No sooner did I post last week’s curation of predictions for the semiconductor industry, and particularly for 3D ICs, than several more popped up online. I suspect this type of thing to continue for the next few weeks, and I will try and continue to cherry-pick the ones that made reference to 3D integration.
In a Semiconductor Packaging News Viewpoint, Doug Goodman, CEO, Ridgetop Group, Inc. talks about the company’s recent work with developing instrumentation for in-situ methods of reliability testing between layers for 2.5D and 3D IC testing. This is just another example of how real work is being done to address the 3D test challenges.
SemiWiki blogger Paul McLellen offered some pretty bold predictions in posts on both EDA Café and in SemiWiki regarding scaling to 20nm, EUV lithography, 3D ICs, mergers, acquisition, and even the demise of one mobile “giant”. (both posts say essentially the same thing, so pick your site of choice to read them). For 3D ICs, McLellan predicts the not-so-earthshattering news: “TSV-based 3D ICs will start to become mainstream. Memory on logic, and mixed digital/analog on interposer. Expect lots of discussion about “more than Moore” and how 3D is the new way for scaling.” Yes, we know.
It doesn’t hurt the 3D effort for target applications to be on the rise. Digitimes reported that global sales of tablets are expected reach 170-180M units in 2013. Not that any of these will contain 3D devices, but growth in this area can’t hurt the cause. Interestingly, the 7-8 inch versions are “ expected to account for over 60% of total tablet shipments. Not surprisingly, Apple is expected to take the lion’s share of the 9-11 inch market, while Android-based models will own the 7-inch sector. But that’s most likely because until late last year, the iPAD mini didn’t exist.
Why only predict the immediate future when you can predict the next forty years? Mentor Graphics’ Wally Rhines takes a stab at that in his EE Times article, Semiconductor Innovation Won’t Lose Steam. He predicts that change and innovation will continue “at a rapid pace for the next forty years” driven not only by smaller feature sizes and larger wafers, but also the “largely untapped opportunity” in the 3rd dimension. He’s not worried that we’ll reach a limit on the number of transistors in demand, because transistors provide access to more and more information, and there is no limit on the amount of information that consumers crave. “More than anything else, semiconductors and computers facilitate our access to information,” he writes.
He also offers a rather convoluted explanation of how over the next 10 years, based on the learning curve predicting price if you know the expected volume, “we should expect to see a 58x increase in transistor count for your inflation-adjusted dollar.” And this will all be enabled by innovation in memory technology that primarily involves 3D technologies, (if I understand his logic correctly). Let’s just hope he’s right.~F.v.T