4 ms·

I believe the argument they were making was about the electricity cost of securing it if all money was in bitcoin. At the current energy requirements (which are

I believe the argument they were making was about the electricity cost of securing it if all money was in bitcoin. At the current energy requirements (which are escalating per transaction), it would take more than the entire energy production of the world to make bitcoin the sole currency.

thebean11 6y ago
> which are escalating per transaction Wrong. The number (or more specifically "weight") of the transactions that can be processed in a given time period is constant, and does not scale with the hash rate. > it would take more than the entire energy production of the world to make bitcoin the sole currency I do not follow..
- rspeele 6y ago
  Let me see if I can help you. One of Bitcoin's biggest selling points is that there is a limited supply of it. For it to be the world's reserve currency, each 1 BTC would represent a stunning amount of actual wealth. You can find plenty of starry-eyed investors using this principle to explain how their price target of, say, $10M/BTC makes sense. Sound good so far? Next, the block reward is denominated in BTC. Currently 6.25 BTC every 10 minutes, but it'll halve again in 2024. The block reward, plus transaction fees, is what the network pays miners. It makes sense for miners in aggregate to expend about that much money on mining, because if the network is spending less than that, there is still profit available for additional miners joining in. If that 6.25 BTC becomes valued at 10x or 100x as much as it is today, the expenditure on mining will also be 10x or 100x larger. Obviously this can't actually exceed the energy production of the world because that's impossible, but this new artificial demand for energy can soak up lots of capacity that would otherwise go to more productive, yet less profitable, endeavors. The fact that it's using 1/200th of the world's energy production already, while it's just a toy speculation vehicle, doesn't bode well. As you correctly point out though, this additional expenditure on mining will do absolutely nothing to help the number of transactions that can be processed in any given time period.
  - thebean11 6y ago
    > Let me see if I can help you. Nobody here is impressed by you, the condescension sounds ridiculous. > the expenditure on mining will also be 10x or 100x larger This is correct, but the assumption that energy will continue to represent a very large part of that expenditure is faulty. There is a huge incentive for miners to mine more efficiently, especially in this hypothetical universe where Bitcoin is driving up energy prices.. That expenditure could easily be disproportionately spent on more efficient hardware / algorithms. Alternatively Bitcoin might not be using proof of work in this hypothetical future. It's not immutable.
    - tlb 6y ago
      We've had several generations of more efficient hardware, and it just causes the hash rate to increase while energy usage stays at the energy-cost equilibrium described above. I can't imagine an algorithmic or hardware improvement that would stop a proof of work system scaling with energy input.
      - thebean11 6y ago
        Energy is still somewhat cheap relative to the value of the hardware, so there's been no incentive to optimize for energy usage. This won't change until there's mass production and the hardware is ubiquitous. You might be right though, but like I said I don't think PoW is necessarily the final state.
    - rspeele 6y ago
      Sorry, I didn't intend to sound condescending, though I see how I did. I was responding to your "I do not follow...". I suspect mining expenditure would still be mostly spent on energy usage, not finding more efficient hardware/algorithms, since mining is an arms race -- when miner A finds a more efficient algorithm, it's only a matter of time before miner B finds it too, and then they're back on level ground competing over energy. But even if this is not the case, I don't particularly care whether the mining expenditure is spent on energy, or developing and manufacturing new mining hardware, or sacrificing goats to a hash-finding deity. Since the amount of mining the network does is set by a price tag (the block reward), and its intent is to make 51%ing prohibitively costly, by definition it must use up some valuable resource. Whether that's energy or human ingenuity or semiconductor production is just an implementation detail but it'll always be a waste. If Bitcoin switches away from proof of work, I'll stop criticizing it for the inefficiency of proof of work.

thebean11 6y ago

> which are escalating per transaction Wrong. The number (or more specifically "weight") of the transactions that can be processed in a given time period is constant, and does not scale with the hash rate. > it would take more than the entire energy production of the world to make bitcoin the sole currency I do not follow..

rspeele 6y ago
Let me see if I can help you. One of Bitcoin's biggest selling points is that there is a limited supply of it. For it to be the world's reserve currency, each 1 BTC would represent a stunning amount of actual wealth. You can find plenty of starry-eyed investors using this principle to explain how their price target of, say, $10M/BTC makes sense. Sound good so far? Next, the block reward is denominated in BTC. Currently 6.25 BTC every 10 minutes, but it'll halve again in 2024. The block reward, plus transaction fees, is what the network pays miners. It makes sense for miners in aggregate to expend about that much money on mining, because if the network is spending less than that, there is still profit available for additional miners joining in. If that 6.25 BTC becomes valued at 10x or 100x as much as it is today, the expenditure on mining will also be 10x or 100x larger. Obviously this can't actually exceed the energy production of the world because that's impossible, but this new artificial demand for energy can soak up lots of capacity that would otherwise go to more productive, yet less profitable, endeavors. The fact that it's using 1/200th of the world's energy production already, while it's just a toy speculation vehicle, doesn't bode well. As you correctly point out though, this additional expenditure on mining will do absolutely nothing to help the number of transactions that can be processed in any given time period.
- thebean11 6y ago
  > Let me see if I can help you. Nobody here is impressed by you, the condescension sounds ridiculous. > the expenditure on mining will also be 10x or 100x larger This is correct, but the assumption that energy will continue to represent a very large part of that expenditure is faulty. There is a huge incentive for miners to mine more efficiently, especially in this hypothetical universe where Bitcoin is driving up energy prices.. That expenditure could easily be disproportionately spent on more efficient hardware / algorithms. Alternatively Bitcoin might not be using proof of work in this hypothetical future. It's not immutable.
  - tlb 6y ago
    We've had several generations of more efficient hardware, and it just causes the hash rate to increase while energy usage stays at the energy-cost equilibrium described above. I can't imagine an algorithmic or hardware improvement that would stop a proof of work system scaling with energy input.
    - thebean11 6y ago
      Energy is still somewhat cheap relative to the value of the hardware, so there's been no incentive to optimize for energy usage. This won't change until there's mass production and the hardware is ubiquitous. You might be right though, but like I said I don't think PoW is necessarily the final state.
  - rspeele 6y ago
    Sorry, I didn't intend to sound condescending, though I see how I did. I was responding to your "I do not follow...". I suspect mining expenditure would still be mostly spent on energy usage, not finding more efficient hardware/algorithms, since mining is an arms race -- when miner A finds a more efficient algorithm, it's only a matter of time before miner B finds it too, and then they're back on level ground competing over energy. But even if this is not the case, I don't particularly care whether the mining expenditure is spent on energy, or developing and manufacturing new mining hardware, or sacrificing goats to a hash-finding deity. Since the amount of mining the network does is set by a price tag (the block reward), and its intent is to make 51%ing prohibitively costly, by definition it must use up some valuable resource. Whether that's energy or human ingenuity or semiconductor production is just an implementation detail but it'll always be a waste. If Bitcoin switches away from proof of work, I'll stop criticizing it for the inefficiency of proof of work.