I strongly suspect that we'll see a glut of mining hardware on the market in Q3 that hardware manufacturers can't sell at a profit, which will lead to consolidation and some companies folding.
Would you care to elaborate on this? This is difficult to ascertain, in my opinion because we don't have their profitability numbers. I just hope that the scammiest, lying companies will not survive. People in mining are incredibly forgiving, even stupidly so.
Without a technological breakthrough, ~.5W/GH looks to be about the limit using current technology, and we're rapidly approaching the point where multiple companies have fairly equivalent hardware (Bitmain, KNC, Spondoolies, and Bitfury). Future power efficiency improvements are going to be very hard to come by.
There are fixed costs associated with all miners for things like heatsinks, fans, power supplies, PCBs, board components, chips, manufacturing, and shipping, so there is a lower limit on miner pricing.
We know that Bitmain was able to make a profit when they sold the S1 at ~$250, and I suspect they can do the same with the S3, which would put the bottom floor cost of a S3 at around .5$/GH.
Spondoolies sold the SP30 at .69$/GH for the Roadstress group buy, and they have said multiple times that they won't sell hardware at a loss in SP10 discussions, so I strongly suspect their SP30 costs are around .5$/GH.
S3 ROI numbers look iffy at $1/GH and a ~17B difficulty, so who is going to buy all this new hardware at a 34B or 51B difficulty if the price to manufacture the hardware is ~.5$/GH?
Unless the price of BTC goes up, I think we're rapidly reaching the point where it no longer makes economic sense to manufacture new hardware, and will reach that point in Q4.
