PoetPower: Any co-relation in the technologies being described Rainer?
Same target, different ways. I don't think they are cooperating.
IBM's main achievement here is that they have succeeded in putting III-V structures on a silicon substrate. This is hard. And I think that's why Geoff Taylor didn't try that in the first place, but used a GaAs substrate instead.
Okay, so IBM has mastered III-V on Si. What's next? To "fabricate transistors and other structures as complex as today's CMOS transistors" would be nice. But not so fast! "More development and optimization will be needed." Oh, well. So we will have to wait some time until those developments and optimizations have successfully happened. And after that we will have to wait some more time until the transistors and other complex structures are ready.
Aside: Will they have p-type transistors or n-type only? For something equivalent to Si CMOS they will have to develop p-type transistors, which is hard. Remember that the GaAs development stalled in the 1990ies, because companies gave up on p-type in GaAs. As far as I know Geoff Taylor was the only one to carry on. He finally succeeded in making positive charges ("holes") move fast.
Back to IBM's achievements. They are assuming that all this "may also lead to active photonics on silicon substrates, since III-V materials are routinely used to make on-chip lasers and other photonic structures." So although this is nothing imminent but somewhat further into the future, they are optimistic that their approach might lead to photonics some day. Nice!
So overall the future is looking bright for IBM and their TASE approach. It might also come to their mind some day to integrate digital and analog circuits on the same die.
However, POET Technologies is already there today – obviously ahead of the curve. On the other hand competition never sleeps! The POET team shouldn't lose any time, but do their utmost to bring their stuff to market soon. And I am confident they are doing just that while we speak.
