If you were to design an optimal system it just wouldn't have the crypto boxes in the first place. You're starting from a solution and working your way back to

If you were to design an optimal system it just wouldn't have the crypto boxes in the first place. You're starting from a solution and working your way back to a problem. That's a great way to land on a poor solution to the problem as it unnecessarily constraints your search space and installs false constraints. Crypto is not by any means unique in this regard but boy does it happen often. "Assuming I spent $20,000 on a box that converts all electricity fed into it directly into heat, and once in a while, in the process of guessing random numbers, facilitates some very questionable transactions and poops out a magic bean I can sell (only 3% of the time once in its useful life), how can I use it to heat a sauna?" -- is a pretty strange place to start an exploration on the best way to heat saunas. Yes if you start there, a heat exchanger or a heat pump is not a bad idea. But why on earth are we starting there, lol, and you really sure there isn't a better way? If a resistive heater on one side were useful that’s just how it would be done, with or without the false economic incentive.