Another thing is what if bitfury gets the same idea and they release their 300 mh/s bitstream with 5% comission.
Yeah, aside from the fact that he's going to make a killing on his own hardware (which has the best power supply of any system I've seen, by a huge margin) and on hosting for it (which I think is a major untapped market -- finding a place to put my mine was a gigantic headache), I've already explained why in his thread.
The "sea of tiny hashers" approach is really easy to reverse-engineer: there's a repeating pattern you can use to easily separate the hashers from the countermeasures, so you don't have to genuinely understand what's going on. In a 64-unrolled ring every stage is different (and the first 8 and last 4 stages are really different) so you have to actually understand the circuit -- or at least the funky stages at the start and end -- in order to separate the signcryption countermeasures from the hasher.
I also encode my public key in the wiring pattern of the circuit, not the LUT tables. So if you want to change the public key you have to actually move wires, which has a high probability of disturbing nearby critical paths.
Yeah, bitfury said he won't be copying your style. I think if he licenses out the bitstream he was going with a hardware AES scheme. The only threat now is just when/if the open source community can catch up to your speed, which might be a while. I'm interested in what enterpoint co can come up with too but I doubt they will be able to match your performance.
