Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : George Gilder - Forbes ASAP -- Ignore unavailable to you. Want to Upgrade?

To: ekn who wrote (2150)	10/6/1999 4:48:00 PM
From: SteveG	Read Replies (1) \| Respond to of 5853

Irvine Sensors has been working on stacking for years as part of a DARPA process through MD and Boeing (which have faltered in their support recently) But others have also tried things like this for awhile (like 5+ years), with none yet attaining commercial success. According to some smart silicon guys, the basic concept of laying out in 3 dimensions is incredibly appealing - by just adding a dimension you can increase the amount of circuitry you can put in a space dramatically. And (from the IC level) there are apparently some unintended but beneficial consequences which make stacking even MORE appealing - because the design of the system on a chip is divided into several sub-chips which are then stacked, each sub-chip is smaller in area than the whole thing would be in 2-dimensions - meaning a lower number of defects & a higher yield on the wafers. Also, the benefit of having support silicon in proximity to system chip is also there for stuff like propagation delay and other "parasitics". Apparently, if the silicon area overhead & the "defect frequency" of the stacking process are not too large, the result could be a cost-savings over conventional 2-dimensional techniques, even when there really isn't much in the area of board space savings According to an IC engineer - for many of these parts, the board space that a design consumes isn't really limited by the die size as much as by the package surface area necessary to dissipate the power consumption - package size then might grow if the heat transfer from the stack wasn't as good as it would be from a planar piece of silicon to the package - but addressing this, an engineer friend at Irvine Sensors suggested that heat transfer can be helped significantly (and easily) by substituting diamond in the packaging substrate for silicon. here's an IRSN pdf which is about a year old (though technically current) on this if anyone is interested: irvine-sensors.com looking at your profile however, seems you are more interested in fast money pushing IRSN's stock than considering the technology - to which i would say be careful - this is not a "breakthrough" patent as far as anyone I've spoken to can tell)