7 ms·

> Somebody on a different thread estimated you could drive a Tesla down the whole California coastline for the energy used processing a single BTC transaction.

by noch 6y ago

> Somebody on a different thread estimated you could drive a Tesla down the whole California coastline for the energy used processing a single BTC transaction. Bitcoin energy consumption is not a function of the number of transactions. A node broadcasts a transaction which miners then include in a block which, within limits, can have an arbitrary number of transactions. It is the Proof of Work(PoW) contest to find a valid hash for the block at a certain level of difficulty that creates a valid block. Further, with Layer 2 solutions, a single transaction in a Bitcoin block can be "dense" with the value of an arbitrary number of transactions that occurred in Layer 2. That is Bitcoin providing the settlement layer. Importantly, because Bitcoin is a timechain where the current blockheight represents the entire history of Bitcoin. "The average cost per transaction isn’t an adequate metric for measuring the efficiency of Bitcoin’s PoW, it should be defined in terms of the security of an economic history. The energy spend secures the stock of bitcoin, and that percentage is going down over time as inflation decreases. A Bitcoin “accumulates” the energy associated with all the blocks mined since its creation. LaurentMT, a researcher, has found empirically that Bitcoin’s PoW is indeed becoming more efficient over time: increasing cost is counterbalanced by the even greater increasing total value secured by the system."[0] [0] : https://danhedl.medium.com/pow-is-efficient-aa3d442754d3 https://danhedl.medium.com/pow-is-efficient-aa3d442754d3

Grustaf 6y ago
The only reasonable metric is energy usage per NECESSARY effect. If bitcoin spends inordinate amounts of energy on storing historic transactions, that is not our problem. That is not a value most people care about.
- noch 6y ago
  > That is not a value most people care about. People in a free market pay for what they care about. To date, those who use Bitcoin have been happy to pay for the energy the network consumes. In the end, that is all that matters.
  - Grustaf 6y ago
    I'm not saying people are not willing to pay for it, I'm saying it's inefficient and specifically that the metric mentioned is not relevant.
    - noch 6y ago
      > I'm saying it's inefficient Compared to what else that provides a decentralised, censorship resistant monetary network without trusted third parties required to secure it? > the metric mentioned is not relevant It's relevant in explaining the energy usage relative to economic value and architecture of an engineered monetary system. If you have a metric that can explain Bitcoin's design and value proposition with respect to energy consumption better than Dan Held's explanation, you should propose it and logically demonstrate your idea's superior explanatory and predictive power.
      - Grustaf 6y ago
        > Compared to what else that provides a decentralised, censorship resistant monetary network without trusted third parties required to secure it? I don't particularly feel that any of those characteristics are very important, and I think most people would agree. > It's relevant in explaining the energy usage relative to economic value and architecture of an engineered monetary system Again, that is not how you construct a metric for how cost efficient something is. You start with the "job to be done" of the object. And storing all historical transactions forever is seldom desirable. If you want to measure the efficiency of electric cars vs combustion engine ones you look at things like miles/Wh or CO2 emissions/mile. You DON'T look at explosions per second or something like that. It might explain how combustion engines work but it is not the goal of a car.
        noch 6y ago
        > I don't particularly feel that any of those characteristics are very important, and I think most people would agree. Bitcoin doesn't care what you feel. It exists as an engineering artifact, continues to exist and functions to serve the needs of those who use it. Those who use it don't pay for it with their feelings but with their hard earned money. You evaluate a piece of engineering for what it aims to achieves not what you feel it should achieve. Bitcoin at block height 670061 has a market cap >USD800B. That's a lot of people who feel the characteristics of Bitcoin are valuable. Cf. A statement like: "I don't particularly feel HD screens on a phone are important and I think most people would agree that they're a waste and bad for the environment. Therefore HD screens shouldn't be in smart phones." > if you want to measure the efficiency of electric cars vs combustion engine ones you look at things like miles/Wh or CO2 emissions/mile. > You DON'T look at explosions per second or something like that. It might explain how combustion engines work but it is not the goal of a car. Again, Dan Held said: "the efficiency of Bitcoin’s PoW, should be defined in terms of the security of an economic history."[0] That history is perhaps analogous to the entire lifetime travel distance of the vehicle. Let a coin be a vehicle. There are 21M coins. Let a block be a mile long and infinitely wide (for transactions of any value can be contained in a block without limit to the value), for instance, and you can get your miles/Wh. Or just humbly study the Bitcoin timechain and understand what it does and why it does it. [0]: https://danhedl.medium.com/pow-is-efficient-aa3d442754d3 https://danhedl.medium.com/pow-is-efficient-aa3d442754d3
- ric2b 6y ago
  That argument is similar to saying a nuclear power plant is ridiculously expensive because it started operating last week and is currently only connected to a single factory. The energy usage of Bitcoin doesn't meaningfully go up with more transactions, so pretending that each transaction uses some specific amount of energy is just misleading.
  - ForHackernews 6y ago
    > The energy usage of Bitcoin doesn't meaningfully go up with more transactions, No it just always always goes up regardless of transaction volume as the block-difficulty increases infinitely. You've built a system that by design can literally only become less efficient. It's actually even worse and stupider than if it were a per-transaction energy usage.
    - noch 6y ago
      > No it just always always goes up regardless of transaction volume as the block-difficulty increases infinitely. [sigh] The difficulty adjustment is so a block is issued on average every 10 minutes. It therefore does not increase infinitely but both increases and decreases depending on the amount of hashpower in the network. https://en.bitcoin.it/wiki/Difficulty https://en.bitcoin.it/wiki/Difficulty
      - ForHackernews 6y ago
        Has it ever decreased? > What is the maximum difficulty? > There is no minimum target. The maximum difficulty is roughly: maximum_target / 1 (since 0 would result in infinity), which is a ridiculously huge number (about 2^224).
        noch 6y ago
        > Has it ever decreased? Obviously. https://www.coindesk.com/bitcoin-mining-difficulty-large-drop https://www.coindesk.com/bitcoin-mining-difficulty-large-dro...
  - 6y ago
- ForHackernews 6y ago
  Furthermore nothing about bitcoin is "necessary", even in some loose consumer-demand sense of the term. - There are other, better cryptocurrencies that don't bake the planet. - There are other, better financial infrastructures that process more transactions and provide a better stable store of value. - There are other, better distributed computing networks that do more than calculate sha256 over and over again.
sharpneli 6y ago
> within limits And those limits come up to the theoretical maximum of was it 7 or so and in practice around 4 https://www.blockchain.com/charts/transactions-per-second https://www.blockchain.com/charts/transactions-per-second Also PoW is hilariously inefficient. Because the amount of energy one needs to spend is basically related to the amount of value one could grab by cheating, so if all economy would be based on Bitcoin we'd be talking about basically cooking the planet due to PoW. When Satoshi originally devised PoW it was more about owning the hardware and using that as basis for voting. Energy costs were miniscule. He simply didn't foresee the, now obvious, result that it will became a function of burned energy rather than amount of hardware out there.
- noch 6y ago
  > And those limits come up to the theoretical maximum of was it 7 or so and in practice around 4 > https://www.blockchain.com/charts/transactions-per-second https://www.blockchain.com/charts/transactions-per-second The link you provided indicates transactions added to the mempool per second! The mempool is a cache of broadcast transactions from which miners pick transactions to be added to the next block. At the current block height, 2700 transactions were in the block. https://mempool.space/block/00000000000000000007a3688746c1d04d1b9619ebdc4813b843fbf995bc9f0c https://mempool.space/block/00000000000000000007a3688746c1d0... > Also PoW is hilariously inefficient. Because the amount of energy one needs to spend is basically related to the amount of value one could grab by cheating Your statement is entirely unclear. Could you restate it? For Steelman purposes, I'll ignore the non-sequitur about cheating. You claim that algorithm x is inefficient in absolute terms? Assuming you're an engineer, you know that engineering operates by comparisons and tradeoffs. PoW is how Bitcoin provides final settlement of transactions and unforgeable costliness compared to existing systems of money (e.g. fiat) and money transfer(e.g. visa, paypal, swift) which are all centralised and built for censorship, while bitcoin is decentralised and censorship resistant. And it provides final settlement of transactions every hour for thousands of transactions. "Finality" and "Unforgeability" are the terms you should attend to. These are both tied to hashrate which is tied to the profitability of mining the heaviest chain. "Bitcoin has unforgeable costliness, because it costs a lot of electricity to produce new bitcoins. Producing bitcoins cannot be easily faked [..]" [0] "So what are settlement assurances exactly? They refer to a system’s ability to grant recipients confidence that an inbound transaction will not be reversed. Wire transfers using a messaging system like SWIFT are popular in part because they are practically impossible to reverse. They are considered safe for recipients because originating banks will only release the funds if they are fully present in the sender’s account. [...] recipients of a Bitcoin transaction can have extremely high confidence that, once buried under a few blocks, a transaction is unlikely to be reversed."[1] > When Satoshi originally devised PoW [...] Energy costs were miniscule. I'm not sure what you mean. In general, energy costs seem to be dropping over time https://ourworldindata.org/grapher/levelized-cost-of-energy https://ourworldindata.org/grapher/levelized-cost-of-energy > [Satoshi] simply didn't foresee the, now obvious, result that it will became a function of burned energy rather than amount of hardware out there. I'm wary of trying to interpret Satoshi, but here are his own words: "I think the case will be the same for Bitcoin. The utility of the exchanges made possible by Bitcoin will far exceed the cost of electricity used. Therefore, not having Bitcoin would be the net waste. [...] Each node’s influence on the network is proportional to its CPU power. The only way to show the network how much CPU power you have is to actually use it." [2] [0]: https://medium.com/@100trillionUSD/modeling-bitcoins-value-with-scarcity-91fa0fc03e25 https://medium.com/@100trillionUSD/modeling-bitcoins-value-w... [1]: https://medium.com/@nic__carter/its-the-settlement-assurances-stupid-5dcd1c3f4e41 https://medium.com/@nic__carter/its-the-settlement-assurance... [2]: https://satoshi.nakamotoinstitute.org/posts/bitcointalk/327/ https://satoshi.nakamotoinstitute.org/posts/bitcointalk/327/

by noch 6y ago

Grustaf 6y ago

The only reasonable metric is energy usage per NECESSARY effect. If bitcoin spends inordinate amounts of energy on storing historic transactions, that is not our problem. That is not a value most people care about.

noch 6y ago
> That is not a value most people care about. People in a free market pay for what they care about. To date, those who use Bitcoin have been happy to pay for the energy the network consumes. In the end, that is all that matters.
- Grustaf 6y ago
  I'm not saying people are not willing to pay for it, I'm saying it's inefficient and specifically that the metric mentioned is not relevant.
  - noch 6y ago
    > I'm saying it's inefficient Compared to what else that provides a decentralised, censorship resistant monetary network without trusted third parties required to secure it? > the metric mentioned is not relevant It's relevant in explaining the energy usage relative to economic value and architecture of an engineered monetary system. If you have a metric that can explain Bitcoin's design and value proposition with respect to energy consumption better than Dan Held's explanation, you should propose it and logically demonstrate your idea's superior explanatory and predictive power.
    - Grustaf 6y ago
      > Compared to what else that provides a decentralised, censorship resistant monetary network without trusted third parties required to secure it? I don't particularly feel that any of those characteristics are very important, and I think most people would agree. > It's relevant in explaining the energy usage relative to economic value and architecture of an engineered monetary system Again, that is not how you construct a metric for how cost efficient something is. You start with the "job to be done" of the object. And storing all historical transactions forever is seldom desirable. If you want to measure the efficiency of electric cars vs combustion engine ones you look at things like miles/Wh or CO2 emissions/mile. You DON'T look at explosions per second or something like that. It might explain how combustion engines work but it is not the goal of a car.
      - noch 6y ago
        > I don't particularly feel that any of those characteristics are very important, and I think most people would agree. Bitcoin doesn't care what you feel. It exists as an engineering artifact, continues to exist and functions to serve the needs of those who use it. Those who use it don't pay for it with their feelings but with their hard earned money. You evaluate a piece of engineering for what it aims to achieves not what you feel it should achieve. Bitcoin at block height 670061 has a market cap >USD800B. That's a lot of people who feel the characteristics of Bitcoin are valuable. Cf. A statement like: "I don't particularly feel HD screens on a phone are important and I think most people would agree that they're a waste and bad for the environment. Therefore HD screens shouldn't be in smart phones." > if you want to measure the efficiency of electric cars vs combustion engine ones you look at things like miles/Wh or CO2 emissions/mile. > You DON'T look at explosions per second or something like that. It might explain how combustion engines work but it is not the goal of a car. Again, Dan Held said: "the efficiency of Bitcoin’s PoW, should be defined in terms of the security of an economic history."[0] That history is perhaps analogous to the entire lifetime travel distance of the vehicle. Let a coin be a vehicle. There are 21M coins. Let a block be a mile long and infinitely wide (for transactions of any value can be contained in a block without limit to the value), for instance, and you can get your miles/Wh. Or just humbly study the Bitcoin timechain and understand what it does and why it does it. [0]: https://danhedl.medium.com/pow-is-efficient-aa3d442754d3 https://danhedl.medium.com/pow-is-efficient-aa3d442754d3
ric2b 6y ago
That argument is similar to saying a nuclear power plant is ridiculously expensive because it started operating last week and is currently only connected to a single factory. The energy usage of Bitcoin doesn't meaningfully go up with more transactions, so pretending that each transaction uses some specific amount of energy is just misleading.
- ForHackernews 6y ago
  > The energy usage of Bitcoin doesn't meaningfully go up with more transactions, No it just always always goes up regardless of transaction volume as the block-difficulty increases infinitely. You've built a system that by design can literally only become less efficient. It's actually even worse and stupider than if it were a per-transaction energy usage.
  - noch 6y ago
    > No it just always always goes up regardless of transaction volume as the block-difficulty increases infinitely. [sigh] The difficulty adjustment is so a block is issued on average every 10 minutes. It therefore does not increase infinitely but both increases and decreases depending on the amount of hashpower in the network. https://en.bitcoin.it/wiki/Difficulty https://en.bitcoin.it/wiki/Difficulty
    - ForHackernews 6y ago
      Has it ever decreased? > What is the maximum difficulty? > There is no minimum target. The maximum difficulty is roughly: maximum_target / 1 (since 0 would result in infinity), which is a ridiculously huge number (about 2^224).
      - noch 6y ago
        > Has it ever decreased? Obviously. https://www.coindesk.com/bitcoin-mining-difficulty-large-drop https://www.coindesk.com/bitcoin-mining-difficulty-large-dro...
- 6y ago
ForHackernews 6y ago
Furthermore nothing about bitcoin is "necessary", even in some loose consumer-demand sense of the term. - There are other, better cryptocurrencies that don't bake the planet. - There are other, better financial infrastructures that process more transactions and provide a better stable store of value. - There are other, better distributed computing networks that do more than calculate sha256 over and over again.

sharpneli 6y ago

> within limits And those limits come up to the theoretical maximum of was it 7 or so and in practice around 4 https://www.blockchain.com/charts/transactions-per-second https://www.blockchain.com/charts/transactions-per-second Also PoW is hilariously inefficient. Because the amount of energy one needs to spend is basically related to the amount of value one could grab by cheating, so if all economy would be based on Bitcoin we'd be talking about basically cooking the planet due to PoW. When Satoshi originally devised PoW it was more about owning the hardware and using that as basis for voting. Energy costs were miniscule. He simply didn't foresee the, now obvious, result that it will became a function of burned energy rather than amount of hardware out there.

noch 6y ago
> And those limits come up to the theoretical maximum of was it 7 or so and in practice around 4 > https://www.blockchain.com/charts/transactions-per-second https://www.blockchain.com/charts/transactions-per-second The link you provided indicates transactions added to the mempool per second! The mempool is a cache of broadcast transactions from which miners pick transactions to be added to the next block. At the current block height, 2700 transactions were in the block. https://mempool.space/block/00000000000000000007a3688746c1d04d1b9619ebdc4813b843fbf995bc9f0c https://mempool.space/block/00000000000000000007a3688746c1d0... > Also PoW is hilariously inefficient. Because the amount of energy one needs to spend is basically related to the amount of value one could grab by cheating Your statement is entirely unclear. Could you restate it? For Steelman purposes, I'll ignore the non-sequitur about cheating. You claim that algorithm x is inefficient in absolute terms? Assuming you're an engineer, you know that engineering operates by comparisons and tradeoffs. PoW is how Bitcoin provides final settlement of transactions and unforgeable costliness compared to existing systems of money (e.g. fiat) and money transfer(e.g. visa, paypal, swift) which are all centralised and built for censorship, while bitcoin is decentralised and censorship resistant. And it provides final settlement of transactions every hour for thousands of transactions. "Finality" and "Unforgeability" are the terms you should attend to. These are both tied to hashrate which is tied to the profitability of mining the heaviest chain. "Bitcoin has unforgeable costliness, because it costs a lot of electricity to produce new bitcoins. Producing bitcoins cannot be easily faked [..]" [0] "So what are settlement assurances exactly? They refer to a system’s ability to grant recipients confidence that an inbound transaction will not be reversed. Wire transfers using a messaging system like SWIFT are popular in part because they are practically impossible to reverse. They are considered safe for recipients because originating banks will only release the funds if they are fully present in the sender’s account. [...] recipients of a Bitcoin transaction can have extremely high confidence that, once buried under a few blocks, a transaction is unlikely to be reversed."[1] > When Satoshi originally devised PoW [...] Energy costs were miniscule. I'm not sure what you mean. In general, energy costs seem to be dropping over time https://ourworldindata.org/grapher/levelized-cost-of-energy https://ourworldindata.org/grapher/levelized-cost-of-energy > [Satoshi] simply didn't foresee the, now obvious, result that it will became a function of burned energy rather than amount of hardware out there. I'm wary of trying to interpret Satoshi, but here are his own words: "I think the case will be the same for Bitcoin. The utility of the exchanges made possible by Bitcoin will far exceed the cost of electricity used. Therefore, not having Bitcoin would be the net waste. [...] Each node’s influence on the network is proportional to its CPU power. The only way to show the network how much CPU power you have is to actually use it." [2] [0]: https://medium.com/@100trillionUSD/modeling-bitcoins-value-with-scarcity-91fa0fc03e25 https://medium.com/@100trillionUSD/modeling-bitcoins-value-w... [1]: https://medium.com/@nic__carter/its-the-settlement-assurances-stupid-5dcd1c3f4e41 https://medium.com/@nic__carter/its-the-settlement-assurance... [2]: https://satoshi.nakamotoinstitute.org/posts/bitcointalk/327/ https://satoshi.nakamotoinstitute.org/posts/bitcointalk/327/