The thinking in the 1970s for the design of subpicosecond computer architectures seemed to coalesce into two main schools of thought.

One school focussed on superconducting design and the other on optical design.

We have a vested interest in both schools success.

The all optical computer design is based on the assumption of direct photon to photon switching interaction, with minimal delay, in both directions (on and off), for a bistable device.

Some emerging devices have shown limited promise in this reguard, however as yet there are apparently no likely practical candidates for an all optical computer.

One of the most promising devices involves an interferometer with two mirrored facets containing a region of active semiconducting material between.

The remarkable success of this class of devices is the on switching time of a few picoseconds. However the disappointing reality is that the off switching time is in the order of a nanosecond, even after optimization. Apparently the lesson of this configuration was the asymetry of the energy transfer process.

The issue apparently devolves to the rapid alteration of the polarizability (mobile space charge) in the cavity, which is responsible for the complex refractive index related optical path length of the cavity, between the mirror facets.

The solution might lie in the discovery of a new class of nano-engineered material that exhibits suitable symmetry in this reguard, and also posesses reasonable optical properties.