Topic: An Idea To Increase the Efficiency of Bitcoin Mining by an Order of Magnitude (Read 509 times)

I expect whoever actually launch Bitcoin miner to space have funds to run it's own pool which configured to be tolerant with high delay.

totally agree, why not to use our main power source....
the largest source of energy (work) in the solar system (the sun) as a source of coherence (for logical operations...

best look in finding accurate team!!

a 10 seconds latency in BTC mining is very costly since the mining pool won't be paying you money out of their own pocket, most pools will send new work within a few seconds and thus the majority of the shares you submit will be invalid, mining solo with such latency will also increase your stale block size by a large margin, but this idea is still applicable if the cost of setup and operation is a lot lower than mining on planet earth, although I highly doubt that this will be the case.

There's a big problem with heat dissipation in space. How would this setup cool the miners? It would need enormous radiators for dealing with all the heat.

Unless one breaks the 2nd law of thermodynamics this looks quite bad. But then no need to go in space - no problems with heat, noise, space, and most importantly, no need of any power supply. This would break most PoW assumptions though.

i doubt a government would ever go to that extreme to mine bitcoin. not even sure governments are interested in mining bitcoin for example the usa government can just print more money anytime it wants. or make their own digital currency and control it.


5 second delay is both ways to and from. then satellites around earth have to communicate with things down on the earth surface. another delay. the latencies add up. how many miners miss out on a block reward by seconds? how about 10 seconds?

another issue is, this thing that is out at L1 becomes outdated tech as new tech is developed and put into use on earth that has a larger hash rate. making it even less productive over time...and you can't easily replace it or upgrade it or do any type of maintenance.

in theory yes but in practice, there's other considerations. rocket fuel has to be used to maintain the position so when the rocket fuel runs out it won't be able to maintain its position. the total cost of putting the mining hardware into orbit and such could be rather large. assuming a limited lifetime of operation, it has to recoup the costs and make a profit on top of that. i highly doubt even elon musk would gamble on something like that lol.

as an example, the james webb telescope had a cost of about 9 billion us dollars to put it into orbit at L2. they expect it to have enough propellant to keep it in orbit for 10 years.

also, check my math here but L1 is located 1.5 million km from earth. it takes light about 5 seconds to travel that far. that delay would be a problem for bitcoin mining. could be.

I've just learned about Lagrange Points, which are "points of equilibrium for small-mass objects under the influence of two massive orbiting bodies." Lagrange point 1 is located between the Earth and the Sun.

https://upload.wikimedia.org/wikipedia/commons/e/ee/Lagrange_points2.svg

A Bitcoin Mining operation that relies purely on solar power could remain operational 24/7/365 if it was positioned at Lagrange point 1, with no possibility of clouds obscuring the line-of-sight to the Sun.

Yes, that cost is likely due to the use of a heliostat. While a heliostat is desirable, it's not technically a required component for this technique to work. I can think of two alternatives:

1. The fresnel lens is stationary, no heliostat, and it only concentrates the sun's rays into a laser for a short duration each day.
2. Some brilliant mathematician/physicist/engineer devises a 4π fresnel lens that can focus the sun's rays from any angle into a laser, no heliostat required.

Or, perhaps a supplementary PV panel could be used to provide power to the heliostat.

If the mirrors last long enough, it might pay itself off in "creature comforts".

Well, to be fair to ETFbitcoin, I think most people would mine 24/7 if that was also the most economical mining schedule. But you're right, with this technique the mining would only occur at those times when the intensity of available sunlight is sufficient, which means the miners would not "run all the time without downtime". However, sunlight availability was the primary reason why I wrote that the application of this technology would probably be very limited.

However, if I really allow my imagination to run wild, perhaps this technology could be applied to all data centers. Again, I am not an expert, and this would certainly be an enormous engineering challenge... but, theoretically, if sunlight could be concentrated into a single source of very intense and powerful coherent light, and that light could be sufficiently split and distributed amongst many computers... and if the fresnel lens apparatus (which concentrates the sunlight into coherent light) could seamlessly be swapped for an artificially powered, high-intensity laser during the night to produce an equivalent source of coherent light, then I suppose this could also provide large efficiency benefits to regular ol' data centers. However, I think we're still something like a decade away from general purpose optical computers becoming commercialized.

Additionally, I briefly read that solar-pumped lasers could be used to concentrate sunlight to the effect of increasing the efficiency of solar panels, or something like that, but that has nothing to do with the optical computing stuff.

can you give an example of something that needs to "run all the time without downtime"? bitcoin mining is not such an example.

Yes, but I don't understand how's morality related with the discussion at all; otherwise it wouldn't be tiring from me to just add an adverb and avoid confusion.

My moral philosophy? What makes direct usage of the photons' energy more or less moral than the other way?

i'd say i couldnt agree with you on "very limited applications". such a technology seems like it would be useful in general purpose computing. if it can compute hashes i'm sure it could run a c program.

Whether or not it's bad depends on your moral philosophy, so can't determine that for you. Regarding the question of efficiency, the short answer is "yes, it's less efficient to convert photons into electric power and then use that power to accelerate subatomic charged particles, with mass, inside electrically conductive material to perform computation (which inevitably produces a lot of heat because the particles are charged) than it is to simply use the intrinsically energetic and intrinsically massless photons to perform computation directly." Of course, this is all moot because humanity has not developed this technique of using sunlight to perform computation. This answer also depends on the design and engineering of the technique; a bad design can turn a more efficient physical process into a less efficient human endeavor.

I don't understand much from optical computing, but here's a simple question: What's bad with converting the energy of the photons to electricity? Does it make the procedure less efficient?

Optical computing is extremely slow and clunky. Optical stuff is great for communication though.

It's easy to see why it wouldn't work: the photon just takes too much space - with wavelength hundreds of nanometers it just cannot compete with the density of present day chips. Additionally photons don't interact themselves, so at least one electron is needed for anything to happen. Or to happen probably, maybe. It's way worse: the event probably doesn't happen. The computation is lost then. Which means huge switching devices, and relatively slow speed.

What about the power needed? One must use lasers for all this. Very inefficient.

Edit: typo.

Not exactly zero operational costs - you still have the data center maintenance costs - but this manages to eliminate the need for large power sources, and with it, the main talking point that people use against bitcoin mining (that "crypto mining causes pollution" - but that is a discussion for yesterday). It can be deployed in places like Africa with almost zero disruption to the power grid. You'd still need a backup generator for the systems running the Bitcoin nodes submitting the blocks, and a decent internet line, but yeah. It looks doable.

Well, I think that depends on the technology to which you're referring. There's already commercially viable photonic integrated circuits (PICs) being employed in niche areas of the market. As noted in the OP, a company called LightElligence is making PICs for AI applications.

If you're referring to the entire technique, though--the technique of using sunlight directly for computation--then I think that would have very limited applications, mostly because the availability of sunlight is intermittent. But, I think the efficiency that it could bring to the Bitcoin mining market is reason enough to motivate its development, especially if Bitcoin continues to become a more important and prevalent financial technology.

so what else would the technology be used for? is bitcoin it's most compelling application?