To me, the most interesting part of the Intel video is from 10:30 to 12:55 min. where Robert Blum explains that Intel manufactures and tests on wafer-scale. Interestingly, he describes that as “unique”. So, he either does not know what POET does, or he feels that it would better to not mention it. Well, POET also claims that they are the only ones who can create a whole tray of icecubes at a time.

However, bringing some InP onto the Si layer and then forming a laser or other active component out of it, is certainly a different approach than POET’s. While I don’t know about the particular characteristics and limitations of such lasers etc., I do imagine that there are limitations, due to the particular nature of the manufacturing process. A plus is that they don’t have coupling losses because they don’t have coupling.

This is different from what POET does. As Fairchij has already pointed out, POET’s architecture is totally flexible in this regard and can – at least in principle – use any laser that is on the market and that might be better suited to your particular application. The only thing you have to do is to manufacture such a laser with a specialized “plug” that would “click” into the “socket” the optical interposer provides. At the moment, such lasers are developed and delivered by Almae, DenseLight, and Sanan IC.

17:34 min. in the Intel video is also interesting. It shows the internals of an Intel CWDM4 transceiver. While Robert Blum says that it contains „mostly air“, I do not share the same grade of impressivedness. Intel’s solution requires much more space (and chips) than POET’s. No way to place four of Intel’s modules into that housing! So POET is clearly leading in miniaturization and thus enabling new classes of application that Intel can’t, at least not for now.

Like Fairchij, I’d be very interested in our management’s comments on this.