After a month of counting the days, hours, minutes and seconds the Avalon ASIC counter has finally hit zero. The next question that I imagine everyone is asking though is “Are they actually shipping now?”. Well, I have no idea, all I know is that in the latest post on BitcoinTalk BitSyncom say’s they have.
If they make their target there will be 300 brand new 66 GH/s ASIC mining devices about to land on the desks of miners in the next week. That is close to 20 TH/s (or exactly 19,800 GH/s) that may be soon on-line effectively doubling the network hash rate.
Apart from the obvious effect of each unit generating the owner just over 11 bitcoins a day (or just over $170 USD) there are other, more important implications for miners. Implications that may never happen again
The first thing we will note as these devices come on-line is that blocks, that are designed at the current hash rate to take 10 minutes to solve on average will start to be solved faster and faster because of the extra hash power. In fact, with the extra hash/s going on the average block solution may start arriving every 5 minutes, that leads us to the next interesting effect, the re-target.
As you may know, the mechanism to make sure that blocks are solved every 10 minutes on average is the difficulty. The higher the difficulty the longer the blocks take to solve and to make sure that the blocks continue to be solved every 10 minutes every 2016 blocks the bitcoin network adjusts itself (called a re-target) by changing the difficulty to match the network hash rate so that a block is solved every 10 minutes, at 2016 blocks, this should work out to be about 2 weeks.
I theory with blocks being solved every 5 minutes instead of 10 though, its not going to take 2 weeks to solve 2016 blocks, its going to be much faster possibly even as fast as 7 or 8 days.
As it happens, the next re-target is due to happen just a few days after Avalon’s supposed shipping date, so with a few days for shipping the extra hash power could be going on-line just as the re-target happens and we may see the fastest re-target ever. In theory with blocks being solved every 5 minutes instead of 10, it’s not going to take 2 weeks to solve 2016 blocks, its going to be much faster, possibly even as fast as 7 or 8 days.
This all leads to another, and probably the most important effect, the bitcoin block difficulty increase. The current difficulty is sitting at about 2,979,637 and it is not expected to change by much in the next re-target a few days from now, but if Avalon’s ASIC’s then come on-line, dropping the re-target time to 7 days it could be that in 9 or 10 days time with the next re-target we could see a difficulty close to 6 million effectively halving the bitcoin output of every single piece of bitcoin mining hardware in operation (including the new 66 GH/s Avalon ASIC’s).
Add that to the recent block reward halving FPGA’s may still be profitable but GPU miners would have to mine at a loss, especially if they pay for their electricity. Even at a best case scenario of having “free” electricity and using their miners to heat their house you might find that you would be better off selling your mining hardware, buying a heater and using the rest to buy bitcoins directly.
When the difficulty makes this jump I will be certainly turning off my clients GPU rigs (Don’t worry, they already know all about it), and I imagine that a lot of other people will be doing the same (the sane ones at least) and this may lead to another interesting effect, since the difficulty is so high, and the hash power decreasing the next re-target may be in 3 weeks or more, just in time for the next re-target when Butterfly Labs supposedly finally ships their long delayed ASIC bitcoin miners.
All of this is theory of course, not a single ASIC has been seen in operation as of yet and so far all we have seen is delays but if I were you I would be watching.