[old-ager]

> I believe you're missing something.

Right; a tunnel diode has negative resistance in part of its characteristic curve; therefore it has gain.

[stevio]

How does it have gain, as a traditional doide has only two terminals? Is the substrate charged?

[expatpat]

It's true that the negative resistance adds functionality relative to a simple diode. Furthermore, several III-V quantum-tunneling diodes can be made on one site (we once made an A/D converter from a bunch of them). However, the added functionality is more useful for analog circuits (oscillators, ADCs, etc.) than the digital circuits these guys are hawking.