev |
