4 ms·

I’d be very surprised if Visa can clear transactions with less energy than Ethereum Proof of Stake requires. https://ethereum.org/en/energy-consumption/ https:

I’d be very surprised if Visa can clear transactions with less energy than Ethereum Proof of Stake requires. https://ethereum.org/en/energy-consumption/ https://ethereum.org/en/energy-consumption/ The energy argument is correct for proof of work, not proof of stake.

thisgoesnowhere 3y ago
Bit odd to see the comparison of ethereum to YouTube and netflix. What is that supposed to show?
- robcohen 3y ago
  It’s supposed to show relative power consumption. Which it does. What would you like to compare it to?
  - thisgoesnowhere 3y ago
    Other payment networks? Stripe, square, interac? One is a streaming service and the other is a payment network. It's like comparing a construction site to a restaurant in terms of fuel consumption.
  - crote 3y ago
    I could also compare it to the power consumption of the lightbulb in my garage. It is apples-to-oranges to compare a payment network to a video streaming platform, the only useful comparison is to another payment network.
kodah 3y ago
I don't think energy consumption was ever actually a virtue of the people who complained about it. It was just a means to an end for curbing crypto and it's grifts.
- jjtheblunt 3y ago
  Yeah it was: the energy cost (as in electric bill) to settle transactions is so large that all but huge transactions are money losers. Anyway, that was the concern. It contrasts with the fee set by percentage of transaction size used by Visa.
  - anonymouskimmer 3y ago
    And people in locations with big crypto farms can feel the hit in their own energy bills. Even people who never even heard of crypto will start to care when it hits their own pocketbook. https://www.nytimes.com/2023/04/09/business/bitcoin-mining-electricity-pollution.html https://www.nytimes.com/2023/04/09/business/bitcoin-mining-e... > In Texas, where 10 of the 34 mines are connected to the state’s grid, the increased demand has caused electric bills for power customers to rise nearly 5 percent, or $1.8 billion per year > The program pays miners, and other companies, for promising to stop using electricity upon request. In reality, they are rarely asked to shut down, but are still paid for making the pledge. > From midnight to nearly 4 p.m. on June 23, Riot earned more than $42,000 from the program while continuing to mine Bitcoin. (Overall in 2022, Riot made nearly $9.3 million from participating in the program nearly 85 percent of the time, the data shows, though the grid operator asked companies to actually lower their use for about 3.5 hours.) > Around that time, the company switched to the second technique: avoiding fees that Texas charges to maintain and strengthen the power grid. It did so by briefly shutting off almost completely. > To incentivize big customers to conserve electricity, those fees are based on how much electricity they use during several peak summer moments. Riot reduced its power use by more than 99 percent. > By 6:30 p.m., the company had resumed mining. If Riot had been fully operating all day, it would have incurred an estimated $5.5 million in fees — costs that are largely made up by other Texans. Over the course of the year, this saved Riot more than $27 million in potential fees. > One final mechanism lets some companies make extra money when electricity prices spike: They can stop mining and resell electricity to other customers. That earned Riot roughly $18 million last year.
- anonymouskimmer 3y ago
  It's why when I first read about crypto back in 2013 or so I thought about getting into Peercoin whenever I had the money to do so. And it's why years later, when I finally had the money to do so, I never bought any crypto, even when FOMO was hitting hard. It's why I stopped taking late night drives in my car to wind down, instead taking walks. And why I try to drive to maximize fuel efficiency. It's why I felt guilty every time I flew a plane while a student pilot and never finished a full pilot's license (I never would have started in the first place had I not been pushed to do so by my father). Yes, some of us care.
anonymouskimmer 3y ago
Everything else held equal it will always take more energy to have multiple computers ready to verify transactions, as well as the computational overhead of determining on the fly which computer actually will verify which transaction, than to have one computer verifying transactions. From your own link, PoS Ethereum is 1/30,000 the energy cost of PoW Ethereum. From this link (https://arxiv.org/pdf/2203.03717.pdf https://arxiv.org/pdf/2203.03717.pdf ), back in 2021 PoW Ethereum spent 125.36 kWh per transaction, while VISA spent 0.0015 kWh, an 83 thousand fold difference. Depending on how these numbers all shake out VISA is still superior to PoS Ethereum by about 2-3X per transaction.
- toomim 3y ago
  I think you're ignoring the energy spent on physical buildings, cars, and vehicles that power the banking/credit-card complex.
  - thereddevil 3y ago
    What does this even mean? Visa cards can be used for online purchases just like crypto. However crypto has a minuscule presence when it comes to physical stores. Why are you comparing apples to oranges.
  - anonymouskimmer 3y ago
    These costs exist for the blockchain networks, too.
- sharemywin 3y ago
  What's the average transaction fee for Ethereum versus visa?
  - anonymouskimmer 3y ago
    Currently it's at about 77 US cents per average transaction, with the fee paid by the sender. A month ago it was briefly above $2.70. VISA transactions, with the fee paid by the receiver, obviously pull more off the top in excess of electricity costs (for larger transactions), but they also have purchase protection, customer service, and things such as points built it for particular cards. So not a complete apples-to-apples comparison.
oblio 3y ago
There's no world in which a decentralized algorithm is more energy efficient than the equivalent centralized algoritm, all other factors being equal.
ndriscoll 3y ago
Everything I've read suggests the Visa network only processes ~5-10k transactions per second, which you could do on a single modest server. It is apparently built to handle up to 75k TPS[0]. Certainly it should be doable with less than 300 kW average. [0] https://usa.visa.com/dam/VCOM/global/about-visa/documents/visanet-factsheet.pdf https://usa.visa.com/dam/VCOM/global/about-visa/documents/vi...
- sowbug 3y ago
  Visa the company has over 20,000 employees. Presumably that network wouldn't function without them. Shouldn't the entire company's energy footprint be included in that 300kW?
  - anonymouskimmer 3y ago
    VISA provides more services than transaction verification. Should the exchanges be included in the Ethereum energy footprint?
    - junofan 3y ago
      No.
greiskul 3y ago
I would be very surprised if it didn't. Ethereum still has to deal with a hugely distribuited database updating all nodes, where VISA only has to update however many copies they keep, which will certainly be order of magnitudes less. Both of course would be much cheaper then Proof of Work systems.

thisgoesnowhere 3y ago

Bit odd to see the comparison of ethereum to YouTube and netflix. What is that supposed to show?

robcohen 3y ago
It’s supposed to show relative power consumption. Which it does. What would you like to compare it to?
- thisgoesnowhere 3y ago
  Other payment networks? Stripe, square, interac? One is a streaming service and the other is a payment network. It's like comparing a construction site to a restaurant in terms of fuel consumption.
- crote 3y ago
  I could also compare it to the power consumption of the lightbulb in my garage. It is apples-to-oranges to compare a payment network to a video streaming platform, the only useful comparison is to another payment network.

kodah 3y ago

I don't think energy consumption was ever actually a virtue of the people who complained about it. It was just a means to an end for curbing crypto and it's grifts.

jjtheblunt 3y ago
Yeah it was: the energy cost (as in electric bill) to settle transactions is so large that all but huge transactions are money losers. Anyway, that was the concern. It contrasts with the fee set by percentage of transaction size used by Visa.
- anonymouskimmer 3y ago
  And people in locations with big crypto farms can feel the hit in their own energy bills. Even people who never even heard of crypto will start to care when it hits their own pocketbook. https://www.nytimes.com/2023/04/09/business/bitcoin-mining-electricity-pollution.html https://www.nytimes.com/2023/04/09/business/bitcoin-mining-e... > In Texas, where 10 of the 34 mines are connected to the state’s grid, the increased demand has caused electric bills for power customers to rise nearly 5 percent, or $1.8 billion per year > The program pays miners, and other companies, for promising to stop using electricity upon request. In reality, they are rarely asked to shut down, but are still paid for making the pledge. > From midnight to nearly 4 p.m. on June 23, Riot earned more than $42,000 from the program while continuing to mine Bitcoin. (Overall in 2022, Riot made nearly $9.3 million from participating in the program nearly 85 percent of the time, the data shows, though the grid operator asked companies to actually lower their use for about 3.5 hours.) > Around that time, the company switched to the second technique: avoiding fees that Texas charges to maintain and strengthen the power grid. It did so by briefly shutting off almost completely. > To incentivize big customers to conserve electricity, those fees are based on how much electricity they use during several peak summer moments. Riot reduced its power use by more than 99 percent. > By 6:30 p.m., the company had resumed mining. If Riot had been fully operating all day, it would have incurred an estimated $5.5 million in fees — costs that are largely made up by other Texans. Over the course of the year, this saved Riot more than $27 million in potential fees. > One final mechanism lets some companies make extra money when electricity prices spike: They can stop mining and resell electricity to other customers. That earned Riot roughly $18 million last year.
anonymouskimmer 3y ago
It's why when I first read about crypto back in 2013 or so I thought about getting into Peercoin whenever I had the money to do so. And it's why years later, when I finally had the money to do so, I never bought any crypto, even when FOMO was hitting hard. It's why I stopped taking late night drives in my car to wind down, instead taking walks. And why I try to drive to maximize fuel efficiency. It's why I felt guilty every time I flew a plane while a student pilot and never finished a full pilot's license (I never would have started in the first place had I not been pushed to do so by my father). Yes, some of us care.

anonymouskimmer 3y ago

Everything else held equal it will always take more energy to have multiple computers ready to verify transactions, as well as the computational overhead of determining on the fly which computer actually will verify which transaction, than to have one computer verifying transactions. From your own link, PoS Ethereum is 1/30,000 the energy cost of PoW Ethereum. From this link (https://arxiv.org/pdf/2203.03717.pdf https://arxiv.org/pdf/2203.03717.pdf ), back in 2021 PoW Ethereum spent 125.36 kWh per transaction, while VISA spent 0.0015 kWh, an 83 thousand fold difference. Depending on how these numbers all shake out VISA is still superior to PoS Ethereum by about 2-3X per transaction.

toomim 3y ago
I think you're ignoring the energy spent on physical buildings, cars, and vehicles that power the banking/credit-card complex.
- thereddevil 3y ago
  What does this even mean? Visa cards can be used for online purchases just like crypto. However crypto has a minuscule presence when it comes to physical stores. Why are you comparing apples to oranges.
- anonymouskimmer 3y ago
  These costs exist for the blockchain networks, too.
sharemywin 3y ago
What's the average transaction fee for Ethereum versus visa?
- anonymouskimmer 3y ago
  Currently it's at about 77 US cents per average transaction, with the fee paid by the sender. A month ago it was briefly above $2.70. VISA transactions, with the fee paid by the receiver, obviously pull more off the top in excess of electricity costs (for larger transactions), but they also have purchase protection, customer service, and things such as points built it for particular cards. So not a complete apples-to-apples comparison.

oblio 3y ago

There's no world in which a decentralized algorithm is more energy efficient than the equivalent centralized algoritm, all other factors being equal.

ndriscoll 3y ago

Everything I've read suggests the Visa network only processes ~5-10k transactions per second, which you could do on a single modest server. It is apparently built to handle up to 75k TPS[0]. Certainly it should be doable with less than 300 kW average. [0] https://usa.visa.com/dam/VCOM/global/about-visa/documents/visanet-factsheet.pdf https://usa.visa.com/dam/VCOM/global/about-visa/documents/vi...

sowbug 3y ago
Visa the company has over 20,000 employees. Presumably that network wouldn't function without them. Shouldn't the entire company's energy footprint be included in that 300kW?
- anonymouskimmer 3y ago
  VISA provides more services than transaction verification. Should the exchanges be included in the Ethereum energy footprint?
  - junofan 3y ago
    No.

greiskul 3y ago

I would be very surprised if it didn't. Ethereum still has to deal with a hugely distribuited database updating all nodes, where VISA only has to update however many copies they keep, which will certainly be order of magnitudes less. Both of course would be much cheaper then Proof of Work systems.