3 ms·

> A lot of bitcoin miners use stranded power, which would otherwise go to waste. Often claimed, rarely supported by evidence or numbers. It's also unlikely to

by hannob 3y ago

> A lot of bitcoin miners use stranded power, which would otherwise go to waste. Often claimed, rarely supported by evidence or numbers. It's also unlikely to be very practical, because that'd essentially mean running bitcoin miners in load-balancing mode, and not running them most of the time. Given that bitcoin hardware tends to loose value quickly, as the next generation of mining hardware comes to the market, this is unlikely to be a feasible model.

latchkey 3y ago
I have evidence and numbers. Update: I got downvoted for not posting them. I did in the link above though. https://www.coinmint.one/ https://www.coinmint.one/ is the data center. They have about 500MW of power going to them from the Moses-Saunders dam. They don't need to shut down cause it is hydro and 24/7. They actually help keep the dam running cause they balance the load coming from it. Just like the aluminum smelter before it did. The location of the facility and dam are near the border of Canada and the US. It is very remote and in the middle of nowhere with enough population to consume the 500MW. There are main grid lines going past them, but it would likely need new infrastructure to connect to it at that much power. I've seen their power costs (including transmission), I can't post that, obviously. But, a large chunk of their costs, is transmission, which pays for the install and maintenance of the lines running the few miles from the dam. These are large / tall physical towers. Disclosure: I'm a former very large scale bitcoin/litecoin/ethereum miner and now building an AI bare metal gpu service.
- sirspacey 3y ago
  Would you be willing to post them?
- rainsford 3y ago
  That's actually an interesting situation, but it doesn't support the assertion that the power would "otherwise go to waste" or that this is something a lot of bitcoin miners do. The argument that it's non-trivial to just transmit the power elsewhere is a persuasive one, the argument that bitcoin mining is somehow uniquely able to take advantage of that power is not as persuasive. Even if the argument is that compute heavy data centers are the only use-case that makes sense, there's lots of compute use cases other than bitcoin. And obviously the former presence of the aluminum shelter suggests data centers are not the only good use-cases. I understand the appeal of the argument that bitcoin mining has less energy impact than people think because it can uniquely take advantage of weird edge cases, but it doesn't make sense to me.
  - latchkey 3y ago
    You're trying to apply persuasive logic to something that doesn't need persuasive logic. It is what it is. The fact that the smelter shut down and literally nobody picked it up and put it to use, speaks volumes. Heck, Alcoa was so desperate to unload it, they let those "dirty" bitcoin miners move in, over any other business potential. Here is another one... the Quincy/Wenatchee area of WA state. Also sparsely populated, but more hydro power than anyone knows what to do with. This is where a huge number of data centers are, including bitcoin miners.

by hannob 3y ago

latchkey 3y ago

I have evidence and numbers. Update: I got downvoted for not posting them. I did in the link above though. https://www.coinmint.one/ https://www.coinmint.one/ is the data center. They have about 500MW of power going to them from the Moses-Saunders dam. They don't need to shut down cause it is hydro and 24/7. They actually help keep the dam running cause they balance the load coming from it. Just like the aluminum smelter before it did. The location of the facility and dam are near the border of Canada and the US. It is very remote and in the middle of nowhere with enough population to consume the 500MW. There are main grid lines going past them, but it would likely need new infrastructure to connect to it at that much power. I've seen their power costs (including transmission), I can't post that, obviously. But, a large chunk of their costs, is transmission, which pays for the install and maintenance of the lines running the few miles from the dam. These are large / tall physical towers. Disclosure: I'm a former very large scale bitcoin/litecoin/ethereum miner and now building an AI bare metal gpu service.

sirspacey 3y ago
Would you be willing to post them?
rainsford 3y ago
That's actually an interesting situation, but it doesn't support the assertion that the power would "otherwise go to waste" or that this is something a lot of bitcoin miners do. The argument that it's non-trivial to just transmit the power elsewhere is a persuasive one, the argument that bitcoin mining is somehow uniquely able to take advantage of that power is not as persuasive. Even if the argument is that compute heavy data centers are the only use-case that makes sense, there's lots of compute use cases other than bitcoin. And obviously the former presence of the aluminum shelter suggests data centers are not the only good use-cases. I understand the appeal of the argument that bitcoin mining has less energy impact than people think because it can uniquely take advantage of weird edge cases, but it doesn't make sense to me.
- latchkey 3y ago
  You're trying to apply persuasive logic to something that doesn't need persuasive logic. It is what it is. The fact that the smelter shut down and literally nobody picked it up and put it to use, speaks volumes. Heck, Alcoa was so desperate to unload it, they let those "dirty" bitcoin miners move in, over any other business potential. Here is another one... the Quincy/Wenatchee area of WA state. Also sparsely populated, but more hydro power than anyone knows what to do with. This is where a huge number of data centers are, including bitcoin miners.