Disco that is going to be a very insightful panel discussion indeed. It is interesting to note (and no I can't provide a source) that a POET optical engine (just the chip) is ball parked at $4 per die. As we know thousands of die can be produced from one wafer...that comment was made publicly in a CC. So if POET was to produce a 100g transceiver the cost should be significantly lower than the targeted $0.25/Gbps optics @ $0.04/Gbps per die (X2 for end to end communications) plus submount cost.
The key to reaching the required price points is wafer level monolithic integration and testing. That is what POET is about and is should enable the lowest cost point in the industry.
