Hacker News new | past | comments | ask | show | jobs | submit login
Bitcoin Energy Consumption Index (digiconomist.net)
256 points by schwabacher on Aug 26, 2017 | hide | past | web | favorite | 280 comments

You can't compare Bitcoin to Visa. Visa is merely a payment sytem -- a means of transfer of money that already exists. Visa is not money.

Bitcoin however is money. As well as being a payment system.

The correct comparison would be to compare the cost of Bitcoin to U.S. dollars or gold. Gold requires hundreds of dollars per ounce and untold energy to extract. U.S. dollars require the existence of a powerful economy and trillion dollar military to keep the currency secure and desirable.

Bitcoin wants to be money, but it really isn't. It fails the traditional account (originally from Jevons); see Kubát 2015 [1]; Mandjee 2015 [2]. More obviously, it is far too volatile to function like money in a normal sense, see Golumbia 2017 [3]

[1] http://www.sciencedirect.com/science/article/pii/S2212567115... [2] https://litigation-essentials.lexisnexis.com/webcd/app?actio... [3] https://motherboard.vice.com/en_us/article/xwwv83/cryptocurr...

And yet some people, like me, use Bitcoin to buy products and services. Your articles merely explain that Bitcoin legally is not money. But in practice it is used as money, it is seen as money, so it is money. Practice trumps theory.

Also these articles fail to mention that volatility is quickly decreasing, thanks to increasing market depths: https://mobile.twitter.com/lsukernik/status/8649208737189519...

Well it's pretty easy to spend an asset that's consistently increasing in value... the real test of it's utility as money is how many businesses and people have started accepting Bitcoin recently.

You should also look at this as % of merchants and users in the world.

Is it 10%,50%,or 0.001%?

What percentage of merchants and users accept HKD?

100% in Hongkong, which is also where the most HKD are circulated.

Pointless comment. Bitcoin is a borderless technology. There will never be, by design, a "Hong Kong-sized" area with BTC accepted everywhere. That is, unless BTC achieves world domination :-)

OP asked "in the world"

If BTC was circulated in only a Hongkong-sized area, that would be a fair comment. It isn't, the area of its circulation is the world.

Or hamburgers? After all, you can trade them for the local currency. Hamburgers must be a currency!

Why would you ever want to spend an asset that's increasing in value? Seems like a useful currency should be mildly inflationary.

Why not dump your entire life savings into it right now?

Because there's risk, it fluctuates, there's no guarantee it will always go up. People need to decide to what degree do they want to invest and be exposed to the risk and reward. Say someone decides they want 1BTC of exposure, and are contemplating buying an item of 0.01BTC. Would that really be a big deal? Small changes in the degree of exposure shouldn't be a financial problem to worry about.

> Why not dump your entire life savings into it right now?

Plenty of the bitcoin true believers have.

Everything you're talking about makes sense for investing in a commodity, not for a currency.

The answer is very simple: no one is absolutely sure BTC will always gain value.

>The correct comparison would be to compare the cost of Bitcoin to U.S. dollars or gold. Gold requires hundreds of dollars per ounce and untold energy to extract. U.S. dollars require the existence of a powerful economy and trillion dollar military to keep the currency secure and desirable.

No it isn't- the energy cost of bitcoin is ongoing and depends on the volume of transactions. Each time money changes hands, it costs energy to verify it. Comparing it to VISA or ACH makes the most sense, because those also incur ongoing costs per transaction.

This is very wrong. It does not work that way at all.

Volume or value of transactions has nothing to do with the energy expended.

The energy is needed to calculate a mathematic proof that makes sure your transaction cannot be reversed or tampered with.

This mathematic proof on average is calculated every 10 minutes, with no correlation to the number of transactions.

You make a transaction, and if the bitcoin network has expended 1TWh after you sent it, at least 2TWh are needed to reverse that transaction (and/or any other transaction). You can see how over time it becomes impossible, even with unlimited money, to reverse one.

The hashing is merely a way to choose a random block. It's not a proof and it doesn't do anything fancy. It just picks a random computer to say "this is how the transaction history looks".

The energy increases with the volume of transactions because there will be more people calculating hashes. The more people mining, the harder they make the hash calculation so that the target is ten minutes. More transactions = more people mining because there is more demand to verify blocks.

I repeat, you got it very wrong.

Transaction number is not correlated with mining energy expenditure.

Mining is used to secure the ledger in a way that the same amount of energy is needed to alter it.

The block hash begins with a number of zeros.

Try for yourself how many tries it takes to find a string that hashes to a hash beginning with 3 zeros.

Bitcoin block hashes begin with 13 or 14 zeros IIRC. This means that you need to try trillions of combinations again if you want to alter a block. And then you must keep finding other hashes with enough zeros fast enough to outcompete the whole network.

But the cost of hashing is the same, with 1, 10 or 1000 transactions in a block. And the block time is on average fixed at 10 minutes.

You don't hash transactions directly, you hash a block header that keeps only the merkle root hash of the transactions. The merkle root is fixed size.

> Transaction number is not correlated with mining energy expenditure.

You're only kind of right. The mining algorithm cares not one bit about number of transactions, so higher tx volume has no direct effect on the mining energy expenditure. But it does have a second order effect.

More transactions usually entails higher fees (as people compete for block space to put their transactions in). More fees means higher mining rewards, means higher incentive to mine, means more mining activity, and, therefore, higher energy expenditure.

> The mining algorithm cares not one bit about number of transactions

Actually it does to some extent because the transaction hashes are used to calculate the block header hash.

Yes, but the hashing of the merkle tree to obtain the merkle root takes like 10^4 hashes for 10^3 transactions. Compared with 2^80 that is about the number of hash operations needed to find a POW block header, it it trascurable.

I'm not sure about that calculation to be honest. The merkle root is very simple to calculate and verify that's why it's so useful. The POW difficulty is in finding a hash starting with the right number of zeroes and containing the previous block header hash.

Did i get this wrong?

No, you're right.

Numbers i said are my guesses, but the order of magnitude should be about that.

I don't remember exactly how many hash ops you have to do to put 1000 txes in a merkle tree, but it should be more than 1000 and less than 10000.

My point was exactly that. Computational cost of the merkle tree is negligible when confronted with the computational cost of the POW

Hash costs are a large fraction of overall costs, but not 100%. Network bandwidth is not free. The more transactions the more beefy the full nodes must be, worse every node pays the full costs or lose out on some transaction fees.

But, it gets worse simply maintaining value takes hashing power to avoid a bad actor double spending and destroying faith in the system. Thus, even as a value store you need to pay for large scale hashing.

You can't just scale the number of transactions in a block forever and still have a stable currency. If you have only a few miners working on massive blocks, then they confer very little confidence onto the transactions in the blocks. More transactions, more mining. The whole point of the system is to verify transactions and it stops working if it doesn't do that.

> You can't just scale the number of transactions in a block forever and still have a stable currency.

If you mean that over time the incentive to centralization become stronger, yes, you are right. There must be competition to enter the blocks, otherwise when mining subsidy ends, there will be no incentive to secure the ledger.

If you mean that ten times the transaction have a computational cost 10 times greater (or 5, or 2), you're wrong.

> The whole point of the system is to verify transactions and it stops working if it doesn't do that.

Plenty of cryptocurrencies are mining tons of empty blocks. on the short term, if there are no transactions, mining continues with the same difficulty.

The effect is long term: if noone is using the currency for transaction, it has no value so less and less people mine it. The difficulty drops, and the security drops, pulling value down even more.

You are almost right, but it is a very indirect effect, and takes years to manifest itself.

You have a very distorted view of how it works.

The number of transactions is entirely unrelated to the number of people mining. There could be an increase in transactions while the number of miners falls, or vice a versa. This is why the hash difficulty adjusts, to ensure that the new block rate (and therefore the transaction rate) stays the the same regardless of the number of miners.

If the number of miners dropped, energy use would drop, but transaction throughput would remain exactly the same.

Also, transaction size (in value) is entirely unrelated to transaction size (in kilobytes) which is what effects the number of transactions that can happen. Most transactions that just transfer some bitcoin from one assess to another are exactly the same number of kilobytes irrespective of how many bitcoin are being transferred.

> The number of transactions is entirely unrelated to the number of people mining

I also commented elsewhere to this effect, but this is not quite true: number of transactions correlates very strongly with higher transaction fees, which further incentivises mining activity. Higher transaction volumes also usually come hand in hand with higher prices, which is another incentive for mining activity. Just because there is no direct effect (which you're absolutely right — the mining algorithm cares not one bit about the transaction volume) doesn't mean there aren't second order effects at play.

Yeah, I saw that, you are of course right, but like you say that is a 2nd order effect.

The post I replied to was claiming that more transactions strictly required more miners to process them, which isn't true. Yes, more transactions might incentiveise more miners, but it doesn't require them. And it's only one part of the complex interplay of incentives that miners face.

Say price dropped sharply, that could result in an increase in transactions as speculators scrambled to get their coins to an exchange to sell, while at the same time we might see a drop in miners because with lower bitcoin priced there is less payout for mining.

Ultimately the sequence of incentives and disincentives that drive miners is complex. Because of course, although a drop in price might deter some miners, if others believed that the drop was only temporary may continue mining and just hold the bitcoin to sell later when the price recovers. I think it's hard to say anything totally concrete about the expected miner behaviour following any event.

If a single bitcoin transaction uses enough energy to power 5 households for a day, that certainly makes me not want to do any bitcoin transactions anymore.

I knew it was wasteful, but this article is really putting it into perspective. And because of the proof-of-work system, the wastefulness is intentional; you have to prove you've wasted enough time and energy to be allowed to mine a bitcoin.

To say that one transaction costs that much power is a misunderstanding of what the power is useful for. The amount of electricity spent mining a block is a function of the block reward, which is a function of the value of the total system. A block doesn't just protect the transactions inside of it, it also protects every single transaction in the whole history of bitcoin.

And, the high value transactions tend to dictate the value of the whole system. Not every transaction is equal in that regard. If all the high value transactions were to move to another system, the coin price would plummet, and so would the amount of energy required to mine blocks, and thus the 'per transaction' cost for low value transactions would plummet.

The right way to think about it is that we spend around a billion dollars per year protecting $70 billion in assets. Put that way, it doesn't seem so bad.

If you look at the percent of the US GDP spent on military defense, you'll see that bitcoin is quite attractive by comparison.

It's hardly justifiable to imply that the entire military defense of the United States is simply to protect "$70 billion in assets."

If the entire monetary system of the US were replaced by BitCoin, do you think there would be no need for an army? Would the block ledger be "protecting" all the actual physical assets, land, and people?

You can argue over whether the US should have a military, or its size, but to imply that it's just to protect the dollars isn't going to sway anyone in an argument.

I meant only to imply that the US military alone is about 3.5% of our economy, and that the protection to value ratio there is about 1:30

I did not mean to imply that bitcoin could replace the entire military, only that the ratio of spending vs. assets defended is better.

The protection to value ratio is drastically beyond 1:30. Your value calculation is worthless if it only takes into account annual GDP.

The total value of the United States, all assets, private and public, is closer to $250+ trillion [1]. That's a 1:400+ ratio.

US household assets alone are near $95 - $100 trillion at this point [2]. That's a ~1:160 ratio just for household wealth.

You missed by an extreme amount in your speculation and that's just based on present value. If we were going to be rational about it, we'd want to consider that present military expenditures help to keep safe future value creation (I'll note that I'm not arguing whether those expenditures are $n too high or too low).

[1] https://en.wikipedia.org/wiki/Financial_position_of_the_Unit...

[2] http://www.reuters.com/article/us-usa-fed-households-idUSKBN...

The military is also not the only protection system in place. You are right, just considering the GDP is too limited, but considering just the military is also too limited. Police, fireman, courts, security systems, the SEC, locks, alarms, etc all contribute to protecting assets in the US.

It's a hard number to measure. Bitcoin inflation has a fixed measurable cost though (12.5 coins per 10 minutes), if that's acceptable to you for the benefits you get, then it's money well spent burning electricity.

"If a single bitcoin transaction uses enough energy to power 5 households for a day, that certainly makes me not want to do any bitcoin transactions anymore."

This is another point where BECI is misleading... the author managed to screw up so many things on his site :-(

Feel free to transact more. This will NOT consume more energy, as explained by many other commenters. The energy per transaction decreases as more transactions are processed, see argument #5: http://blog.zorinaq.com/bitcoin-mining-is-not-wasteful/

> If a single bitcoin transaction uses enough energy to power 5 households for a day,

Care to share a source for this claim? I think you might have gotten transaction and new block generation mixed there, but I may be wrong, of course.

I got it from the article, but several responses say that it's a misrepresentation by the article, and that more transactions would not consume significantly more energy.

I too think that's the case but I'm always willing to learn.. thanks for answering.

Energy consumption is in no way related to transaction volume. Whether there are thousands of txs per block or only one the energy used will be the same.

Which is part of the problem, since the max number of transactions Bitcoin can process a second is around 10.

Sure, but it means it when it can handle thousands of transactions the energy usage doesn't need to increase.

> the energy cost of bitcoin is ongoing and depends on the volume of transactions

Wait, what? That's not correct at all.

The energy cost is based on the "difficulty", which is derived from the moving average of the hashpower in the network. Bitcoin effectively forces the network to take a specific amount of time to mine a block irrespective of the number of mining nodes. The transaction volume has nothing to do with it.

For a fair comparison we should look at "energy" in an abstract manner. Time and money is energy too. Every time gold exchanges hands, it also costs energy.

Think about the energy expended to take that gold out of earth. And to make the chemicals for extractions....

If read the article correctly, he's including the cost of mining Bitcoin, not merely energy cost to verify a coin (which is absolutely trivial). Once all Bitcoin is mined (yes, there's a built-in maximum), or it becomes prohibitively expensive to mine, there won't be any mining costs.

You can look at the current Bitcoin energy costs as the startup cost to establish a new gold mine or a new country with its own currency.

> Once all Bitcoin is mined (yes, there's a built-in maximum), or it becomes prohibitively expensive to mine, there won't be any mining costs.

In that future you will still mine blocks, but there won't be any new bitcoin generated in that block. Sure, you might want to use a different word for that than "mining", but it will still have very real and significant energy.

"Validators" is the more generic for term what are also called miners or stakers.

Not quite, even when the cap is reached, new blocks will still need to be created for new transactions, which will require proof-of-work to achieve consensus, and thus energy expenditure.

Once all bitcoin is mined, there will still be mining costs, but they'll be 100% paid by tx fees.

Bitcoin will probably look extremely different by then, so most bets are off anyway.

Mining always needs to happen. Once it stops producing coins, the cost burden just moves over to the people using the currency. The mining cost only decreases if the proof-of-work difficulty decreases.

If mining becomes less profitable, fewer people will mine and the proof-of-work will tune to be less difficult, but the cost never disappears and there's still a huge amount of energy wasted in the selection process for adding blocks. If too few people mine the currency will be totally destabilized.

> Gold requires hundreds of dollars per ounce and untold energy to extract.

A valid comparison. Gold mining is spending resources to increase the amount of gold in existence, without delivering any other value.

> U.S. dollars require the existence of a powerful economy and trillion dollar military to keep the currency secure and desirable.

Not a valid comparison. Outside of a few relatively inexpensive activities at the US Mint and inside banks, no resources are being spent specifically on increasing the money supply. The value of the dollar is the byproduct of economic activity that exists primarily to achieve other things.

Note that bitcoin mining does not primarily exist to increase the amount of bitcoin in existence. That's just part of the incentive to mine.

Bitcoin mining exists to timestamp transactions so that the network can achieve consensus on the state of the ledger. It is designed to use as much energy as possible, so as to make it expensive to cheat.

In fact, in the white paper it's just described as the decentralized time stamp server and that's its entire purpose. The term "blockchain" and all the hype about its magic powers came later.

Gold is also not money, and arguably also mining it is largely a waste of resources and environmentally disastrous.

>Bitcoin however is money.

Bitcoin is not money as we think in a modern way. If you think of money a way to exchange goods then yes Bitcoin is money. Modern money (printed piece of paper labeled in currencies in its materialized form. Just a sequence of digits on your bank account) is dept. That's why currencies have interest rates (IR). It is also possible to create money out of nothing. IR and volume are adjusted according to the economy. Bitcoin is closer to a commodity in the sense that it is finite and requires work to be extracted.

> Bitcoin is closer to a commodity in the sense that it is finite

Bitcoin could be forked in the future to become inflationary. It has no permanent rules, things can change through consensus.

It isn't like gold (gold can be devalued through poeple losing faith in it, etc., but the supply of it doesn't inflate other than through discovery/mining).

Bitcoin can also be devaluated by lost of trust. Also the current supply of Bitcoin is through mining only.

If rules are changed by a fork then you will have 2 different coins. We already saw it with bitcoin cash. Another way would be that the current consensus on Bitcoin itself changes. But that's a risk. How can people then predict such a consensus change. How risky is it to have Bitcoin since the day after the rules may change ?

US Currency budget for 2016 was 709M USD: https://www.federalreserve.gov/foia/files/2017currency.pdf

Ecological concerns aside, this puts a bit of a perspective on the overhead costs of maintaining a currency.

Fiat currencies require a stability of power of course, which is impossible to quantify.

>Gold requires hundreds of dollars per ounce and untold energy to extract.

The gold you trade is already almost all extracted, I doubt there's 1B$ a year in gold extraction investment but maybe there is.

>U.S. dollars require the existence of a powerful economy and trillion dollar military to keep the currency secure and desirable.

All currencies require a relevant economy using them to be valuable so that's not a real requirement. As for a trillion dollar military that's doubtfull. There are plenty of currencies with the market cap of Bitcoin that don't have nearly as much of a military backing them.

Accurate figures exist. In 2016 world gold production was 3,100 metric tons, worth US$128 billion. Above ground gold stock is something over 180,000 metric tons.

New mining increases gold stock 1-2% per year, but very dependent on market conditions.

Gold has double the volatility of equities with less total return than U.S. treasuries. It doesn't really belong in most people's portfolios.

Compare it to cash and coins seems the most appropriate. How much energy does harvesting cotton, making into bills, and distributing it all over the place take? Don't forget those fun armoired trucks that have to cart it from place to place. Pennies famously cost more than $0.01 to produce because of the cost of metal. Metal that has to be dug out of the earth, trucked, smelted, trucked, stamped, trucked again.

According to this it costs almost 21 cents to "make" a dollar of bitcoins. According to the mint's 2015 annual report[1] they made 1,114 million dollars in circulating coins at a cost of .51 cents per dollar in coins.

However bills cost 4.9 cents/dollar for $1 bills and half a cent per dollar for $20 bills. So the vast majority of money is much, much cheaper than a bitcoin.

[1]: https://www.usmint.gov/wordpress/wp-content/uploads/2016/06/...

That's to make it. What about trucking it around, accounting for it, securing it, etc.?

So many Bitcoin fans, so many "whatabouts". I would say the burden is on you to show that producing and moving cash requires ludicrous amounts of energy like Bitcoin does. It sounds unlikely.

Yes, but you have to make new bills to replace old bills.

I don't agree with the comparison to "making" coins or cash.

Coins and cash actually get consumed as they degrade and/or get lost or destroyed. You need to replenish the supply.

But you can circulate bitcoin forever.

You need to compare transactions.

How do I send you cash? You can give me an address to meet you at. We both drive our cars or take public transit there and consume some amount of energy and time to meet up. Then I give you cash.

Or you could just tell me a bitcoin address to send to.

I can also put that cash in a sufficiently opaque and uninteresting-looking envelope, put it in the mail, and the Post Office will take over. There is time and energy cost incurred, but it is a cost already being paid. Mostly by advertising, right now, since an assortment of lawmakers have decided in their infinite wisdom that it's not something our tax dollars should be supporting any more.

A literal comparison to paper bills isn't quite right. A small sheet of linen/cotton paper could represent $1, $100, or hundreds of millions of dollars (eg., a stock certificate of a million shares of Apple stock).

You have to compare Bitcoin to the concept of money, like U.S. dollars, but not literally the paper. A comparison to bars of gold is OK however since that is money and not a representation.

> A comparison to bars of gold is OK however since that is money and not a representation.

Bars of gold certainly are a "representation" of money. They are only worth something so long as we all say they are. (Besides the little bit of intrinsic value from electrical wiring.) If people stopped saying gold was worth anything, it wouldn't be.

Cash and coins are not a store of value and are also a fiat currency associated with a central authority. Gold is a better analogy.

You can't buy a pizza with gold. You can with cash. BTC is trying to do both, and is failing as a store of value right now due to volatility though that may change.

It's neither cash nor gold, but shares more properties with gold. And it certainly is not failing as a store of value, with a steady increase in value over an appropriate time window.

Btw, you can buy pizza with gold, I guarantee someone will take your gold for pizza.

I'm pretty confident that you will find more pizza places that would be willing to sell for gold than for bitcoin.

War economy also functions on gold.

If it wanted to, Visa could "become a crpytocurrency" tomorrow. It could make Visabux and continue working without real changes to infrastructure.

Bitcoin is a payment network with its own currency. Visa is the same thing, but the currency is linked to dollars. It is the most apt comparison.

The comparison to VISA is a common one. Whether it makes sense to use BTC in a certain way ultimately depends on the user and how they're using it. But honestly, as a medium of exchange Bitcoin simply doesn't make sense in most cases.

At least until/if Ligtning networks is proved to work.

The fairer comparison would be Bitcoin vs Visa the corporation, which includes buildings, people, air-con, employee traveling for work, to begin with.

Still, let's hope Bitcoin moves away from wasting energy soon.

Bitcoin can't move away from wasting energy - the entire design is based on that aspect.

Ad discussed above, the solution may be moving from proof-of-work to a proof-of-ownership scheme which doesn't require as much power to run

That's not bitcoin and nobody has demonstrated that any such thing is possible.

Also the courts that enforce the contracts visa makes, the FBI that chases people who commit fraud, the police and jails who enforce the decisions of the court, etc etc.

Blockchains can operate without all of those.

A society without contracts, police, or jails is a failed society where you will not be able to reliably power your computer, and where you will be too busy fending off looters to care about blockchain anymore.

Bitcoin doesn't even try to replace contracts. Hard-line Ethereum Classic fans think they can replace contracts (they won't).

No that's that makes blockchain so fascinating. It operates well in countries with no infrastructure. Bitcoin is incredibly popular in Venezuela despite being completely illegal and actively hunted by the government, and despite the who country being very unstable.

You can't use a blockchain to protect your land rights, but you can use it to protect digital rights and that means you don't need external resources like courts to spend their time on digital rights. They can focus entirely on land conflicts and other physical matters, which will make them cheaper overall.

So living off of Bitcoin and the black market in Venezuela is your idea of what we should aspire to?

Even if a few people in Venezuela manage to use Bitcoin (propped up by the US dollar, nobody's selling their Bitcoin for Venezuelan currency), that still easily meets my definition of a failed society.

Payment systems like visa are relatively cheap to operate when being used honestly. There may be overhead when people commit fraud, but to me that seems more prudent than giving honest actors an intentionally inefficient system just to make abuse more difficult.

Since I personally don't worry about centralisation and "trust", I like the idea that a system like the blockchain could be used to revolutionise payments without the need for tin foil hats and proof-of-work.

> money that already exists

What does "exist" mean? The currency exists, but the amounts transfered over visa and other systems is not related to how much physical banknotes exist.

I haven't touched a banknote or coin in over a year. If they didn't exist I wouldn't have noticed.

You could argue that puts me in the hands of banks and visa/mc and you'd be correct. They know more about me than I'd like. But I hope the energy cost of my transactions are pretty low. More and more stores no longer accept cash, meaning they have no transports for cash.

The currency I use is not backed by a large gold reserve nor a huge military. Just a regular national bank of a small country.

So visa isn't money, but a central bank, commercial banks and visa can maintain a currency without the need for a lot of physical money, and hopefully with reasonable energy use.

> compare the cost of Bitcoin to U.S. dollars or gold

All-in sustaining costs for mining an ounce of gold (worth $1,293.70) is between $1,100 and $1,200.

Yes. And Bitcoin, or another digital currency can do better. And will do better.

As more people use it and invest in it, the price will surely become more steady.

That is an absurd amount of energy. This really changes my opinion on this being a feasible currency.

It's worth noting that this Bitcoin Energy Consumption Index (BECI) is not without controversy. A more detailed analysis is this one:


and the author of that has specific criticisms of the BECI calculation:


Miners consume an amount of energy comparable to Christmas decorative lights in the US (alone) I never see anyone refer to these lights as "absurd" or "infeasible" :) http://blog.zorinaq.com/bitcoin-electricity-consumption/

Also note that BECI overestimates the consumption: http://blog.zorinaq.com/serious-faults-in-beci/

1/1000th of global energy production. Or the output of one nuclear power plant, or the same as all energy lost to friction for every car in the world. That really puts it into perspective.

Much of the mining is done next to hydro plants in China.

Still, there is a fixed amount of hydro power available, short of building new dams. This puts Bitcoin's toll on fossil fuel.

(If Bitcoin didn't exist, the hydro power would still be there and would be used for something else, therefore less fossil fuel would have to be burned in its place.)

> If Bitcoin didn't exist, the hydro power would still be there and would be used for something else

That's not a good assumption. There are many power plants in China in the middle of nowhere unconnected to much. It's completely plausible that the entire bitcoin network could run on clean energy that otherwise couldn't be used for much else because it's too remote (geothermal in Greenland or something).

I hope we're not building hydro power plants for the sole purpose of supporting the Bitcoin network. The energy may be "clean", but these projects have their own impact on the surrounding ecosystem.

So far they're built next to pre-existing ones and using excess power

There's huge energy oversupply in Sichuan province, you can't just keep storing hydropower infinitely, it has to be released to protect the dams.

All this outrage is misguided, the market incentives mean miners go to where the cheapest power is, all the industrial scale mining is located next to large renewables sources in China, Pacific NW, Scandanvia/Iceland.

The largest ethereum mine in the world is 100% renewable.

I agree. I've been racking my brain on ways to connect proof of work to something useful (like folding proteins, general purpose distributed computing, etc.)

There's primecoin, which uses a proof of work based on finding a special kind of prime number, which (they claim) has use in the outside world, at least the academic one.


Generally, to have a useful PoW, it needs to meet some criteria:

1) Easily verifiable

2) Necessary work on the problem is easily quantifiable and estimable

3) Problem can be programmatically generated from random data.

One model might be NP-complete problems. Users feed in problem instances they want solved, with a bounty. You find out how many guesses it should take. A valid proof of work then requires solving enough such problems, combined, to exceed the difficulty threshold. (You'd also need to solve one based on the current known transaction history.)

The problem there is that it's hard to know if a given NP complete problem instance is "one of the hard ones" and to find such instances.

There's Gridcoin, which, if I understand correctly, is essentially the "coinification" of BOINC:


It would be great if it ever became profitable to mine this.

Do that and it would no longer be useful as proof of work. Proof of work-i-was-going-to-do-anyway is not useful for securing the ledger.

That's incorrect. All that matters is you can't game the system by taking shortcuts or deciding what the problem will be beforehand; the work cannot be even partially computable without knowing the previous block.

No, you are incorrect. The important aspect of proof of work is that it is very expensive to create an alternate history (maybe one where you didn't send $15,000,000 to the exchange, for example). If you are doing useful work, someone can pay for it, and you can use that payment to make an alternate history for free, defeating the security assumption of blockchains.

Blockchains only work by probably wasting in order to build a history. If you aren't wasting, then it wasn't expensive to build the history (it was subsidised), and it wouldn't be expensive to build a different history (that one could also be subsidized)

But "socially useful" does not imply saleable (to the bane of venture capitalists everywhere).

It's possible for something to provide a social benefit but not make sense for anyone to pay for as a for-profit business because others would get the same benefit and free ride off them. (Hence the problem of public goods in economics.)

So if the proof of work is only solving some general research problem, then it might not be feasible to double dip like you've described. That's why primecoin was able to avoid the problem.

There are very few things that are actually useful (socially or otherwise) but don't provide economic incentive. As pointed out by others primecoin isn't really a very good example. A good application to shoot for, if there are any budding cryptographers in the audience, is a ticking timelock encryption. A set of keys known in advance, and each block solves the next key. This would let you encrypt messages into the future, where the cost to decrypt is equal to the cost of minting that many blocks.

I've seen partial schemes for this, but no complete ones that would justify implementation. If a full and workable scheme could be devised, I would bet on it replacing sha256 based proof of work, eventually.

Every public goods problem is mathematical research is a case where the solutions are socially useful but not saleable. The problem (pointed out by the sibling) with primecoin was that it worked too well, such that further solutions were uninteresting and past the point of diminishing returns.

Edit: See my here for a model of how NP complete problems could fill the role of you could predict difficulty in advance. There wouldn't be double dipping because the person wanting the answer wouldn't pay both bounties.


No the problem with primecoin is the mathematical "problem" it is solving isn't scientifically interesting.

Primecoin also ultimately had limited utility. It found the world's largest prime number thousands of times, continually beating it's own record. At some point we had more prime numbers than had mathematical utility.

My unprovable guess is that any other non-sellable problem will ultimately have the same outcome. It's interesting and useful for a bit, but when you have a billion dollars of hardware grinding nonstop for years in a row, you eventually exhaust everything interesting about the original problem and then it's back to the same waste you started with.

Doesn't that mean it worked too well, and we just need to make a Coin with harder and more useful problems?

It doesn't matter if the work is profitable because you only need to create an alternate history for a few blocks.

If you attack the network you can steal billions - so if you have that much computing power available for your useful work, you can temporarially use it against the block chain.

As long as that mining power is available anywhere the chain isn't secure.

But there is no known function that has these properties and is useful in real world.

Sure and that puts it in the same camp as PoS and other currently unworkable schemes. A possible way forward; we don't know yet.

The energy expended is the security of the currency.

If a transaction is confirmed by 1 TWh of proof of work, that much energy is required to alter it, meanwhile the bitcoin network has expended even more energy.

It is still young, and the exponential growth in energy consumption is going to stop as an equilibrium is reached.

It's only a temporary situation. There's a lot of research going on into replacing "Proof of Work" blockchains with "Proof of Stake" which doesn't require any mining. Ethereum is likely to get Proof of Stake fairly soon. Bitcoin and other cryptocurrencies will likely follow if it works out.

Energy is whats backing Bitcoins value. You need a lot of energy to create this valuable document - block

That's a good point -- in earlier times slaves worked the mines to extract gold from raw earth. Gold was a good store of value because of its scarcity. Likewise, Bitcoin mines the electrical grid.

Are you arguing that using the electricity you paid for is immoral, or are you arguing that using slaves is not immoral as long as you use them for something good?

I don't see how your analogy applies.

Slaves are unimportant to his point. Gold being scarce means it was hard to counterfeit gold coins in a time when you couldn't mint coins in a reliable way. Likewise, the purpose of bitcoin mining is to verify transactions in a way that is hard to counterfeit, making it more difficult to get a majority share of mining. The difficulty enforces scarcity.

Sorry, I thought it was a thinly-veiled moral argument equating bitcoin mining to use of slaves. Glad I was wrong.


This is completely true; miners will mine the coin that gives them the best ROI. If Bitcoin's value decreases without an equal decrease in difficulty, the whole thing could crash quite fast.

Energy doesn't back bitcoin, it just stops the currency from inflating. Or it would, if mining could always produce new bitcoins, but IIRC its supposed to stop after a while.

The difference being that a bitcoin does not entitle you to some unit of energy or work produced from it. The energy used to verify bitcoin transactions is just lost.

Energy backs bitcoin in the sense that "cheating" (reversing transactions or double spending) costs inordinate amounts of energy.

"backing" a currency usually refers to exchanging fiat money for a good with intrinsic value. That does not describe the relationship between bitcoin and electricity at all.

Energy ensures bitcoin's ownership, not its value (although it does impact its value indirectly).

The more energy Bitcoin network uses, the more valuable Bitcoins are... If Bitcoins value falls down, energy usage will fall down too. Energy is used to create valuable document - The Block. And block is not necessary new Bitcoins, its transactions too.

As the network hash power increases that value will only go up.

did you mean price will go up? either way it doesn't offset that high cost of use, so if humans here still value efficiency they might find it worthwhile to switch tools.

Energy usage rises if Bitcoins value grows. Miners are mining to get profit. If they will use energy and don't get profit, they will stop mining.

Other way round: energy usage will go up if price rises.

You right

> Electricity consumed per transaction: 165 kWh

Thats about $20 in the US.

If a transaction is that expensive, how can this system even work for transactions with a value of that order? Are large transactions "sponsoring" small transactions?

What happens if more people use the system for small transactions? Will BTC become unusable?

Yeah, that's inaccurate. The major portion of that energy expenditure is keeping the bitcoins that aren't moving secure.

Put another way, this is saying it is estimated to take about $800M it keep $40B of bitcoin secure.

This is probably a bit expensive in terms of comparing to fiat currencies (But then we are probably not covering all the real costs of those currencies)... but bitcoin is still in the technology adoption life-cycle and has not yet become a currency.

When the inflation of BTC has declined and it has become fully adopted (assuming this happens) then the numbers will likely start to make a lot more sense.

Also, the vast majority of transactions in about 2 years will likely be happening off chain using lightening network and the like.

Those transaction fees will happily be covered by merchants who are currently paying %2-%5 per transaction to Visa et. al. and will be more than happy to pay %1 or less to Bitcoin to manage a lightening channel.

Really, it will be companies disrupting Visa building payment networks out of lightening channels and they will have to keep fees really low because LN is open source and anyone can compete with them.

Prices are set by what the market is willing to pay, rather than what the service costs to run. Visa, et al, charges what it does because most merchants are willing to pay it, evidenced by the relatively few "cash/check only" businesses out there.

This doesn't mean that merchants wouldn't switch to a lower-cost payment system if it were available and there were enough customers using it, but it does mean that you can't just assume that Visa, et al, would not lower their prices if they felt the competition was serious enough. What their actually cost is, in such a circumstance, is harder to say. Their basic payment processing cost is probably quite low, even accounting for the people who maintain it, but they also provide fraud protection and frequently kickbacks to customers (cash back, airline miles, points); whether they could switch to chip-and-pin exclusively (in the US) and severely curtail kickbacks while still offering a desirable service to customers as well as merchants...

[Their credit services are presumably well-funded through their high interest rates.]

Large transactions are not sponsoring small transactions. Transactions are priced by their size in bytes, so the monetary value of the transaction is totally irrelevant to the fee.

People holding bitcoin are essentially sponsoring all transactions, if you want to look at it that way, because miners are making money both from fees and from new bitcoin that is created, which dilutes the value of all existing bitcoin.

So, if you look just at transaction cost the system clearly cannot sustain itself. But as more people buy into BTC, the system keeps from collapsing. If you ask me, that's starting to look like a Ponzi scheme in disguise ...

Check your premises-- your transaction cost is inaccurate.

Check jaekwon's comment above.

That sounds like the opposite of a Ponzi scheme. You're saying it becomes more sustainable as more people participate.

Wow. Nice rough math! Thats ridiculous. Hopefully Ethereums POS (proof of stake) system will fix this.

Check out Tendermint and the Cosmos Network.

A transaction here is a block. A block averages 2000 or so transfers, so that $20 becomes $0.01 for any individual transfer in the block.

Electricity price in the US is also irrelevant. As the article shows just over 0% of blocks are mined in the US, or a rounding error. Bitcoin mining chases cheap electricity.

No, a transaction here is referring to a transaction. The block reward on Bitcoin is already 12.5BTC is 12.5 × $4000+. That incentives the expenditure of significantly more energy than $20 worth.

You're right. Sorry. After re-reading the article I see the entirety of my comment is wrong.

I could have sworn there was still sizeable Bitcoin mining done near the cheap hydros such as in Washington State(around 3c/kWh).

Sure China and Ukraine is cheaper but not that much.

The statistic that stands out to me from this article is that 1 bitcoin transaction uses enough energy to power 5.58 us households for a day.

I'm hoping someone more knowledgeable about bitcoin can comment - is it likely that this will continue as the mining reward decreases? How expensive will transactions be after that happens? And if smaller rewards reduce total mining and power consumption, how vulnerable does the blockchain become to attack?

What i'm wondering is if transaction costs in the 'end state' of bitcoin can be competitive with centralized competitors like credit cards, paypal, etc. given this level of power consumption?

    I'm hoping someone more knowledgeable
    about bitcoin can comment
I did and was quickly voted to the bottom.

    bitcoin can be competitive with
    centralized competitors like credit cards
In it's current form it can't. It's already running at max capacity. No further increase in transactions per day is possible.

If there will be a cryptocurrency that can handle as many transactions as Visa, it will be something else then what we currently call Bitcoin.

Lightning network allows trustless instant and low fee payments. It is at the embrional stage yet, and the whole structure has yet to be defined, but the point to point protocol for trustless off-chain instant transactions is ready.


Correct. Glad someone has heard of it at least on hacker news.

So, the block reward is indeed increasing, but two other things are increasing very fast. First is the price per Bitcoin, and second is the number of fees per block. Just the day before yesterday miners received 1800 coins from mining and an additional 500 from fees. That's already over 20% of the total daily reward. This two things are more than compensating the loss of block rewards (in BTC terms). Things aren't expected to improve based on this.

It's kind of insane if you compare to other payment services (if that's how you use BTC). Not only do you pay $5-$10 per transactions, but that transaction also consumes 20,000-30,000 times more energy.

Well, transaction costs are already super high. I made a transaction yesterday for $5 and it was single input/output. But it was urgent, so I paid generously.

As the network grows, the demand for transaction grows too. The fees will not increase because the reward decreases. Simply miners will stop mining as it is no longer profitable. But this will happen in 2090-2140. So it is very far.

I think in the future, lightning networks and centralized wallets will be the way to go. The blockchain will be used by only a few big guys and transaction costs will be over $100/tx.

Well... Right now we have blocksize limited to 1 mb. If we will remove this limit and Bitcoin becomes a global currency, we can earn a lot from transaction fees. nChain estimates about 0.05 for a transaction if it will became us big as VISA. If you interested where is one youtube video talking about possible fees.

Long term outlook for mining isn't good. Ethereum, for example, is moving toward Proof of Stake (https://github.com/ethereum/wiki/wiki/Proof-of-Stake-FAQ), and Bitcoin will become less profitable for miners soon, beginning with the activation of Segwit.

The reason for the recent Bitcoin Cash fork, was that miners are terrified of losing control given the substantial investment required to maintain profitability. I don't think Satoshi ever thought mining would become so profitable, centralized, and political. If Satoshi had, he or she may have rethought the mining aspect of Bitcoin.

It's the end of days for mining as we know it.

Miners can choose to include more profitable transactions, skiping segwit ones.

POS will not work. you need to back Bitcoins value with something, and its electricity. Gov backing with gold.

Incorrect. Bitcoin isn't "backed by electricity". You can't turn Bitcoin back into energy. The energy is used to secure a distributed ledger to prevent double spend attacks on ongoing transactions, in a really inefficient way, but don't call it "backing".

PoS will succeed. It already has, we've set all the pieces. The thing is, you don't want to live in a world where Bitcoin and PoW dominate our financial transactions. That world is not livable.

Bitcoin transactions are secured by spent electricity. PoS transactions are secured by cryptographic signature schemes and reputation of operational security. It's how the world already functions, just more replicated.

It takes a cryptographically provable amount of energy to reverse a Bitcoin transaction, and that amount is equal to the money spent on electricity mining blocks that confirm a transaction.

Proof of Stake systems have no equivalent. They operate by trusting the staking parties are not colluding to create an alternate history. But if they do decide to collude, there is little thermodynamic cost to rewriting history.

It's a fundamentally weaker system, and one that depends on trust. The whole value of proof of work is that it allows you to escape trust.

PoS as well as PoW depend on the assumption that a majority (either in terms of value invested in hashing power or invested in tokens) of the network is following the rules. If you try to double spend in a PoW system you are betting your current electricity usage on a new block becoming the longest chain and in a PoS system you are betting the tokens themselves - from an economic point of view those two systems are indifferent (if your attempt doesn't succeed you have wasted either electricity or tokens). Collusion can happen in both cases. PoS systems introduce new (probably yet unsolved) problems but they aren't inferior from an economical point of view imho.

No, in a proof of stake system collusion to rewrite history is essentially free.

In a proof of work system, even if you collude you still have to spend a non-trivial sum of electricity to create that alternate history. Electricity that would be making you money via the block reward if you were using it honestly.

It is provably more expensive to reconstruct history in a PoW setting, even when everyone is happily colluding.

Whether it is more expensive or not is irrelevant. If a majority of the network (either in hashing power or number of tokens) decides to collude and start modifying network states for their own benefit then they will ultimately succeed. You argue that in a PoW system rewriting older blocks is more difficult than in a PoS system but that doesn't matter, because when a majority of the network becomes hostile and starts blocking all of your transaction attempts then the value of your coins immediately goes to zero. In a PoW system the records of your older transactions might be around forever but that's little more than damage control.

In such an adversarial environment proof of stake even has advantages over proof of work. If in a proof of work system a majority of the network starts to double spend or censor transactions, what can you do? You are essentially powerless because any fork (that tries to fix their malicious blocks) can be quickly taken over by the colluding miners. However in a proof of stake system you could simply fork once and destroy/invalidate all the tokens of the malicious miners.

If you are resorting to a hardfork to manage malicious actors, you can always change the proof of work algorithm, which would damage the malicious miners to a tune of billions of dollars in bitcoin.

Rewriting a year of history in a proof of stake system takes a couple days on consumer hardware. Rewriting a year of history in Bitcoin is going to require burning a literal hundred million dollars of electricity, if not an order of magnitude more.

In a proof of stake system, typically less than 30% of participants by coin amount actually do the proof of stake. And that's for systems that don't require you to bond coins for months. Exchanges frequently control that many coins. And if an exchange abuses their power, are you going to burn the coins of every single person using the exchange?

It gets worse than that though. You don't need current coins to attack the system. You only need old keys of coins that you sold. After you sell your coins, why wouldn't you sell your keys to an attacker too? If the attacker is paying for the keys, and you sold the coins anyway, there's nothing to lose for you.

Or an exchange can rotate their coins to new addresses and use the old coins to attack the system.

How is it more difficult to rewrite history in pow? Maybe only if you want to stick to the same exact algorithm going forward. But a modified node can do whatever it wants, including rewriting the blockchain from the beginning to use a different hashing scheme, making custom exceptions for individual block numbers, etc. It all depends on how the modifications align with the incentives of the coalition and user base.

51% coalitions can also create alternate histories in PoW by agreeing to run a modified version of the client. There is negligible thermodynamic cost to rewriting history when you control a majority of the nodes.

Pow is really an incentive system to keep selfish miners flocking to one chain, by making betting on one chain more profitable than betting on lots of chains at the same time.

The PoS challenge is to do the same without the egregious amounts of wasted energy, by making miners bet their money directly, instead of indirectly through their computers and electricity.

Running a majority of the nodes does not give you the ability to arbitrarily decide to accept an alternate history. One of the big value adds of bitcoin is that every node can decide on its own, without consulting other nodes, whether a transaction is valid or not.

There is absolutely a thermodynamic cost to reversing a transaction in bitcoin. Having the miners run an alternate client doesn't change that.

Just another thing people went wild for without thinking about externalities. You don't need to be a genius computer scientist to realize a thing that requires a constantly rising amount of processing power just to keep pace will eventually need an unreasonable amount of power.

Bitcoin does NOT need constantly increasing computing power. The hashrate and difficulty level adjust up or down as needed.

It's a temporary situation. Proof of Work which requires mining is likely to be replaced by Proof of Stake and other methods which don't require any mining. Ethereum is planning to try Proof of Stake in the next year or so.

Humanity needs a constantly rising amount of processing power to keep pace.

The energy expended is a guarantee of security and immutability.

The fact that real world energy is expended is the thermodynamic guarantee of the irreversibility of a transaction.

If you have access to the VISA database, and change something, you cannot have the cryptographic proof that it has not been tampered with. At the very best, you trust a VISA root key.

With Bitcoin you trust math and physics, telling that if the whole network has expended 1TWh after my transaction, at least twice that energy will be necessary to reverse it, and the only data i need to prove that, is the blockchain.

I get that the energy is being used for something useful and cool, but when the energy cost to secure a single transaction could power 5 households for a day, then something is wrong and needs to be fixed.

Surely even fans can agree there's a level at which it just doesn't make sense any more.

That energy is not just confirming the transactions in the existing block, it is protecting all transactions in the blockchain history, adding security to each of them. That represents a history of more than $70 billion in value. Suddenly it doesn't seem so expensive anymore.

That's actually revealing.

First people try to argue that all this gigantic energy expenditure "buys" security.

And then – in the fine print – you are learning that you don't actually "buy" security, you have been sold a subscription. This gigantic waste of energy doesn't do anything unless you follow up with even more energy.

So the calculations that x amount of energy "secures" y amount of economic value is flat out wrong. Because x should be x + x' + x'' ad infinitum.

Oh, and the longer you stay in, the more difficult is getting out, because "sunk cost" isn't a fallacy in this scheme, it's actually real.

Yeah, the inflation is a tax more or less. It's a tax on your holdings to pay for the security and integrity of the network, and to make sure things keep moving forward.

Bitcoin is nearing the point where most security will be paid by fees instead of inflation, but imo inflation is the more fair model to pay for security.

Yes, there is a level at which it doesn't make sense any more. In a sufficiently efficient market, that is the point we are at, at any given point in time.

I believe the bitcoin mining market is efficient enough already that we're more or less there.

That shows the primary failings of putting all trust in an efficient market. The "makes sense" breaking point in an efficient market is when it stops making financial sense for the miners. It relies on rational miners who don't don't succumb to the sunk cost fallacy after investing large sums upfront in mining rigs. While it also ignores the negative externalities involved in increasing global electricity usage like climate change.

Yeah, I hadn't really considered how simple economics essentially forces this to grow to exactly the point where the externalities are unsupportable based on the benefit provided.

> something is wrong and needs to be fixed

It's already fixed with PoS... IMO the winner of the cryptocurrency-race will be a PoS-based coin.

PoS deserves research but it's far from shown itself to be viable as the dominate consensus mechanism.

A lot of research was done by heavyweights for PoS in 2012 and 2013. The consensus was that it was a fruitless pursuit and that PoS could not match PoW in terms of security and decentralization.

Several papers have been written about it.


Well, as you may know it's currently being used in some coins like bitshares [0] and it works fine, Ethereum also is migrating to PoS [1]. So the consensus for a lot of people in the ecosystem is that it works.

[0] https://bitshares.org/technology/delegated-proof-of-stake-co...

[1] https://www.coindesk.com/ethereum-casper-proof-stake-rewrite...

I'm just trying to not be absolutest by stating it can't work. I'm in the camp that believes it's just obfuscated PoW.

In addition to math and physics, you also trust the durability of infrastructure, the correctness of the software, the moral compass and cohesion of the developers, that there will be no 51%+ mining coalitions, that if there are they'll be benevolent, and probably believe in relative global peace.

You also trust totally mundane things like internet resilience against phishing attacks, web browsers and other basic tools, the https standard, dns and certificate authorities, etc. etc.

And this trust in the developers and the system has already been broken (for benevolent reasons, but still):


Please note "The right people were online and available in IRC or could be contacted directly."

And of course you trust your own skills keeping your devices, passwords, keys, totally secured from outside access, and not making the zillion tiny userland mistakes that get your Bitcoin lost or stolen.

Yes we all understand this. The question is is it worth it? BTC offers some clear benefits over existing money transferring systems, but is also has clear drawbacks. Simply restating some benefits doesn't close the issue.

*at least an equal amount of energy is needed to reverse it

Spending 800M$ a year in electricity alone to run a digital currency is massive and at 8$ per transaction seems incredibly inneficient. The comparison to the VISA network is incredible. Independently of your opinion of the value of Bitcoin in the long term the current network is definitely not more useful to society than the VISA network.

800M is not to run the network, its hard money. If 1 mb limit was removed today, this network can handle loads of VISA and more. Electricity is spent to create new blocks... Hard work is done to create new Bitcoin and verify transactions.

BECI is flawed, I wrote a critic of it some months ago: http://blog.zorinaq.com/serious-faults-in-beci

This motivated me to research and publish a more precise energy comsumption estimate, and I have been published in Bitcoin Magazine: http://blog.zorinaq.com/bitcoin-electricity-consumption https://bitcoinmagazine.com/articles/op-ed-bitcoin-miners-co...

I always like to remind people that Bitcoin miners consume about the same amount of energy as Christmas decorative lights in the US. Kinda puts things in perspective...

> I always like to remind people that Bitcoin miners consume about the same amount of energy as Christmas decorative lights in the US. Kinda puts things in perspective...


The first piece is merely a defamation project and contains a lot of false information. You forced me to publish a statement (and release the emails we exchanged) on that here:


The second piece contains a lot of cherry picking that I asked about in these emails as well, but never got a response to (instead my emails were quote mined to support those false statements).

Nothing else to say to you. Everybody else is welcome to ask questions and discuss as usual.

Apparently "terawatt-hours per year" is a unit that people use in public policy, but it strikes me as strange. Converting 16.22 TWh/year to a "more metric" unit, that's a sustained power usage of 1.85 gigawatts.

So Bitcoin is constantly using more than enough power to make Doc Brown's eyes bug out.

I too hate seeing __ watt-hours per year. What's the point of the double time conversion? It's like saying "The movie starts in 3 kilometers per mile-per-hour"

I mean, I understand why, it's because often you want to relate it to cost, and electricity is priced in $/KWh, not in ($/sec)/watt.

I do wish energy were priced per megajoule. But I suppose this is probably a symptom, like miles-per-hour vs meters-per-second, of the non-metrication of time units at the relevant scales. (It takes order of an hour to drive somewhere, and order of an hour to run a laundry machine.) If we had a metric hour of a thousand seconds, a kWh would just be a MJ, no problem. And if we had a "metric year" of a million seconds, a "TWh/yr" would just be a GW. Unfortunately, it's pretty important to humans to track time in a way that lines up with earth's rotation and revolution, and there's no way to make that metric. (If French Revolutionary Time caught on, a second would be 1e-5 days instead of 1/86400 and an hour would be 1/10 of a day, and that would be a start, but the year/day ratio is pretty intractable.)

There it is again though! A Watt is a Joule/sec, so you shouldn't even use Watt-seconds as a unit. You mean $/Joule.

That's an arbitrary price though. Instead of 10¢ per KWh, we could price it as 36¢ per megajoule.

It's the same thing.

This is one of the reasons I prefer proof-of-stake schemes, despite the complexity, low power consensus is preferable given roughly equivalent tamper resistance. State collusion with mining pools is a credible threat to proof-of-work schemes.

Proof of Stake is just Proof of Work with hidden work.

Miners will crank whatever handle they can crank to increase their payout, up to the point where they spend more marginal effort cranking the handle than the corresponding marginal payout is worth. In PoW (in bitcoin) that handle is SHA256. In Proof of Stake, it's more obscure, but they will find it, and they will crank it, and you'll just be back where you started.

I'm not convinced PoS is a practical improvement on PoW.

This is a misleading argument. Yes, miners will always spend more until marginal revenue == marginal cost. The problem with the proof of work system of bitcoin (and many others) is its inefficiency that only allows for a low transactions throughput. Proof of work has no inherent value for the system, it's simply a way to cap the network throughput to about 1MB per 10 minutes because anything lower than 10 minutes will make the network increasingly vulnerable to adversarial attacks [1]. There's got to be a way to increase network throughput beyond this while maintaining censorship resistance and decentralization.

[1] C. Decker, Information Propagation in the Bitcoin Network: http://www.tik.ee.ethz.ch/file/49318d3f56c1d525aabf7fda78b23...

The point is expend work to create signatures, creating a crytoeconomic cost to state changes.

The purpose of PoW is to reduce the number of forks and also making double spending attacks very expensive. Why can't the computing power of the entire network be used to just verify the integrity of all transactions and skip the nonce finding part? This would immensely increase network throughput. The reason is because Bitcoin's PoW implementation can't provide the same level of consensus and security with a higher number of transactions. Newer PoS systems might.

The problem with POS is that you do not have a mathematical guarantee of immutability. It costs nothing to try to stake on different blocks, and every counter-measure relies on the fact that the network is not partitioned.

This would be fatal in case of a remote crash vulnerability, because an attacker could partition the network at will.

Or a nation/state could cut internet access.

Without real world energy, you lose byzantine fault tolerance.

The only proven countermeasure so far is delegated proof of stake, which has "trusted" parties doing the staking. But this goes against the idea of decentralisation.

Some people wonder if pure proof-of-stake is even possible. Ethereum is likely going with DPOS, yet to see a single solution to the nothing-at-stake problem.

PoW needs to die.

PoS can be superior to PoW in every way except the problem of initial wealth distribution. PoW solves this problem elegantly. PoS currently has no meaningful alternative to this distribution method.

What Ethereum is doing with using PoW to bootstrap wealth and then transition to PoS looks like a winning combination.

So the calculation is incredibly sensitive to the assumptions, that miners spend 60% of revenues on electricity and that they have an average cost of $0.05 per kw-h.

Anyone have any sense of how accurate either of those is?

Just want to note that 60% is the expected future (ultimately), but the current % is more like 22 (found by taking the total cost # divided by the total annualized mining revenue #).

So in that case it's only ~2 houses rather than 5.58? Is this your website? Do you think the figure should be made more accurate?

Ah no, it's 5.58 for the current 22%. If it actually gets to 60% it would be much worse.

If you aren't going to disclose your actual assumptions and methodology you should probably delete all the text that purports to do that.

Will save people time.

Huh? It's all in the infographic. 60% is the target, but "production takes time" (see content here) hence the "ultimately". I'm not hiding the current number, it's in the key statistics.

If you're interested in where the 60% comes from in the first place > it's based on the lifetime costs of a number of machines. It was on the page a while ago, but the index page itself isn't a research paper so I removed it. I'll give it a new spot soon, probably on this page (already containing the foundation of the method):


If you just want to see some real-world numbers, check out my break-down a farm recently:


(Note: subtle difference between part spent on elec & part spent on ongoing costs. I take costs and then 1 kWh for every 5 cents spent, so including overhead.)

Okay, I'll rephrase: An explanation of where each number in the table comes from would be more interesting than the current text.

It would also be useful to separate the table so that one contained the information meant to be conveyed by the index and another contained the impressive comparisons to countries and households and stuff.

... Speaking of environmental externalities,

In the US, "Class C" fire extinguishers work on electrical fires:

From Fire_class#Electrical:


> Carbon dioxide CO2, NOVEC 1230, FM-200 and dry chemical powder extinguishers such as PKP and even baking soda are especially suited to extinguishing this sort of fire. PKP should be a last resort solution to extinguishing the fire due to its corrosive tendencies. Once electricity is shut off to the equipment involved, it will generally become an ordinary combustible fire.

> In Europe, "electrical fires" are no longer recognized as a separate class of fire as electricity itself cannot burn. The items around the electrical sources may burn. By turning the electrical source off, the fire can be fought by one of the other class of fire extinguishers [citation needed].

The reward for mining a block will go down over time. In the end, all mining will be financed via the transaction fees. Then the market will figure out how much energy should be used per transaction.

Yeah... doesn't seem promising to me. When transaction fees need to pay for mining / transaction validation, people will probably stop using Bitcoin for purchases they could make with other methods of payment.

It's the other way round. The energy consumption is defined by the fees people are willing to pay. Mining will always be profitable. No matter how low the fees are.

Does that justify wasting enough electricity to power Dallas, Texas in the mean time?

Obviously it generates enough value for some people that they are willing to pay for it.

Watching television probably uses up even more energy. And I would argue it has a negative impact on life quality. But who am I to decide what people should value?

Incandescent bulbs provided light and people were happy to use them.

They were outlawed for the simple reason that the problem they solved wasn't efficient compared to other ways of solving the same problem.

The same thing will happen to cars with internal combustion engines.

People may value the sound of combustion engines or the warm glow of incandescent bulbs but as a society we don't care what they value since we share the effects of externalities.

Watching television does not use up more energy than mining bitcoin, that's the point of the linked article.

The article states that the "electricity consumed for a single transaction" is equivalent to what 5 households use up in 5 days. Which seems to make you think that one would "use up" less energy by watching TV then by doing a Bitcoin transaction. But that is not true.

Making a Bitcoin transaction uses up hardly any energy.

The miners mine for the block reward. That is completely unrelated to transactions. They even mine if there are no transactions and generate empty blocks.

If nobody made a Bitcoin transaction for a month, the energy consumption of that month would stay pretty much the same.

It's not "wasted". The miners wouldn't spend the energy if they weren't getting a good return on their investment.

Just because you don't like bitcoin doesn't make it a "waste" of energy, any more than mining gold is a "waste" of energy if you don't like gold.

You're thinking of waste too narrowly. That power wasn't available for other things, so more had to be produced to cover Bitcoin and the other things, which meant more of all the negatives that come with energy production. Does Bitcoin provide something valuable enough to human civilization to account for that? Could that value have come from a more efficient source?

Failing to account for externalities is how we get rapid ecosystem destruction (before even bringing in climate change). Those Bitcoins aren't innocent little bits on drives. They have a cost to the planet they're produced on and the life that inhabits it.

If you think you can replace bitcoin with something more efficient, please do so. The millions of people who use bitcoin every day would gladly switch to a cheaper version if it was truly just as good.

I don't take it as a given that Bitcoin is a good solution to the problem those people solve with it. The best of the bad solutions is still bad, and Bitcoin isn't looking too good by its power consumption. It's okay to stop, really think about a problem, and find a good solution instead of charging ahead with a bad one oblivious to the brick wall ahead.

What do you think is the best way to store+exchange value across borders in a way that can't be censored or reversed by any third party and doesn't require trusting or asking permission from any third party?

Personally, I think the answer is bitcoin.

Made an account just to upvote your comment. This is an incredibly powerful point that appears to be completely missed by the majority of threads I'm seeing. Maybe PoS will turn out to be effective and reduce energy costs for cryptocurrencies, but there is absolutely nothing like them. The anonymity of money is valuable for any free and open society, and Bitcoin is currently providing that to anyone willing to create a wallet.

Who told you bitcoin was anonymous?

"Wasting" energy is required to make PoW work, but if we could achieve the same with PoS (and that seems to be the case since ETH is switching too) then it would really be a waste.

If gold could be found without excessive energy use then it would be a waste to spend a lot of energy digging for it.

If there was a good low energy alternative to PoW then it's a waste of energy to use that method of validation.

Like incandescent bulbs are a wasteful way of making light when there are more efficient ways.

What the article fails to take into account is the huge volatility of BTC relative to the USD. So if you are going to make this calculation over a whole year then you have to split it up into many small calculations (for instance, per block with the going rate of BTC:USD at the time) and then accumulate in USD.

On another note: a very large fraction of all bitcoin mining is happening in China and electricity cost there is quite different compared to the USA.

So I seriously question the prime assumption leading to the quoted energy consumption.

It's such a waste of energy in a planet already burdened with climate change we worsen annually. Greed is crazy. What this is wasting is painful to see.

This is a stupid comparison. The number of transactions can be increased by a factor of a 1000 easily while holding the amount of energy expended constant.

If it's so easy, please describe how.

The how is easy to explain with just the name of an implementation/fork: Bitcoin Cash.

The reason it has not happened for original Bitcoin is internal politics. If the number of transactions are limited then they are more profitable.

I do not believe that Bitcoin Cash is 1000x more efficient than Bitcoin. It is an extraordinary claim and requires extraordinary evidence.

It is not, each block in Bitcoin Cash can only hold 8x more transactions.

Also, and more important: it is not necessary to for it to be 1000x more efficient, because no cryptocurrency currently has such a big amount of transactions per hour. It would be more wasteful to support 1GB blocks when they would go empty.

But, and this is actually my claim, if volume transaction increases enough to saturate a Bitcoin Cash block, then they will increase block size, as many times as it is needed, as that is the proposed Bitcoin Cash strategy. In stark contrast with the Bitcoin 1MB blocks forever strategy.

(I am assuming you don't think efficiency is measured in watts per transaction, because if you do, /facepalm)

>Number of U.S. households powered for 1 day by the electricity consumed for a single transaction: 5.58

The one that surprised me the most

It's a bad stat. The electricity that goes into a block doesn't just confirm the transactions in the block, it also confirms all the transactions in all blocks prior to that block, and those transactions have a cumulative value of about $70 billion.

The bleak picture painted on the article is an incomplete one by far.

Why is it a bad stat? Is having to confirm all transactions in all prior blocks not a direct result of a transaction? Why is the cumulative value relevant? I can see how there is a perceived value there but if the transaction didn’t happen it wouldn’t have been needed?

If the block had zero transactions in it, it still would have cost just as much energy to create.

Similarly, if the price of the coin were 1/1000th but the block were full, it would have cost 1/1000th the price to confirm those transactions.

What determines the cost/mining reward of a block is not the transactions in it, but the value of the chain as a whole.

Ah got it. Thanks.

Why don't we compare Bitcoin's energy footprint with that of all the banks, remittance shops and associated infrastructure instead?

The PoW is quite expensive, but remember that mining is a subsidised activity amounting to billions of dollars worth of rewards for the miners, many of whom make extensive use of hydro-electric and solar power.

There's the question of wether we need all this hashing power to protect the billions of dollars worth of bitcoin market cap. The truth is that ASICs mining was not anticipated and neither was the "arms race" that made mining into a profession. With Proof of Stake consensus algorithms taking front stage in Ethereum, EOS and other blockchain projects, PoW is bound to be abandoned or significantly changed.

Both ASIC mining and the arms race were anticipated as early as 2008:

"[As] the network grows beyond a certain point, it would be left more and more to specialists with server farms of specialized hardware." https://www.mail-archive.com/cryptography@metzdowd.com/msg09...

"We should have a gentleman’s agreement to postpone the GPU arms race as long as we can for the good of the network. It’s much easer to get new users up to speed if they don’t have to worry about GPU drivers and compatibility. It’s nice how anyone with just a CPU can compete fairly equally right now." https://bitcointalk.org/index.php?topic=12.msg54#msg54

You're right.

What kind of unit is a TWh/year? Why not just use watts?

By the way, 10 TWh/year is about 1.21 GW.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact