So, do you think BFL will hit their oddly rounded target of 1Gh/s per watt?
We've heard a lot of power estimates from other ASIC manufacturers that are mostly in the range of 200-500Mh/J, with BFL claiming 1000Mh/J, do you think they'll hit this, or is it all likely to turn out incorrect like the early "1000Mh/s @ 20W" estimates from the original BFL single?
Just curious to find out what the community thinks - personally I don't take issue with the idea that BFL will come out with the most power efficient product (they've had more money to throw at the problem than anyone else for starters) but given that the last lot of estimates they made from simulations gave a claimed efficiency of five times what was achieved with the final product I'm somewhat skeptical. Will be interested to see what happens with the BTCFPGA products - supposedly molex powered, implying a limit of ~110W @ 54Gh/s unless they're using multiple plugs, so if BFL can produce a product slightly more efficient they may well hit their numbers.
I've also made up a quick bet if anyone wants to put their money where their mouth is over this:
http://betsofbitco.in/item?id=782