5 ms·

Your position seems to be that CO2 isn't linked to computing specific transactions, it's just the amount of mining on the network at any given time. But that co

Your position seems to be that CO2 isn't linked to computing specific transactions, it's just the amount of mining on the network at any given time. But that could still very easily be expressed in terms of CO2. I.e. "during the time these transactions were monopolizing the network, the carbon equivalent of n acres of rainforest were burnt". Proof of stake doesn't actually eliminate the carbon impact of these systems, it just improves the efficiency. There are still many computers running for the express purpose of buying autographed drawings of monkeys.

rglullis 5y ago
It improves the efficiency 10000-fold. You can run a Ethereum validator node on a 10W celeron, underclock it, and still be fine. There are ~380k validators on the beacon chain now. There are 20M PS5 + Xbox sold already. A video game console consumes as much as 300W during usage, and even on idle can consume as much as a validator.
- kennywinker 5y ago
  Seems like someone should go update the wikipedia article then https://en.wikipedia.org/wiki/Proof_of_stake https://en.wikipedia.org/wiki/Proof_of_stake > In 2021 a study by the University of London found that in general the energy consumption of the Proof-of Work based Bitcoin was about 1,000 times higher than that of the highest consuming proof of stake system that was studied 1,000x is not 10,000x > They also couldn't find the energy consumption of a proof of stake system on a large scale as such a system does not exist at the time of the report.[16][17] Big systems have big side effects. As others have said, I'll believe it when I see it.
  - rglullis 5y ago
    You can see it already [0]. Eth PoS network already exists, what is missing is for the transition to be completed (aka, the "Merge") [0]: https://beaconscan.com/statistics https://beaconscan.com/statistics
    - kennywinker 5y ago
      The 380k validators concerns me a little, that's a lot of computing power burnt making sure we know who owns the monkey jpg. But I don't expect you would see the true reality of unexpected side effects until a system like this goes live and there is $$ to be made.
      - rglullis 5y ago
        That's not just a "monkey jpg", and your comment would be a lot more valid if you dropped the sarcasm. Anyway: 380k validators at 10W and assuming the cost of the kwH to be $0.25 (which is more than the world's average cost) amounts to ~8 million USD /year. That seems to me more than acceptable if it means that we are securing a blockchain that is already holding more than ~500B in assets.
        kennywinker 5y ago
        I can not, and I will not, drop the sarcasm. If blockchain is the future, it deserves to be mocked every bit as much as the bits of paper it replaces, probably more. If it's not the future, then it's even funnier (in a sickening way) that humans decided to create 500B worth of monkey jpgs while the climate imploded around them. Also, for the record, your back-of-the-envelope estimate - 32,000,000 kwh assuming I reversed it correctly - corresponds to about 27,200,000 lbs of carbon @ 0.85lbs/kwh - the average for all US electricity production in 2020. Convert from lbs to 12337.7125 metric tons, divided by the 141 mT of carbon stored per acre of rainforest. Which makes the answer: 87.5 acres of rainforest burnt per year by PoS, before full adoption.
        rglullis 5y ago
        Adoption of the network is not related to its electricity consumption. The number of validators is a function of how many people want to invest (capital and resources) to secure the network. And forgive me if I don't act all-too-shocked by your numbers, judging from all your posts about iOS and Apple stuff, I can easily see that your talk is cheap. Come back when your consumer habits actually match your concern about the environment.
joshcryer 5y ago
The entire ETH VM is running at about 0.8 MIPS right now. Yes. That is not an exaggeration, current gas limit is 30 million / 3 gas for simpliest add instruction / 12 seconds per block. The calculation power of the network is absolutely inconsequential to the operation of the network. 99.999% of all the energy used is just for Proof of Work. Once they move to Proof of Stake, if they even do, most of the power usage will be to run the hard drives to hold the increasingly massive ledger that will spawn from it.
- quickthrower2 5y ago
  For humour sake, if I had time i would set up a raspberry pi to run psql and node and provide an alternative centralised blockchain that would compete computationally.
  - rglullis 5y ago
    If you manage to convince all the millions of people that the data on your raspberry pi is valid and you pinky-swear that you are not going to manipulate their funds, go for it.
    - quickthrower2 5y ago
      Works for binance and many other centralized chains!
- kennywinker 5y ago
  The issue isn't how much computing power (aka carbon) does the network NEED. The issue is how much computing power the network consumes. Depending on the implementation of PoS, there could be all kinds of incentives or pressures for a lot of distributed computing power to get thrown at it, despite it only needing a fraction of that. None of that is a problem if the system is #1 displacing other higher consuming systems and #2 not actively harming people. Re: #1 So far most of what I've seen out of blockchain seems additive - it's not replacing old systems, it's layering on new systems. But it's still early days, maybe that'll change. As for #2, well, it's hard to trace the exact benefits and harms of blockchain we've seen so far. Which leaves me to fall back on history. You can call me a commie if you want to, but a system that benefits capital over labour usually shakes out to benefit a few while harming many ¯\_(ツ)_/¯
aeternum 5y ago
Still nothing compared to the amount of CO2 and environmental destruction caused by tourists competing for likes on their social media posts.

rglullis 5y ago

It improves the efficiency 10000-fold. You can run a Ethereum validator node on a 10W celeron, underclock it, and still be fine. There are ~380k validators on the beacon chain now. There are 20M PS5 + Xbox sold already. A video game console consumes as much as 300W during usage, and even on idle can consume as much as a validator.

kennywinker 5y ago
Seems like someone should go update the wikipedia article then https://en.wikipedia.org/wiki/Proof_of_stake https://en.wikipedia.org/wiki/Proof_of_stake > In 2021 a study by the University of London found that in general the energy consumption of the Proof-of Work based Bitcoin was about 1,000 times higher than that of the highest consuming proof of stake system that was studied 1,000x is not 10,000x > They also couldn't find the energy consumption of a proof of stake system on a large scale as such a system does not exist at the time of the report.[16][17] Big systems have big side effects. As others have said, I'll believe it when I see it.
- rglullis 5y ago
  You can see it already [0]. Eth PoS network already exists, what is missing is for the transition to be completed (aka, the "Merge") [0]: https://beaconscan.com/statistics https://beaconscan.com/statistics
  - kennywinker 5y ago
    The 380k validators concerns me a little, that's a lot of computing power burnt making sure we know who owns the monkey jpg. But I don't expect you would see the true reality of unexpected side effects until a system like this goes live and there is $$ to be made.
    - rglullis 5y ago
      That's not just a "monkey jpg", and your comment would be a lot more valid if you dropped the sarcasm. Anyway: 380k validators at 10W and assuming the cost of the kwH to be $0.25 (which is more than the world's average cost) amounts to ~8 million USD /year. That seems to me more than acceptable if it means that we are securing a blockchain that is already holding more than ~500B in assets.
      - kennywinker 5y ago
        I can not, and I will not, drop the sarcasm. If blockchain is the future, it deserves to be mocked every bit as much as the bits of paper it replaces, probably more. If it's not the future, then it's even funnier (in a sickening way) that humans decided to create 500B worth of monkey jpgs while the climate imploded around them. Also, for the record, your back-of-the-envelope estimate - 32,000,000 kwh assuming I reversed it correctly - corresponds to about 27,200,000 lbs of carbon @ 0.85lbs/kwh - the average for all US electricity production in 2020. Convert from lbs to 12337.7125 metric tons, divided by the 141 mT of carbon stored per acre of rainforest. Which makes the answer: 87.5 acres of rainforest burnt per year by PoS, before full adoption.
        rglullis 5y ago
        Adoption of the network is not related to its electricity consumption. The number of validators is a function of how many people want to invest (capital and resources) to secure the network. And forgive me if I don't act all-too-shocked by your numbers, judging from all your posts about iOS and Apple stuff, I can easily see that your talk is cheap. Come back when your consumer habits actually match your concern about the environment.

joshcryer 5y ago

The entire ETH VM is running at about 0.8 MIPS right now. Yes. That is not an exaggeration, current gas limit is 30 million / 3 gas for simpliest add instruction / 12 seconds per block. The calculation power of the network is absolutely inconsequential to the operation of the network. 99.999% of all the energy used is just for Proof of Work. Once they move to Proof of Stake, if they even do, most of the power usage will be to run the hard drives to hold the increasingly massive ledger that will spawn from it.

quickthrower2 5y ago
For humour sake, if I had time i would set up a raspberry pi to run psql and node and provide an alternative centralised blockchain that would compete computationally.
- rglullis 5y ago
  If you manage to convince all the millions of people that the data on your raspberry pi is valid and you pinky-swear that you are not going to manipulate their funds, go for it.
  - quickthrower2 5y ago
    Works for binance and many other centralized chains!
kennywinker 5y ago
The issue isn't how much computing power (aka carbon) does the network NEED. The issue is how much computing power the network consumes. Depending on the implementation of PoS, there could be all kinds of incentives or pressures for a lot of distributed computing power to get thrown at it, despite it only needing a fraction of that. None of that is a problem if the system is #1 displacing other higher consuming systems and #2 not actively harming people. Re: #1 So far most of what I've seen out of blockchain seems additive - it's not replacing old systems, it's layering on new systems. But it's still early days, maybe that'll change. As for #2, well, it's hard to trace the exact benefits and harms of blockchain we've seen so far. Which leaves me to fall back on history. You can call me a commie if you want to, but a system that benefits capital over labour usually shakes out to benefit a few while harming many ¯\_(ツ)_/¯

aeternum 5y ago

Still nothing compared to the amount of CO2 and environmental destruction caused by tourists competing for likes on their social media posts.