That was a great write-up. I've avoided BTC from the start because I always figured someone would commit the algos to silicon and crush everyone else. That said I'm doubtful there is such thing as a ASIC resistance system in anything that has a direct economic value to highly-repeated calculation ratio. Many years ago people would say of graphics systems "You'll never get X poly's/second" etc. But then gaming came along and all those hungry 3d gamers kept buying more and more powerful GPU's. Then data sciences came along and said "wow we can use these GPU's" and onward marches the power of the GPU. I have lived to see so many things over the years go from "impossible" to "common place" in technology.
Another thought that came to mind several times over the years is that China or USA could spool up enough hardware to swamp the BTC networks and crash them. If you think about advancements in these ASIC's and how cheap power can be to large governmental organizations all that is missing is the incentive to attack BTC. I take the recent slow but steady recognition of BTC by various governments with a certain cynical grain of salt. I think others see these things as proof that BTC is getting more generally accepted, but I see them as very large and powerful entities thinking, "If you can't beat them join 'em". It's entirely possible that these steps are more about staying closer to the enemy then being friends with BTC.
When I read this article making claims about heat and electricity being major blockers to centralization I started to think about how google and facebook have been solving these problems with amazing success in the last couple of years. Imagine an "BTC Factory" in a nice cold climate where power is cheap and labor to maintain the facility is cheap. If BTC continues to grow and become predictably tradable for other currencies then these farms could be built with great ease. All we're lacking here is a VC who sees 15% as a reasonable return and a handful of nerds who don't mind building the worlds most powerful mining environment in a 505,000 sq ft facility in North Carolina. Of course this would be a massive investment and could signal an amazing time for BTC or it could signify the centralization of a decentralized process that destroys the original dreams of BTC and yet again puts the people with the most power (literally) in control.
The only thing that stands between today and that potential future are forward thinking people trying to keep the BTC concept alive and continuing to improve what is today's state of art to stay ahead of tomorrow's state of art. I'm still on the sidelines watching but am enjoying the show!
10 megawatt. In Sweden. Datacenter should be at full capacity by now, so ~10 petahash/sec (= 10% of the global Bitcoin network) assuming ~1 watt per Ghash/sec. My numbers are all +/-30% depending on how much hw is 28nm (KnC's Jupiter systems) vs 20nm (Neptune).
How about relatively small BTC mining room heaters - plug them into an electrical supply they provide heat and offset the heating costs with returns from mining?
As it happens I was actually offered a Xerox Lisp Machine to take home when a university department I was associated with was throwing some old kit out. I was living in a tiny flat at the time so it wasn't really an option.
However, now we have plenty space and a property that is very difficult to heat and we need heating ~6 months a year, changing the property is not really an option (planning restrictions are really tight where we live - for good reasons, and any improvements would be incredibly expensive) - BTC room heaters might actually make sense....
The linked article touches upon that concept as a feasible application of mining heat:
"In a home, however, if the outside temperature is less than about 20′C, the cost of electricity is zero; all electricity spent by the miner necessarily eventually turns into “waste” heat, which then heats the home and substitutes for electricity that would be spent by a central heater."
The zero-cost figure is... um... wrong. Your central heater probably either does not use electricity (presumably using a cheaper fuel source) or is a heat pump (and thus gets efficiency greater than 100% on the electricity by stealing heat from the ground). The real cost is still lower than the nominal cost, but it's not zero.
> I started to think about how google and facebook have been solving these problems with amazing success in the last couple of years.
I read once that the way Google solves the heat problem is by not requiring people to be able to walk around the servers. This way they can keep high temperatures in server rooms (which the machines handle fine) and don't spend that much electricity on cooling things down.
Another thought that came to mind several times over the years is that China or USA could spool up enough hardware to swamp the BTC networks and crash them. If you think about advancements in these ASIC's and how cheap power can be to large governmental organizations all that is missing is the incentive to attack BTC. I take the recent slow but steady recognition of BTC by various governments with a certain cynical grain of salt. I think others see these things as proof that BTC is getting more generally accepted, but I see them as very large and powerful entities thinking, "If you can't beat them join 'em". It's entirely possible that these steps are more about staying closer to the enemy then being friends with BTC.
When I read this article making claims about heat and electricity being major blockers to centralization I started to think about how google and facebook have been solving these problems with amazing success in the last couple of years. Imagine an "BTC Factory" in a nice cold climate where power is cheap and labor to maintain the facility is cheap. If BTC continues to grow and become predictably tradable for other currencies then these farms could be built with great ease. All we're lacking here is a VC who sees 15% as a reasonable return and a handful of nerds who don't mind building the worlds most powerful mining environment in a 505,000 sq ft facility in North Carolina. Of course this would be a massive investment and could signal an amazing time for BTC or it could signify the centralization of a decentralized process that destroys the original dreams of BTC and yet again puts the people with the most power (literally) in control.
The only thing that stands between today and that potential future are forward thinking people trying to keep the BTC concept alive and continuing to improve what is today's state of art to stay ahead of tomorrow's state of art. I'm still on the sidelines watching but am enjoying the show!