barring some massive investment in bitcoins driving the price up, the pattern will be pretty simple: decreasing number of miners as block rewards go down crossing the threshold for power costs and winter heating serendipity will result in only asic miners with marginal power rates as the last ones standing.
the main issue I see is the funding of the next gen of asic. the majority of the gpu miners were people with disposable income(and a few who went into debt) who spent a large chunk to acquire the next gen hardware with the presumption of future profits offsetting the outlay.
Im assuming only the really dedicated or the ones with low electricity rates were the ones who made the jump to asic or fpga. they however funded most of that development with profits from the gpu mining era. with each block reward halving, the potential money available from mining to be reinvested into hardware development will fall off pretty quick. imagine your R&D budget getting chopped every year, how many companies will have the latest and greatest products every cycle?
anyone who goes past the gpu mining era will have to watch projected earnings vs upgrade cycle very carefully around block halving.
the final group of hashers(will no longer be mining) will be the infrastructure for the bitcoin system, responsible for transaction processing only with very marginal profits/transaction fees. that means the number of asic orders will reach a cutoff point where the volume will be very low making the asp very high. i dont know if the profit margins at that time will be enough to justify hardware replacement on the irregular schedule of system failures.
the main issue I see is the funding of the next gen of asic. the majority of the gpu miners were people with disposable income(and a few who went into debt) who spent a large chunk to acquire the next gen hardware with the presumption of future profits offsetting the outlay.
Im assuming only the really dedicated or the ones with low electricity rates were the ones who made the jump to asic or fpga. they however funded most of that development with profits from the gpu mining era. with each block reward halving, the potential money available from mining to be reinvested into hardware development will fall off pretty quick. imagine your R&D budget getting chopped every year, how many companies will have the latest and greatest products every cycle?
anyone who goes past the gpu mining era will have to watch projected earnings vs upgrade cycle very carefully around block halving.
the final group of hashers(will no longer be mining) will be the infrastructure for the bitcoin system, responsible for transaction processing only with very marginal profits/transaction fees. that means the number of asic orders will reach a cutoff point where the volume will be very low making the asp very high. i dont know if the profit margins at that time will be enough to justify hardware replacement on the irregular schedule of system failures.