Space-based Bitcoin Mining

Preface: if you’re a bitcoin expert, gird your loins because I’m probably going to be making a LOT of technical errors in this post.

So: space based solar power, but instead of beaming the power to Earth, we mine bitcoin with it.

OK. I’m indifferent to Bitcoin. I generally think it’s a dumb idea but that has some interesting technology. But it’s useful conceptually for imagining a straightforward way to turn computing power into money. Or, once progress on mining hardware slows down to the now-slow Moore’s law of 4 year doublings, a way to turn electricity directly into money while only having to transmit a relatively short amount of data.

What it is, very briefly, is a kind of “cryptocurrency” designed to do away with a central ledger, instead using a sort of consensus of many distributed computers as a ledger. Money is added to the money supply by “mining,” which in this case just means doing some pointless but hard and verifiable calculation until you randomly free up a “block” which is a bunch of bitcoins (each bitcoin is like $750 right now… easily subdivided, by the way). Mining difficulty is adjusted to keep the pace of adding to the money supply at a steady but very slowly decreasing rate.

I googled “bitcoin mining hardware,” and this popped up: https://www.bitcoinmining.com/bitcoin-mining-hardware/
The best hardware listed uses about 1200 Watts of electricity (all turned into heat) and produces roughly $4 of BTC per day and costs $400-something (…while consuming about $400 of electricity at 14 cents per kWh…). And weighs about 4 kg, so 300W/kg. So a dollar gets you 3 Watts of useful bitcoin mining, and it takes about 100 days for it to pay for its own hardware cost assuming mining difficulty doesn’t change. (Which is a terrible assumption, as more and more mining hardware comes online, the actual yield of bitcoin mining reduces to compensate.)

But only 33 cents per watt-used is great! That means we can replace that super expensive laser or microwave transmitter and receiver with cheap computer chips and radiators. And double or quadruple the power is available, too, because we don’t have all those transmission losses. But we do need to radiate all that heat away, so we’ll need a few more radiators, but in principle that’s not too bad. Radiators in this temperature range are like 100W/kg, so if we assume about 200W/kg for the solar power and a little extra for comms and pointing, we’re looking at about 50Wbitcoin/kg launched to orbit.

So at $100/kg to orbit (same price as the mining hardware, by the way!), and another $100/kg for hardware for a total cost of $200/kg, We have about $4/Watt-bitcoin. At 4 year replacement cycles (probably the power and thermal system would last a lot longer, but hardware would need to be updated), that’s about 11 cents per kWh of bitcoin processing, which is competitive (since it includes the mining hardware as well as the power). At longer intervals, even better.

But of course, this assumes bitcoin remains popular WHILE bitcoin mining hardware stagnates. The latter is not a terrible assumption, given that we’ve essentially reached the end of Moore’s Law (we’ll still see large improvements, but over longer timeframes, like 4 years instead of 2 years): http://arstechnica.com/information-technology/2016/02/moores-law-really-is-dead-this-time/

The following two tabs change content below.
This entry was posted in Uncategorized. Bookmark the permalink.

2 Responses to Space-based Bitcoin Mining

  1. Bob Steinke says:

    Okay, let’s napkin-business-plan this one.

    Competitiors:

    Terrestrial bitcoin mining – This one is actually pretty easy to bound. As mentioned, if there’s a big increase in mining activity the returns (valued in bitcoin) drop. But the only reason there would be a big increase in mining activity would be if bitcoin popularity was going up so more dollars per bitcoin. You can probably say that the dollars reaped by a given mining effort will hover a little above the terrestrial cost of that mining effort. So the net present value of this idea is any reduction in cost over terrestrial mining using grid electricity.

    That’s going to be the important calculation to make. I don’t know much about the state of the art in bitcoin mining, but I could imagine they use off-peak electricity and pay less. Maybe they piggyback on unused cycles of machines used for something else. It would be really important to get your ducks in a row on this calculation and make sure you aren’t underestimating your competition, but once you’ve got a good number you can calculate at what launch cost it would have a positive net present value.

    Risks:

    COTS mining hardware will be glitchy in orbit. Rad hard mining hardware will be expensive (liklihood high, severity negligible to managable)

    Each individual calculation is relatively unimportant so throwing out a fraction of them just impacts productivity. You should be able to predict the rate of glitches and derate your productivity accordingly.

    Bitcoin falls out of favor and crashes to zero (liklihood ???, severity catastrophic)

    Based on this risk, investors would probably ask for 20-30% IRR. If you can provide that at available launch costs you could probably get funded.

  2. N/A says:

    The thing with terrestrial mining rigs is, if the rig was profitable (hardware+facility+electricity+ops costs < bitcoin value), why is the rig manufacturer selling rigs when they can mint their own coins? Effectively, you are betting the rig is profitable for bitcoin mining when the manufacturer thinks it's marginally profitable or a loss maker (for mining, not for hardware sales per se).

    So you get weird situations like chinese soft brown coal mines in the hinterlands burning the polluting (thus low price) soft coal locally to power bitcoin rigs in open wall shacks with free air cooling (and some fans), or rural dams with poor transmission infrastructure preventing power export. Low ops, low facility, low electricity cost.

    Now, mining may be a fools game for most people, but exchanges and network supernodes are a different situation. If the bitcoin network (which agrees on the blockchain ledger) had a number of critical nodes in orbit, which one can communicate with via direct transmission (LPI RF or lasercomm), bypassing national telecomms infrastructure, one can provide ledger security services while potentially operating supranationally (well, at least until the sat operator gets taken over).

    If not bitcoin per se, one might consider a lasercomm cubesat network for something like GNU Taler or zCash.

Leave a Reply

Your email address will not be published. Required fields are marked *