Seems like a lot of people are assuming that the Gen 3 will be a success.
It is still too early to make such a call considering that the Gen 2 had to be cancelled due to the under performing chip design.
Even if the Gen 3 chip does perform, there are still 3-4 months before they are implemented in functional miners, and that is a very long time in the Bitcoin world.
If gen 3 is lower cost to run it won't matter when it comes out. It will simply outhash everything out there and make older chips obsolete.
Yes, fair enough.
But look at what happened to BFL, Avalon, Cointera, HashFast, ActM, and even AsicMiner. Everyone's roadmap was quite different to what they initially planned. The chances of the chip design to under perform or having a big delay is quite likely.