> It's like Finland has suddenly shut off its power grid
So, have we observed global energy usage go down by about one Finland?
Shouldn't that be observable somehow? Shouldn't there be some power stations reducing their output as a reaction to reduced demand?
Anyone know how this would be visible, and on what kind of time frame we expect it to become visible?
I'm not claiming it hasn't happened. I just feel surprised to not see more coverage of that in this article, nor here in the comments. Energy efficiency is largely the point of this major change. Shouldn't there be graphs of the power grids everywhere showing a big drop? Maybe my expectations are just off on that.
The entire world generates about 25,000 TWh of electricity every year.
Finland consumes about 87 TWh of electricity every year.
An entire Finland of electricity use thinly distributed over the entire planet disappearing is a negligible rounding error on the grand scheme of things. It's about a ~0.3% change.
News media uses word imagery like "an entire Finland of electricity use" because it sounds huge and scary to the average person who doesn't understand that absolute numbers are meaningless out of context. Zooming out, you realize that while Finland is big from an individual human scale, it barely exists in the wider cacophony of human civilization.
And that's just looking at electricity, which is a percentage total human energy use. Much more energy use comes from transportation, industry, and heating, which is much more carbon intensive than electricity generation, since much of our electricity comes from hydro, nuclear, and increasingly solar and wind.
The thing is, in the wider cacophony of human civilization, Ethereum and cryptocurrencies also don't exist.
If I and my grandma and everyone's grandma would use Proof of Work cryptocurrencies to buy peanuts at the supermarket, PoW energy usage would probably rival that of China.
Now it's just used for speculation by a bunch of rich folks, crooks and marks. Probably only a few thousand transactions per second, I imagine.
> If I and my grandma and everyone's grandma would use Proof of Work cryptocurrencies to buy peanuts at the supermarket, PoW energy usage would probably rival that of China.
I can't speak for all PoW cryptocurrencies, but this is a common misunderstanding about Bitcoin. Energy usage does not scale with the number of users. The transactions per second is largely fixed due to the block size limit. If your grandma is using Bitcoin, she will probably be transacting on a second layer, e.g. Lightning or Coinbase.
No, but you don't have to. The point of blockchains is to allow people to opt out when they want to. If you don't want to or need to, you have that option.
It's actually a worse situation that bitcoin synergy usage scales with its price.
So same number of people could be using bitcoin but if the price jumps up by 10X, then over a period a few months it's energy usage would also go up by 10X.
This is exactly what you’d expect to see with market manipulation. It’s the pump before the dump. Or less cynically, it’s just herd mentality seen through data.
Using the words "centralized" or "decentralized" when describing crypto (and probably many other tech things) is perhaps frequently the worst oversimplifications one can engage in.
I think what you want to discuss when using these words is the extent to which it is either likely or possible that a small number of players can unduly influence a thing, but there are frequently MANY MORE FACTORS at work here than "that number." You have to look at how the thing is set up, what governance looks like, what technical limitations there are, and -- perhaps most importantly -- what incentives are there in place for the players to do so.
A lot of the "centralization" fears are bad because they're the sort of thing, that -- if executed -- would destroy the value of the thing to the centralizers themselves, and thus they would literally never do it.
None of it is truly decentralized. Most Bitcoin mining comes from three corporations. You could replicate the whole network by setting up a cross-cloud system that's balanced between AWS, Azure, and GCP.
Wealth always concentrates to few entities in all forms of free market capitalism.
You cannot have decentralized money without distributed wealth that is not concentrated . This usually means a successful form of socialism ( none have proven successful, or looks likely to do so today)
This merge to PoS is just exposing that people with wealth always have a say on how the system works, whether it is miners or stakers it was and is always controlled by few .
> This merge to PoS is just exposing that people with wealth always have a say on how the system works, whether it is miners or stakers it was and is always controlled by few .
Miners nor stakers can decide how the system works, Ethereum does not have on-chain governance. 51% of miners/stakers can't just post invalid blocks crediting them with free money nor can they force anyone else's wallet to do anything that wasn't signed by the private key. The worst you can really do is refuse to include transactions in your blocks for as long as you are controlling the network, all it takes is one honest actor to include your transaction and it will finalize.
Couple of hundred years?[2] Many models look sustainable[1] in the short term - few decades is short for any economic model to show its weaknesses. The soviet model did survive for 5-6 decades years quite well after all.
[1] Sustainable does not mean good or that I (or any else) condone it, just that it can work and survive .
[2] That is the number to make it comparable to modern American Econo-political system that has survived largely intact in pretty much same form from the early years of the revolution. Historically feudal monarchies or other systems like the Tokugawa Shogunate had survived much longer for variety of reasons however that are difficult to compare in the modern techno Geo-political era .
They're brown and 'marginalized' and occupy a small slice of Mexico. You and I may know about them because we probably monitor anarchist/leftist circles. I'm quite anti-socialist myself but I readily admit the Zapitistas have proven 'successful' about as well as they could hope for their situation. But few outside of a few circles wants to acknowledge it as it goes against the conventional thought that democratic or pseudodemocratic society can be trumped by a bunch of fucking 'backwater' community-focused brown people with guns and their own prerogative, who don't fit the mold of the typical example that when society 'breaks down' it must be taken over by savage worlords or drug barons.
You can guarantee if there were a story of a bunch of American rednecks even trying to pull off anything similar it'd be downvoted to hell here and nearly universally hated by American society. Soccer moms everywhere would clap to their execution by the National Guard. Some ATF-esque character would stand over the rubble for photographs like they did over the char of the dead kids at Waco and the public would eat it up.
Admitting the Zapatistas are a 'success' challenges a lot of people's philosophies and assumptions. Hell it even admits some of my own were wrong. This can be difficult to reconcile. It's not surprising that answers to your question are not exactly forthcoming.
But that's kind of the problem, isn't it? It may be obscure, but its success does disprove a very non-obscure claim.
(Side note: my definition of "success" for socialism is fairly modest: to be sustainable long term economically, and to preserve personal freedoms at least to the same degree that modern capitalist liberal democracies have. That's the only way I see it as a meaningful alternative to the status quo.)
There's also Rojava to consider, and it's perhaps somewhat more widely known because it was covered somewhat during the Western press' fascination with ISIS and war with them. These guys are perhaps more interesting than the Zapatistas long term because they're doing it in a somewhat more urbanized & industrialized area with 10x population, and they are explicitly designing a governance framework with intent to scale to nation-state size and beyond. But, well, it hasn't even been 10 years yet, so I don't think it counts as solid evidence of anything just yet.
They talk a big talk about socialism but honestly I wasn't seeing it in the cities like Qamishlo. Full of capitalist street vendors and the like, and poverty on the outskirts. Honestly they seemed pretty vanilla capitalist outside the 'mustache-jesus' (Apo) talk you see in YPG, etc. I do recall free bread/naan line on the street.
In the firmly 'Rojava' 'owned' cities like Deyrik the general structure of the people and markets reminded me quite a bit of Barzani controlled KRG. Of course, I spent most my time with the militia and not civil life, so I'm sure there's lots I missed. I think many of the 'true believers' in the old Kurdish socialist causes probably migrated more to the Iraqi Mountains, sadly those people are getting pounded on by the Turks.
From their agitprop, my impression was that by socialism they mostly mean minimizing economic exploitation, so e.g. street vendors, craftsmen etc who aren't wage workers aren't really incompatible with that approach. I'd be more curious to see the statistics on larger businesses - how many have a hired labor force, and how many are co-ops.
Also, the most recent iteration of the social contract that I have - one from 2016 - has these bits in it:
"There shall be a right to invest in private projects which, while taking account of environmental balance, provide the services necessary for economic development, respond to the needs of society and help to stimulate economic activity in the community.
The right to private property shall be safeguarded in such a way as not to conflict with the public interest, and it shall be regulated by law."
Now this is all rather vague, but it doesn't read like a hardline stance against any hint of capitalism. And given the bottom-up political system, so long as democratic institutions work as declared, it seems that individual municipalities should be able dial the balance as they see fit. From what I heard, while PYD enjoys broad popular support, that doesn't necessarily translate to specific policies they advocate for. So perhaps it's basically as socialist as people actually want it to be at the moment?
Which would make it even more important. While the economics of it is interesting by itself, I think the key part of the experiment is decentralized council democracy. That should be enough to give people actual voice on matters that concern them, as opposed to the sham that representative republics inevitably end up at scale. Once communities actually have control over their commons, they can figure out the economics according to their needs and preferences.
But, again, this is assuming that their governance model is as advertised and can persist in the long term. What was your impression on the ground?
Sorry for the abbreviated response I'm going to give here.
I didn't spend a lot of time in civil life. I saw no large businesses. I saw these in KRG Iraq (even big dealerships, like for farm equipment) but not in Rojava. Basically all the businesses I noticed in urban areas were what I describe as roughly family-unit sized operations run in your typical middle eastern stall. Perhaps a hardware store or something was a bit larger. The biggest 'business' I saw was vehicle repair shop with 4 or 5 stations and that was run by YPG. In civil city such a repair shop would basically be a single booth with the storage-shed style big room and they probably worked on the car while it was outside. I never saw something large-businesses esque like the kind of supermarkets you may see deep in latin America. I also did not notice much presence of 'government operated' retail business and if that was happening for anything other than bread, it was invisible to me (PYD/YPG does run some co-ops, like agriculture and textiles related stuff).
I have no idea how the farms were run. It's possible these were run by the state. The farmers are very dedicated, they gave no fuck and just kept going while active shelling/fighting was happening proximally. It is possible as I mentioned above to get bread for free in the cities. You just walk up and ask for it (of course, you could ask practically anyone for food anywhere, they would probably give you some). No money was seen changing hands, but that is at specific stalls. I do think whatever their food distribution system is, it seemed to be working better than in ISIL controlled territory -- when we took over one village people were begging in a desperate way and fighting over the little food our people were able to give. I never saw such desperation in Rojava proper (lots of beggars in Rojava, but for money not food).
Overall I did not find many devout followers of some particular brand of political writings, although I'm sure some politicians are, but I'm not much a politician myself :) A few 'Westerners' joined the Kurds with extreme political views, but anecdotally I found those people were not very well liked. I have my doubts whether the average people even give a fuck about whatever the PYD or people in charge think. The Kurds have a long history and most of that doesn't include Apo's (Ocalan's) brand of socialism, although they would surely view him highly charitably. There may be significantly different opinions in (the tip of) Northern Iraq where the hardcore socialist elements hold more sway. Then again, given the temperament of the Kurdish people I think it would be hard to impose any sort of anything on them.
I would wager as long as you don't perform some egregious crime you could get away with a lot in Rojava without the government imposing their will on you. I visited a prison and some petty criminals escaped as I was there, the reaction of the guards was just kind of 'meh' as they got away. I do not think the government has the strength to fight their existential threats while simultaneously enforcing any strong form of socialism on the populace, although doing so on a large business would be easy enough as they'd be a big centralized target.
The main attractions I would describe to Rojava are pretty extreme freedom (minimal government, you can have guns, you seem to be able to run a business pretty openly without much fanfare, by middle east standards extremely liberal view on religion, booze is sold openly, people are open and friendly and helpful, personal rights like privacy and ownership appear to be well respected, etc). If I were looking for a socialist utopia, or anything near approximating it, I would not go there.
If I could describe Kurds, I'd call them Americans of the middle east. Instead of America's de facto religion of Christianity, their 'religion' is Apo (Ocalan, the socialist). But like America, the average person seems to be not much a practitioner and Apo is more of a symbol like the cross so many people hang as much for cultural reasons as anything else. Nevertheless their culture is incredibly hospitable, so perhaps their brand of socialism can better described as just being a good neighbor and not the government putting a foot up your ass to redistribute your wealth.
I can pretty confidently say that all of Visa+Mastercard is way more than a few thousand transactions per second, I'm familiar with several companies that push hundreds of transactions per second through Visa+Mastercard and there's no way they're a significant portion of their business.
Visa says they process "150 million transactions every day in 175 currencies" (see page 3 at https://usa.visa.com/dam/VCOM/download/corporate/media/visan...). That's ~1,800 per second. Mastercard is smaller, so this would be the upper limit for them. Both combined should still fit into "a few thousand transactions per second".
True. The latest financial report from Visa [1] says 164.7B transactions in 2021, or ~5,000 per second. This number is 3x larger! Mastercard is slightly smaller, but comparable at 140B [2].
Imagine we can achieve that throughput with a single server without breaking a sweat![1]. The number of economic transactions all humans engage everyday including cash is perhaps 100x of that: so just in order of 500,000 TPS or less that feels quite small to be honest.
[1] Yes these systems are complex and very distributed and have lot of checks and balances and the actual transactions apps and DBs are running on infra in hundreds or thousands of servers in DCs all around the world.
On the busiest day of the year at peak hour here in the Netherlands the debit card transactions (creditcards are uncommon) reach just over 700/s. [Search for 'pin transacties per seconde' for news items covering this]
Comparing visa and Mastercard transactions with Bitcoin is something that doesn’t make a lot of sense. Do people really think visa settles transactions between two different bank accounts? Those card transactions can take days to settle.
Slow settlement and reversible transactions are a boon for buyers (and for fraudsters abusing chargebacks to get free stuff). For sellers they're a bit of a nightmare since you never know when money you thought you had been payed in good faith might disappear from your accounts with little or no recourse.
Zkrollups on Ethereum can do a couple thousand tx/sec, without security compromises. The plan now is to use the base layer mainly to support rollups, and use data sharding on the base layer to multiply the capacity of rollups. That should get it to about 100,000 tx/sec. That will be a pretty big change, but not as big as what they just did, and they've already got the design mostly worked out.
I doubt that for credit cards. Carrefour or pick your favorite supermarket chain alone probably generate hundreds of transactions per second worldwide during peak periods, and there are tens of major supermarket chains you've never heard of. Add regular grocery stores, cinemas, etc, I'd be really interested in an order of magnitude, but at peak times it has to be in the tens of thousands if not low hundreds of thousands.
For crypto I was trying to be super generous, I know they're incredibly slow.
in PoW, energy consumption scales linearly with the block reward value which is denominated in bitcoin thus scales with value of btc. Reward for mining is [ ( your hashrate / network hashrate) * block reward ].
so it's not usage that scales energy consumption, it's value of btc. Arguably though the two are intrinsically related (this should be obvious - if bitcoin somehow becomes the defacto global currency, a shitload of investment will pour in and bring its value much much higher)
in truth it's a moot point. even with lightning protocol, bitcoin as a protocol couldn't handle that much capacity. If a few hundred million people just wanted to publish a single transaction (putting money in a wallet and thereafter doing everything on lightning protocol), it would take literally years to process that many transactions.
Edit: just to be explicit, the reason energy consumption scales with block reward value is because it becomes very profitable to buy more mining equipment and mine more if the reward say doubles. It’s not an instant effect of course because there’s capital acquisition and operational capacity involved
Misinformation, both from proponents and opponents is the name of the game in crypto.
It's incredibly difficult to find the truth from either side, especially if you're a lay person who doesn't have any grasp of how the technology works.
I can tell you as an opponent of cryptocurrencies.
Experiments on social systems, currencies and stores of value should start small and work their way up OVER DECADES to tens and hundreds of thousands of people.
Experimental stuff put into production, like Communism, can bring suffering for decades, if not centuries, to tens and hundreds of millions of people... billions, even.
So if the proponents want to revolutionize the system immediately for a 5% efficiency improvement, especially since many of them have so MUCH MONEY AND POWER to gain from this revolution, be very, very SKEPTICAL.
Frankly their odds to not be grifters or (smart) idiots are minuscule.
> If I and my grandma and everyone's grandma would use Proof of Work cryptocurrencies to buy peanuts at the supermarket
Beyond your misconception that work scales with users rather than value, it's also a misconception that bitcoin is competing with technologies like Visa and Paypal. It's not supposed to be an alternative payment system.
It's competing with gold and sovereign currencies.
> If I and my grandma and everyone's grandma would use Proof of Work cryptocurrencies to buy peanuts at the supermarket, PoW energy usage would probably rival that of China.
You don't understand how any of this works, do you?
> Good try, but trite criticism regarding crypto is required to include the words "Ponzi" and "tulips"
Thank you, I'm trying to come up with an "anti bullshit bingo" since the bullshit bingo from cryptocurrencies has already managed to raise tens of billions of dollars. Glad you like it.
> Quite an imagination! You're very far off, do you have any experience in this space?
I was trying to be generous since crypto supporters always like to point to the newest barely working bleeding edge centralized "layer" called lightning or thunderbolt or some other electricity derived thing, which is supposed to greatly accelerate the glacial rate of crypto transactions.
Again, can I use crypto of any kind to buy $2 peanuts at the supermarket in Bucharest and grandma $1 popcorn at the cinema in Djibouti, without turning Earth into Venus?
> Again, can I use crypto of any kind to buy $2 peanuts at the supermarket in Bucharest and grandma $1 popcorn at the cinema in Djibouti, without turning Earth into Venus?
Probably not, but you can neither use USD (or any sort of dollar) for that either. In Bucharest you'd use Romanian leu (RON) and in Djibouti you'd use Djiboutian franc (DJF).
It's all about finding people in the middle, to agree on what you both have. In this case, you wouldn't be able to buy anything in those locations.
Like my normal dollar based credit cards that allow foreign transactions and have no foreign transactions fees...thr processor converts thr currency to dollars for me and puts it on my statement.
0.3% is a rounding error if you look at the electricity generation of a region, but it's huge if you look at the impact it will have over many years. That's 0.3% less of the global energy supply that needs to be replaced with nuclear or renewables+storage. It's like getting ten free nuclear plants.
>News media uses word imagery like "an entire Finland of electricity use" because it sounds huge and scary to the average person who doesn't understand that absolute numbers are meaningless out of context.
With the media tactics out of the picture, you don't think this is a huge amount of electricity for crypto alone to have been using? Seriously? 0.3% of global electricity use is ABSOLUTELY HUGE.
> The entire world generates about 25,000 TWh of electricity every year.
I know this isn't you, many people use these units, but I cringe every time I see abuse of units like this. 1 TWh/year is 113MW, so you are saying 2.8TW global; Finland is 9.4GW.
No, because of two reasons. First, many miners simply pointed their hardware to mine other cryptocurrencies such as ETHW or ETC. For example we have evidence that about a quarter of Ethereum's mining farms moved to ETC over the last 24h. Second, contrary to sensationalist headlines Ethereum miners only represent a drop in the bucket of the global electricity consumption: only 0.1%. Yes that can be a "country's worth of electricity" but in relative terms, 0.1% would be barely visible on charts you might examine.
Also, digiconomist, an often quoted source of Ethereum miner's energy consumption statistics, was grossly overestimating the figures. The actual consumption was probably around 20-30 TWh/year instead of the ~80 TWh/year figure they estimated. Just look at their chart: it made no sense, for example between Sep 2020 and May 2022 Ethereum hashrate grew 5-fold from 200 to 1000 TH/s, whereas in that same time-frame digiconomist estimated the consumption grew 15-fold from 6 TWh/year to 90 TWh/year. If anything, hardware has become (a bit) more efficient over time, it didn't become 3 times less efficient...
But that's not too surprising, given the author of digiconomist has a history of exaggerating his figures, like he did for Bitcoin see https://blog.zorinaq.com/serious-faults-in-beci/ But nowadays his Bitcoin estimate is more in-line with more reputable estimates such as Cambridge's https://cbeci.org/ Last time I looked he was within ±30%
> Ethereum miners only represent a drop in the bucket of the global electricity consumption: only 0.1%
1 out of every 1000 watts of energy produced in the whole world going to ethereum mining is tragic. It’s an absolutely massive waste, and the fact that crypto miners think a number like that isn’t a big deal is horrifying to me.
You care about 1 in 1000 ? Did you know that 660 out of 1000 watts of primary energy used to create electricity are wasted by the time the electricity arrives at the customer meter? (source: https://www.enerdynamics.com/Energy-Currents_Blog/How-Much-P... )
Logically, you should be 660 times more "horrified" by this waste than by cryto mining, and yet you are not. In fact most likely until today you were not even aware of such waste. This is what bothers me the most about these discussions. People have priorities in the wrong places. You want to show you care about waste? Then start caring about the most important problems instead of being laser-focusing on 0.1%.
Your 66% number includes generation losses. You're literally arguing that we should be horrified that solar panels don't produce 1 watt of electricity for every watt of light which hits them.
Would it be great if solar panels and wind turbines and nuclear reactors converted 100% of the heat/light/wind energy into useful electricity? Sure. But that's also not something we have the technology to achieve, and likely never will. People are constantly researching ways to improve power plants, but they will never reach 100% efficiency.
What we _do_ have the technology to achieve, however, is turning off things which use a ton of energy but provide no value. Saving 1‰ of earth's global electricity consumption just by turning off a useless gambling toy is a huge win, and the fact that it was allowed to use 1‰ of our global energy use in the first place is horrifying.
The transmission losses are something inherent in the system that can’t be avoided, and if someone could solve that problem they’d help a great deal with greenhouse emissions and I’d certainly applaud that.
But this 1 in 1000 watts going to ethereum mining is entirely useless, contributing nothing to society except a Ponzi scheme that, like most Silicon Valley bullshit, only exists because regulations haven’t caught up yet. I don’t care that 1 in 1000 watts is a small number. The wattage going to Finland actually gives tens of millions of people electricity to live their lives. Crypto mining is just tech bros playing with funny money, robbing people of real money in the process.
«are something inherent in the system that can’t be avoided»
Firstly they aren't inherent. Losses can be minimized. But you don't seem to care, even though losses account for orders of magnitude more energy waste than mining. Your position is illogical.
Secondly, this isn't only about transmission losses. The link explains, for example, that some of this waste comes from incandescent light bulbs that waste 90% of electricity as heat. Again, logically you should be "horrified" by this, but you aren't because media doesn't write click-baity articles about incandescent light bulbs "boiling oceans".
> Firstly they aren't inherent. Losses can be minimized
“Can be minimized” and “Inherent” are not antonyms. It can be both. And yes, I care.
> some of this waste comes from incandescent light bulbs that waste 90% of electricity as heat. Again, logically you should be "horrified" by this
I am. Every reasonable human being would look at incandescent light bulbs and say “yes, those are wasteful”. Because a reasonable alternative exists, and if you’re still burning incandescent light bulbs, you’re being wasteful.
PoW cryptocurrency however, is a system that is designed to be wasteful, and if the crypto bros had their way, we’d do all transactions this way. When an obvious alternative exists: Don’t fucking use cryptocurrency. Crypto solves zero problems, has zero benefits, and wastes electricity. The faster we abandon it, the better for literally everyone.
> First, many miners simply pointed their hardware to mine other cryptocurrencies such as ETHW or ETC. For example we have evidence that about a quarter of Ethereum's mining farms moved to ETC over the last 24h.
We'll see if this is maintainable.
Miners can't mine if the reward doesn't cover their costs. So, the only way the mining remains sustainable is if these tokens rise drastically in value.
It is possible that many miners are huge holders of ETH, in which case we may see them massively dump ETH and buy ETC to try to invert the price and keep their business going.
> For example we have evidence that about a quarter of Ethereum's mining farms moved to ETC over the last 24h.
...which is a stop-gap, holding-pattern sort of action to take, because there is OpEx to running a mining farm, and the lower trading value of ETC vs ETH (1 ETC = 0.025ETH) means ETC block rewards likely won't be enough to be positive-margin for these farmers. This is just a way to reduce burn while they try to sell their rigs. (Which they likely won't be able to do, because nobody wants overheated near-EOL mining GPUs. Mining farms will likely just end up bankrupt with assets liquidated at fire-sale prices.)
Power demand varies wildly during the day (Like +/-50%, maybe more depending on climate). A 1% change will get lost in the noise of "Oh, it's 3 degrees warmer today... HVAC working harder"
Ethereum Classic’s hash rate has gone up by about 25% of Ethereum’s hashrate, so at least for now it looks like a lot of the energy use is just moving as miners point their GPU rigs at alt coins. Very curious to see if ETHW, ie Ethereum without the merge, maintains a significant amount of hashing power. Another thing to watch will be if those alt coins are profitable to keep mining or if miners will start selling their rigs.
Aha, thank you! I felt like this change should be visible in some kind of graph somewhere, and you're right, the ETC hashrate has roughly quadrupled in the span of two days:
There will probably be some really interesting network effects with this. Since a lot of the other PoW coins that are ETH-hardware compatiable have low volume, I think we're going to very rapidly see the profitably of mining these go way down (to the point of going negative in some cases) as all these extra miners suddenly start mining these coins, but this takes time. So in other words, I would expect the real global energy usage reduction to happen weeks or months after the merge since it will take a while for all the altcoins to get overmined.
So, with all the new hashers coming from ETH, ETHW still has an order of magnitude less activity than ETH had before the change. Evidently it didn't absorb all the hardware.
The question if it is visible in the electricity generation numbers is still very relevant. (If it is, we will probably only be able to see it in an year, when international organizations compile their numbers.)
Whoever was using this electronics switched to other BTC variants, but in long term this reduced profitability and should harm people using energy in this way.
But sadly no immediate impact, unless there are electronics that could be profitably consuming power for Ethereum and it is not profitable for alternatives.
In some sense that's true, but missing the point. The amount of energy worth buying to mine crypto is exactly equal to the value of the crypto mined. What we should expect to see[1] is that the value of "Proof-of-Work ETH" (which is still a functioning blockchain[2], just like Ethereum Classic is) will drop as attention is focused on PoS ETH. And so energy devoted to it will drop in tandem.
It's also true that there are second order effects, like for example all the mining hardware dedicated to ETH needs to find a new home, which will depress prices for new mining hardware for "chains that are hardware-compatible with old ETH", and thus probably support their prices a bit.
[1] And do, I think. IIRC there was a stat rolling around a few months back showing electrical grid usage dropping due to the crypto crash, but can't remember where it was or how reliable the source seemed.
[2] Though AFAICT no one is tracking exchange rates for it yet, so your guess is as good as mine as to its value.
> The amount of energy worth buying to mine crypto is exactly equal to the value of the crypto mined.
That's like saying a stock price is directly proportional to the p/e ratio. Things have both intrinsic and extrinsic value. You are only considering the intrinsic value. In reality, people mine stuff at a loss all the time because they think it might be work more later, i.e. speculation.
Why would they mine it at a loss relative to the market when they could just buy it at whatever the current market price is, if they think it's going to go up in the future?
I don't think that's a solid argument. Some coins were structured such that there would be a finite supply (21M in the case of bitcoin IIRC). The health of a coin seems to me to be its transaction volume, and somewhat related its ability to be directly exchanged for physical goods or services.
So back to a previous user's suggestion: instead of buying energy to mine coins at a loss, one could instead buy the coins at a lower price which seems on many levels to make more sense than to continue mining.
There's a whole wasteland of people mining to keep coins alive. But not very many people. If ten people are running one graphics card each, out of nostalgia or just in case it catches a rocket to the moon, this is no environmental catastrophe, any more than a personal Minecraft server is.
Exactly! Which is why we expect the energy expended on mining PoWETH to drop to zero as the coin dies due to lack of interest, which was the upthread point you were arguing against.
I don't understand your point. If you want to speculate on crypto you can hand a credit card to Coinbase at near-zero cost, you don't need to buy mining hardware to do it. If miners are buying electricity to speculate, they're making stupendously bad decisions. They should sell that hardware and buy crypto; their leverage will be much higher and their costs will be vastly lower.
No, mining is economic activity, not investment. You pay stuff to get stuff. Whether you then invest your profits in crypto has nothing to do with where you got them.
True, and yet nearly every original bitcoin millionaire became one specifically because they decided to mine something that was worthless at the time. Sometimes you don't have money, but you have a GPU and someone else is paying for your electricity. I understand that doesn't describe most mining these days, but the point still stands: people will speculate at a loss.
Again, you're confusing "mining" (economic activity that produces new coins and fees that happen to be measured in BTC) with "investment" (acquiring BTC via any means with the intent to hold).
They are not the same concept, in fact they're completely orthogonal. You don't mine to get coins for investment, because you can get coins much (MUCH!) more cheaply via other means. You mine to get income in the present.
Wouldn’t you have better luck speculating if you just used your money to buy the coins directly, instead of buying GPUs and then paying more than the cost of the coin in electricity to mine it?
There's virtually no other ASIC resistant (i.e. can use GPU) PoW coin left to mine. There's the proof-of-work ETH fork, but it only has a market cap less than 2% of than real ETH. So even though they juiced the block rewards, miner rewards are more than 90% lower, which isn't enough to pay for electricity of the previous hash rate.
Yep. But with that the difficulty is going through the roof and earnings are dropping. Ethereum had A LOT of hash power while ETC isn't really used by anyone so where are the earnings supposed to come from. It's not going to be economically viable because the electricity is a fixed cost, even if some individual miners are trying it out.
There is a new PoW fork that was started by miners that want to continue mining. Some of the work could go there assuming it doesn't tank. I personally don't see the utility though, doubt it will be able to attract many users.
Turned off is more profitable than running at a loss for miners that need to fund their opex by selling what they mine.
It looks like it's already happening. After the merge there were a number of coins that saw huge spikes in hash rate which drove them to absolutely unprofitable levels. A lot of that hash rate has since gone elsewhere (most likely offline) and it looks like many coins are settling at a level in the short term that is breakeven at $0.06-0.08 kWh which many (most?) miners can't be profitable at as that is below their electric rate.
I think this is potentially the most interesting comment in here. If all of those GPUs just flipped to another cryptocurrency, global energy reduction would be zero.
In the short term, that's true because the GPUs are already purchased. In the long run, people will invest in new cryptocurrency mining rigs very much in proportion to the profitability of running one.
Not necessarily. It's true if the person making the decision is getting paid directly from the mining. But perverse incentives abound in the cryptocurrency space. For example, consider Celsius:
As they thrash around in bankruptcy, they have proposed that they will mine their way out of the hole. Will this work? I doubt it. But will it let the CEO stay in charge for a while longer, taking in more investor money and continuing to get paid? Possibly! So actual economic efficiency may not matter.
If you're the type of person who poured capital into mining hardware and you own it, which you likely do because it's more cost effective, you still have all that hardware sitting around. You're going to repurpose it to other mining endeavors or quickly find a way to try and eek more money out of if, because you were already that type of person.
I don't keep up with crypto and mining but until it becomes unprofitable or you can't pull money for and start operating in the red, you're going to continue consuming similar power.
Yeah, I've got 21 3080s/3090s, and I'm still mining (NiceHash switched algorithms for me automatically). But I've also listed my machines on vast.ai to rent out for deep learning. When they're not being used by a client, they're mining still. My electricity is super cheap, though $0.0875/kWh. That said, it's hardly profitable at all. I'm just speculating and breaking even at this point.
Ethereum miners were using GPUs. AMD, Nvidia, and others. They could switch their GPUs from mining ethereum which is about $1500 per coin to Ethereum classic but that is only about $40 per coin. I’d guess that wouldn’t be worth continuing with since it would be much less than than their electricity bill. They could sell their GPUs on EBay or other secondary markets, switch to protein folding, cloud-based password cracking, or SETI sky scanning. Maybe they use them to play high resolution video games like most people. Some smaller percent might notice next month when they eventually see that their ETH wallet hasn’t grown over the next month and Google why that’s the case.
Crypto mining occurs mainly where electricity is either cheap or free. Hydroelectric and Geothermal tends to produce the cheapest energy so many large mining outfits were relocated next to Hydroelectric and Geothermal plants. Many are in remote northern areas where computer cooling costs are less expensive too. $HIVE blockchain technologies was running enough ethereum miners to mine 7675 ethereum ($11.5 Million dollars worth) during the 3 month period ending in June 30, 2022 according to their quarterly earnings report. 100% of all of their miners used renewable energy sources.
In case of Ethereum that wasn't true afaik, a lot of it was mined in the US for example. Since Ethereum wasn't all ASICs like Bitcoin, there was more decentralization in that sense with individuals running miners from home, not always at the highest efficiency. Your first paragraph is correct regarding profitability for most miners though, based on the numbers I've seen.
One thing that I find interesting in the electricity debates is that if we took the gaming example and looked at the collective consumption of all people playing video games around the world, you'd arrive at even larger numbers of power usage and emissions. Yet this isn't ever discussed, even though an immutable public ledger like Bitcoin arguably has more utility for society than playing games. A lot of HN users probably play video games, CO2 emissions are of course mainly an issue caused by other people and activities oneself doesn't take part in. In Europe, there's also a trend of public anger against SUVs and there are groups slashing tires of cars, simply based on the shape and ignoring the actual energy efficiency. Happened to a friend of mine who couldn't understand why they targeted her car and left 30 year old gas guzzlers in the same street alone. I think a lot of it has to do with emotions more than rational considerations around sustainability.
>One thing that I find interesting in the electricity debates is that if we took the gaming example and looked at the collective consumption of all people playing video games around the world, you'd arrive at even larger numbers of power usage and emissions.
Okay, this is the second time I've seen an argument of the form: If you're not okay with the energy consumption of PoW cryptocurrencies, you can't be okay with X.
The first time I called it out, was someone pointing at the energy consumption of making, transporting, and storing ice cream. You are bringing up gaming.
You are advocating for making our children's/descendants lives worse, our lives worse, and trying to throw a valuable industrial subsidizer of the state of the art in many sub-fields of computer science...
...to defend the least efficient, wasteful,least empowering form of computation we've ever discovered. You're taking and making hostages of something that has objectively wrought joy and innovation to millions of lives.
If you find yourself on one side of an argument, and ice cream/video games on the other... I'd recommend having a long hard think about how you managed to get there.
I think that pattern of tire slashing is rational. Old cars will die, and the thing you want to hurt most is the incentives for making more oversized vehicles.
If it's a hybrid SUV then it's more complicated, but a gas SUV makes sense as a target even with an engine tuned for efficiency.
>even though an immutable public ledger like Bitcoin arguably has more utility for society than playing games
Video games' secondary impacts to simulation, film making, and modeling have and will provide more benefit per energy used than any proof of work system ever will.
I guess hydro/geo is better than them firing up a coal fired power station to do the job. But all the "mining" nonsense just sucks up clean power than could be used in the real world which hopefully pushes out more polluting sources.
All to supposedly create artificial scarcity for .jpg files.
At least that illness is over now somewhat. Hopefully all the miners go bankrupt.
I find it incredible how many arguments rely on a uniform distribution assumption here. There are markets where crypto miners are double digit percentage of utilization. They have very favorable conditions for mining, like the Pacific North West. The “one Finland” isn’t smeared over all power consumption, it’s highly congregated in a relatively small number of locations. The argument that a sudden devaluing of the use of electricity has no impact in the power infrastructure where it’s concentrated is absurd. I’m not saying it has or hasn’t happened - I’ve no idea. But it will be news if it does happen because operators will see double digit drops in demand locally and it’ll be noteworthy. But I don’t think it’ll be like energy prices in the EU improve - the mining happens in places with huge gluts of power they can’t otherwise sell or distribute for more.
> Shouldn't there be some power stations reducing their output as a reaction to reduced demand?
No, because ETH mining is/was quite distributed globally, let's say across thousands and thousands of power grids. A single grid or power station shouldn't be able to notice the difference.
Think about it this way: all of a sudden, domestic fridges consume 1/100th of electricity, compared to before.
Fridges are 0.1% of average power consumption. Thousands of fridges in a given area are powered by the same power station. The power station barely notices the ~0.1% reduction in power consumption, compared to the day before. Shrugs.
Is it just me or does that sound like an insignificant amount? Finland has a tiny population, if ETH mining only used that much electricity sounds like it it was pretty great to begin with.
The people who were mining ETH didn't give a toss about the environment. They invested in PoW hardware and they're not going to stop using it because ETH is now PoS, they'll just mine something else.
I just hope the PoW markets collapse now ETH is moving on.
That is rude. I care about the environment, still I am mining.
How is it good for the environment? I only mine when my house and battery does not need electricity and our solar is producing over 1kw. Everything I earn (around 100-300$ a month between February to Oktober) I invest in the next environment project. My miners need about 500w. This is a fraction of the amount I put back into the grid.
If I would not do this, our new heating system wouldn't be financed yet.
You are certainly right.
However I am under the assumption that you only mine where electricity is cheap. If you are not stealing it, there is only nuclear, wind, water and sun produced energy left => Miners are kind of forced to use environment friendly energy.
It's better for the environment if we build expensive cables to send that power to areas still emitting CO2, instead of selling it cheaply to miners. At least up to some pretty large distance.
There's a case to be made for using the spikes of solar power to run miners, on a grid dominated by renewables, but that requires the economics to work out just right with very cheap silicon that can sit idle a big majority of the time.
Yes, germany is already doing that with electric cars. I have a Wallbox in our garage, with seperate electric because it is sadly 300 meters away from our house. So I can't use our solar. However this connection is terminated via GSM when many people are using the grid. This is usually around morning and evening. If the grid isn't exhausted they will turn the electricity back on. I really like this approach. It works well.
It would be nice, if this would be mandatory for big official mining operations. Pretty sure they would just move somewhere else but still.
Unless your “new heating system” is a euphemism for a gpu array, you’re being pretty dishonest here.
Mining takes energy that is likely created by fossil fuel, that would otherwise be used to some necessary purposes, and burns it.
Unless you are generating green energy yourself, you’re likely hurting the environment.
I’m not saying you need to stop, and I’m not saying mining rigs are the end of the world (lord knows at least they aren’t frivolous airline flights), I’m just saying the previous poster isn’t being rude… it’s a valid point.
I was under the impression that when they said "This is a fraction of the amount I put back into the grid", it implied that they are generating their own electricity to use for this, most likely from solar panels.
This is an assumption though, they could be running a large diesel generator for all we know.
I think anyone involved in a PoW system (holding, mining, spending) is legitimising wasting energy. Even if you're using a sustainable source, others aren't, and the power you're using could have been used elsewhere.
Your personal impact might have been small, but it still takes serious compartmentalization or blissful ignorance to be involved in crypto and profess care for the environment.
My understanding is that Ethereum PoW used traditional GPUs whereas for example Bitcoin uses ASICs - "PoW hardware" is not as interchangeable as you're implying.
RandomX is pretty GPU-hostile. If you have a lot of VRAM, you could theoretically make it work okay, but I don't know of a GPU miner that has any semblance of performance (or perf/watt) that is close to a CPU.
Gotcha, CPUs have been most of my experience with it.
I dabbled with GPUs but it seemed like I'd have to do a lot of clock/voltage tuning to make it worthwhile. This was way before energy prices jumping, too
There are lots of other coins that still use GPUs to mine, though. I imagine most people who are running profitable operations will switch coins. Hopefully the popular coins will switch to PoS and the mining operations cease their profitability and shut down.
There are a 1000 other coins using GPUs for mining. E.g. "Ethereum Classic". Miners will switch to these. There are ETH ASICs btw, they just aren't as dominant as BTC AntMiner ASICs.
It is only worth it though if the shitcoin is worth more than the cost of electricity used to generate it. At the moment and for many miners, that just won’t be the case.
Most shit pins probably have few miners, and are pretty susceptible to market manipulation. I large influx of miners across the market could dramatically change the price.
Also, many miners are already using software that automatically changes their hardware to target whatever coin has the most profitability. It’s expected that this will continue just without ETH
They've actually forked Ethereum to keep the original mined version, ETHPoW ($ETHW). They'll keep mining on the original PoW chain as long as it remains profitable.
I'd expect the value of ETHW to crash fairly rapidly, though, because there are not many buyers interested in buying into a deprecated chain without official support or ecosystem buy-in, and lots of sellers who need to sell their mined ETHW to fund mining operations. Then we'll see miners shut off and leave the network, as the mining rewards can no longer support the electricity costs of mining. At some point it might get 51%'d, but at that point nobody will care.
With the price of ETHW currently at $22.80 and the price of ETH at $1500, I can’t imagine mining ETHW would even cover the electric bill. It’s also not clear to me what advantages ETHW would over ETC which is about $37 ea.
Previously ETC was often used as a lower cost dev-testing server for ETH applications that might not be ready for the main-net.
With ETH off there’s going to be a significant loss of profit to be had and people stopping because the expected value of mining has gone negative. Store stocks of graphics cards are booming. There absolutely will be a reduction in PoW power usage globally.
Maybe. What are they gonna move to though? If everyone moves to some other coin (not Bitcoin, because that's different hardware), it'll quickly become unprofitable to mine. They'll just sell their hardware, most likely. Hopefully.
Well, there are options. I do not want to point to obvious candidates, because I am biased and it could easily be misconstrued, but RVN just just saw a decent influx of those ETH miners based on recent spike.
Life hates vacuum. Short of outright ban, nothing will change in that space.
I will admit that it is hard for me to make a good prediction here. I ridiculous amount of it boils down to psychology and perception and not any kind of fundamentals. The following statement can be true in some instances, but I am not sure if it is true across the board.
I wonder what will happen now that the ponzi structure is now gained new velocity as energy constraints have shut down proof of work mining. PoW creates centralized collusion between those who can afford the best miners and this helped the velocity of the chain itself since it is only as good as the last mined block.
Now that layer is gone, its this proof-of-stake which is a bit funny since, Ponzi schemes are also proof-of-stake, where the previous investors stake's performance signals the next until the order books flip to a highly skewed with a very long til, it results in the last group who were late to the party, get caught with the bags.
I wouldn't be surprised if there are many whales dumping as they would know (and I hope so) what the new paradigm shift is in this digital ponzi gold rush.
Also rather anxious for these fellas who promoted securities written on ethereum. The SEC flat out came out and said almost all cryptocurrencies passes the howey test recently. This PoS seems perfectly timed for the occassion.
You're confused, in a ponzi earlier investors are paid from the investments of future investors. That's not how PoS works. In Ethereum, everyone is diluted to pay stakers and the time of their investment does not matter.
> in a ponzi earlier investors are paid from the investments of future investors.
Also, a Ponzi scheme requires fraud, where the earlier investors are being lied-to about where the money comes from in order to paint a false financial picture of the company.
Not directly related to your point, but I wanted to put that out there since it's a pet-peeve of mine that "Ponzi" gets frequently misused as a label for anything the speaker thinks is unsustainable.
A pre-mine might make it a security, but it's basically the opposite of a ponzi. Not everything is a ponzi. I think you're just confusing different financial constructs with one another
You're confused. In The Ethereum Ponzi, early token holders 'stake' their tokens and are paid by future transactors.
But here's the funny thing that no-one gets: all assets are a ponzi scheme (stocks are only worth something now because future rubes will buy them for more later), what makes bad ponzi schemes is when the underlying asset that everyone is speculating on doesn't do anything useful.
> stocks are only worth something now because future rubes will buy them for more later
Stocks are valuable because they are a claim on the assets and future profits of a company, as well as a claim on the ownership and control of the company.
Some people buy stocks just in the hope that a Greater Fool will buy them for more money later, but that's not the same thing as calling stocks in general a "ponzi scheme".
But no, that's not the same. If the owners of a stock received a fee when somebody used the company's product, that wouldn't be a "ponzi". It lacks the key element that earlier investors are paid via the investments of future investors. That's just not what happens on Ethereum.
I am, because it's true. For example I own a lot of Ethereum and didn't pay any transaction fees to obtain it. Most ETH transactions work this way because they happen on centralized exchanges.
>Most ETH transactions work this way because they happen on centralized exchanges.
Well then what does Ethereum do actually? And why exactly can't those centralized transactions be denominated in "Coinbase points" without all the crazy "Proof-of" code?
Because if Coinbase did that, nobody would use them, but people do use Ethereum.
Btw, Binance has done this with BNB and their USD stable coin, and they are top ranked coins by market cap, so in fact sometimes it is the case that having a highly centralized and liquid coin backed by a major player is good enough.
>Because if Coinbase did that, nobody would use them, but people do use Ethereum.
And that doesn't feel anything like a ponzi scheme to you? Because, again, as you state, people are not using Ethereum (that would require paying fees to old investors) they are buying it (in hopes that they aren't too late to become an old investor themselves).
Or are you literally saying that the value of ETH is that "Aesthetically, it's nicer to see 'ETH' at the end of an account balance than it is to see 'USD'"?
People are using Ethereum, they are paying transaction fees to use it. Those fees mostly do not go to stakers, they are mostly burned. Even if that weren't the case, it's not even close to a ponzi because the stakers aren't paid out by future stakers, they are paid out by diluting everyone and via fees for usage.
Also to address your second point, I'm not saying that, but that would not make it a ponzi at all. Art has this property, art isn't a ponzi.
> People are using Ethereum, they are paying transaction fees to use it. Those fees mostly do not go to stakers
Why exactly are people staking then? How are they being rewarded for slashing risk? Who do you think gets the gas fees? You know ETH2 still has them, right?
> the stakers aren't paid out by future stakers, they are paid out by diluting everyone and via fees for usage.
Again, you’re absolutely misinformed here, but aren’t crypto maximalists supposed to be all about how bad inflation in fiat is? Or is that old news?
> Also to address your second point, I'm not saying that, but that would not make it a ponzi at all. Art has this property, art isn't a ponzi.
You’re the one who said “most ethereal is bought on centralized exchanges”! You can’t have it both ways.
The people paying a fee are not investors. This is like saying the grocery store is a ponzi because customers pay the shop owner, who was an early investor.
The people paying a fee do not expect to make a profit off of holding ethereum. (Some of them do, but that's a coincidence in the same way that a shop owner might shop at their own store).
So, have we observed global energy usage go down by about one Finland?
Shouldn't that be observable somehow? Shouldn't there be some power stations reducing their output as a reaction to reduced demand?
Anyone know how this would be visible, and on what kind of time frame we expect it to become visible?
I'm not claiming it hasn't happened. I just feel surprised to not see more coverage of that in this article, nor here in the comments. Energy efficiency is largely the point of this major change. Shouldn't there be graphs of the power grids everywhere showing a big drop? Maybe my expectations are just off on that.