Hacker News new | comments | show | ask | jobs | submit login
[dupe] New study quantifies Bitcoin’s ludicrous energy consumption (arstechnica.com)
77 points by fabian2k 3 months ago | hide | past | web | favorite | 150 comments



I provided this statement to some journalists who contacted me to get another viewpoint on Alex de Vries's paper:

His findings are not completely accurate, but also not completely wrong either. My own formal well-documented estimate is that Bitcoin used 0.09% of the world's electricity as of January 11th, 2018 (http://blog.zorinaq.com/bitcoin-electricity-consumption/). After 4 months of growth I estimate we are around 0.15-0.20% today. Therefore reaching 0.50% by the end of 2018 is plausible; not likely but plausible. That said, overall I think Alex de Vries's figures are rather inflated because he makes two errors. Firstly, his model assumes that manufacturers sell at cost and make zero profits, which allows miners to spend more on electricity. In reality we've seen Bitmain, the largest manufacturer, make billions of dollars over the last few years. No one expects them to make zero profits by the end of 2018. Secondly, de Vries makes the mistake of assuming that all cryptocurrency mining chips produced by TSMC on behalf of Bitmain end up in Bitcoin mining machines. In reality many chips are fabricated for cryptocurrencies other than Bitcoin. This altcoin sector is rapidly growing. It wouldn't be unreasonable to assume that Bitcoin mining machines will represent less than half of Bitmain's sales by the end of 2018.

(This is a copy of my comment from yesterday's thread on the same topic https://news.ycombinator.com/item?id=17084253)


> In reality many chips are fabricated for cryptocurrencies other than Bitcoin.

OK, so replace "Bitcoin" with "Bitcoin and other cryptocurrencies" in the title. The point remains the same: blockchain-based technologies are extremely inefficient with energy. That's the nature of "proof of work" algorithms.


«OK, so replace "Bitcoin" with "Bitcoin and other cryptocurrencies" in the title»

Changing the title would not resolve my main critic, which is that de Vries still overestimates the true figure (because of his assumption that manufacturers will make zero profits "by the end of 2018".)


Which algorithms are more efficient at doing what Bitcoin does?


You’re thinking efficient compared to another hashing algorithm. I think op is saying efficient compared to the value it creates. I.e. cryptocurrency is useful for some things, but it’s burning a LOT of energy to do it.


I'm asking what is more efficient at doing what Bitcoin does: censorship-resistant transactions with sound money.


I think the point here is that the need for such a currency is not absolute and may be outweighed by the cost of such a large energy consumption rate.


Cash?


What does Bitcoin (et al) do for someone such as I, who cares not a fig for speculative gambling?

- edit

this was a sort of response to your statement below "Don't get me wrong, I use VISA all the time, but it's a different product", but I couldn't reply to it direct.


It is an asset with a fixed issuance and censorship-resistant transactions.

Some people don't trust the authorities with the powers of issuance and censorship and find it interesting for those reasons. It is likely that you are not in that category.


Presumptuous. I've long had my own concerns about The System, but I don't think the fix is in a highly volatile and comparatively easy-to-disrupt electronic ledger.

Now, Gold OTOH...


"easy-to-disrupt"... no, that's the whole point. Prove me wrong.

Yeah, have you ever tried to send .002 gram of gold to anyone recently?

Leave your comfortable bubble and look at countries with hyperinflation like Zimbabwe/Argentina, how are they going to use a cash (not dollar) in a trustless fashion?


But they do use dollar, right - mano a mano?

If you think that crypto's hardier than real world fungible assets when the shit hits the fan, I imagine it's not me who is trapped in a 'comfortable bubble' as you so condescendingly put it.

Given now infamous recent disclosures about fundamental flaws in our trust in IT at the deepest levels we can imagine, I suspect you vastly overestimate the resilience of digital 'securities'.

Never mind power cuts or flat batteries...


Whatever VISA is doing is orders of magnitude more efficient.


* not censorship resistant

* issuance controlled by a monopoly

Don't get me wrong, I use VISA all the time, but it's a different product.


Bitcoin is neither of those things. Mining, developers and exchanges are all centralized. All a government that cares has to do to control issuance or censor transactions is regulate those entities.


do you really understand how transactions are recorded in bitcoin? Care to educate me how a government can censor a transaction?


One example that the US government's OFAC is considering is blacklisting addresses and tainted UTXOs [0]. Miners would be forbidden by law to confirm a transaction related to a tainted UTXO. This means if the transaction is spending a tainted UTXO or sending to a blacklisted address (and thus creating a tainted UTXO), it wouldn't be confirmed. "taint" in this case travels forward in a chain of transactions and is a well documented property if you google it.

Companies like Chainalysis have become really good at not only clustering related UTXOs but also monitoring taint as UTXOs are attempted to be obfuscated. [1]

[0] https://www.coindesk.com/goodbye-fungibility-ofacs-bitcoin-b...

[1] https://www.chainalysis.com

As an aside, the patronizing tone is unwarranted. It was acceptable five years ago when the jury was out on Bitcoin, but science is slowly revealing the inadequacies of the technology. Examples here [2] and here [3] and here [4]

[2] https://www.cell.com/joule/fulltext/S2542-4351(18)30177-6

[3] https://www.tse-fr.eu/sites/default/files/TSE/documents/doc/...

[4] https://arxiv.org/abs/1801.03998


A bit of paper called cash backed by the state?


* Issuance is controlled by a monopoly

* high transaction costs unless both parties are in physical proximity

* not durable

* not programmable


Tried and tested. Doesn't use a shitload of energy.


To each their own!


Serious question... has there been a bigger waste of electricity in any commodity trading or in any internet trend?

A the cost of a single bitcoin transaction (~300kW) could power a use home for more than a week [1]. How is this even remotely defensible by people who participates in coins and also make posts about climate change? Seems very hypocritical to me, because I’m pretty sure it’s not those “filthy denier types” that are wrapped up in the coin game.

[1] https://www.eia.gov/tools/faqs/faq.php?id=97&t=3


The question about how much costs a single bitcoin transaction is deeply flawed. The amount of transactions a bitcoin network can process does not directly depend on amount of electricity spent. Bitcoin network can work the same way using 10% of its current electricity consumption. Also, lightning network will allow for thousand times more transactions with same energy cost.

Why is everyone obsessed with bitcoin electricity consumption? I wonder how much energy is required by the banking system to operate(imagine those thousands of people burning gas in cars to get to their daily work at bank). Humanity spends vast amount of energy on much more useless stuff with just entertainment value(World football cup, all those fans coming from different parts of world in the jets, logistics...)

Also, energy is a resource like any other and it's generation and price is subject to supply/demand. The fact, that is is used on mining bitcoins just indicate, that people value bitcoin technology more than something else they would spend the electricity on.


>Why is everyone obsessed with bitcoin electricity consumption? I wonder how much energy is required by the banking system to operate

People care because Bitcoin is, by design, massively wasteful. The banking system's energy use can be curtailed with green tech. Bitcoin's entire ledger is powered by proof of work, which by definition requires computers to grind at maximum power calculating millions of hashes. And the difficulty goes up if more people are mining. And on top of that, the Bitcoin network is extremely slow and several orders of magnitude less efficient than other banking systems in terms of transactions per second.

Everyone is obsessed because we're wasting damn near 1% of the world's energy so nerds can speculate on Dunning-Krugerrands and it's not doing anyone any favours in the long run.


> People care because Bitcoin is, by design, massively wasteful. The banking system's energy use can be curtailed with green tech.

Still does not explain how the current banking industry along with others, such as the entertainment industry is not "massively wasteful" by design. Sure, maybe a small portion of the banking system's energy can be curtailed with green tech, but what about the 24 hour ATM, the light posts at the banks, the AC units in each branch, the massive skyscrapers in which all of the banks corporate offices operate? and the energy required to build all of that infrastructure? As stated already, there are countless number of things that "waste energy", and in some cases "by design". Check out "b.s. jobs" on wikipedia.


>Still does not explain how the current banking industry along with others, such as the entertainment industry is not "massively wasteful" by design

a) The entertainment industry is irrelevant to the discussion of Bitcoin vs traditional banking systems. It's a red herring. Cut it out.

b) Bitcoin is using an amount of electricity comparable to the country of Ireland. The banking system might use that much power, but they're doing millions of card transactions a minute - there's actual returns and actual business being done. And literally everything you described is an externality that costs the system money - ATMs exist because they're cheaper than human tellers, the lightposts can use more energy-efficient bulbs, etc etc. None of these are explicitly designed to burn power.

On the other hand, if you add mining power to the pool, it makes it harder to get hashes, using more power. Bitcoin is regressive in its power usages. The traditional system does not actively set out to use more power. While due to inertia it may not make active strides to reduce it, but that's not the point here. It's not by design like Bitcoin is.


> Sure, maybe a small portion of the banking system's energy can be curtailed with green tech, but what about the 24 hour ATM, the light posts at the banks, the AC units in each branch, the massive skyscrapers in which all of the banks corporate offices operate? and the energy required to build all of that infrastructure?

The point is, that waste is all upkeep. Something the system tries to minimize, as it bleeds money through it. If ATMs can be made more efficient, the system saves money. In Bitcoin, OTOH, waste is not upkeep but a feature, which the system tries to maximize.


> which the system tries to maximize

How exactly the system is trying to maximize waste?

In fact, miners did more progress on efficiency in terms of pure hashrate/watt with a million time improvement in past 10 years, than banking industry ever.

It is demand, that is driving the energy usage, not the system on its own. I agree with the person above, that it is basically justified by increasing consumption of the service.


>but what about the 24 hour ATM

:D how many hours of an ATM running do you think it takes to get to 300kW or 1 single bitcoin transaction?

I’m going to guess that a single ATM might not hit 300kW in a year.

Your sense is scale is way wrong.


You're right, I didn't keep scale in mind when listing examples. Didn't think it mattered since there are hundreds of relevant things that consume energy at different scales. Next time I'll avoid listing things that can be cherrypicked to make a relevant counter argument.


> Also, energy is a resource like any other and it's generation and price is subject to supply/demand

Actually most municipalities treat electricity like they treat water, more akin to a public utility. The actual cost isn't necessarily related to supply/demand.

Normally this isn't a big deal as historically the electricity use either was minor (homes) or lead to economic growth (businesses) similarly most businesses would make specific deals if they had heavy use.

Now Bitcoin threw a curve ball by allowing electricity -> money in a system not designed to make the cost of electricity support it.

Yes the fundamental idea of Bitcoin is great but the current external cost to support the "investment market" is quite high.


I mostly agree, except that the fundamental idea is bad because it inevitably leads to an arms race between miners. Using more electricity gives you a bigger share of the pie, but doesn’t actually grow the pie. So the electricity is really wasted. And this at a time when our energy demand is contributing to climate change.


> Bitcoin network can work the same way using 10% of its current electricity consumption.

Except by design, bitcoin miners are incentivized to consume more energy the higher the price goes. If they don't consume as much energy as they possibly can, their mining pool will have less of a chance to get that sweet, sweet block reward (....to convert back into an actual currency in order to, you know, pay their bills and stuff....)

> Also, lightning network will allow for thousand times more transactions with same energy cost.

If the solution to bitcoin's horrific energy consumption is to not use bitcoin.... what is the point of bitcoin?

> The fact, that is is used on mining bitcoins just indicate, that people value bitcoin technology more than something else they would spend the electricity on.

It is valued as a speculative asset and an excellent way to launder ill-gotten money. It's only use is for scams. Bitcoin generates a massive negative externality in that it drives up the costs of electricity for all the people who are doing actual productive things with it.

The whole thing is a abhorrent waste and supporters like yourself should be absolutely ashamed of yourselves for defending it (using some of the most paper-thin arguments, I might add...).


> If the solution to bitcoin's horrific energy consumption is to not use bitcoin.... what is the point of bitcoin?

The lightning network IS Bitcoin, just as using HTTP over TCP is using TCP.

> The whole thing is a abhorrent waste and supporters like yourself should be absolutely ashamed of yourselves for defending it.

It's the only way to survive economically in repressive states.


> The lightning network IS Bitcoin, just as using HTTP over TCP is using TCP.

So its got a decentralized, distributed public ledger recording every transaction made? Since when did that happen?

> It's the only way to survive economically in repressive states.

LOL. No. You use USD.... No citizens in economically repressive states are gonna flip to bitcoin. Not when it gets 4 transactions per second tops. Not when you have to purchase your currency from early adopters in order to participate. Not when you can lose your entire savings by simply exposing your computer to the internet.

Sorry man. You fell for a scam.


> So its got a decentralized, distributed public ledger recording every transaction made? Since when did that happen?

No, that's not how it works. Participants can go on-chain with their transactions only if needed.

> LOL. No. You use USD.... No citizens in economically repressive states are gonna flip to bitcoin

Repressive states make it difficult to get USD.

https://www.economist.com/the-economist-explains/2018/04/03/...

> Not when it gets 4 transactions per second tops.

Again, read up on second-layer scaling.

> Not when you have to purchase your currency from early adopters in order to participate.

Why would that be a problem? Also you can mine it yourself.

> Not when you can lose your entire savings by simply exposing your computer to the internet.

There are may solutions for storing your deposits securely, hardware wallets being one.

> Sorry man. You fell for a scam.

Everybody who tells me that knows very little about Bitcoin


> Again, read up on second-layer scaling.

As in "not using the blockchain" otherwise known as "not using bitcoin".

Lightning Network wouldn't even need to exist if Bitcoin was actually good at what it did.

Speaking of, when will I be able to use LN to purchase a cup of coffee from Starbucks?

> Also you can mine it yourself.

Yes. By "mine it yourself" you mean "burn more money on electric and hardware than the value of the bitcoin mined"? Cause last I checked the days of individuals being able to profitably mine bitcoin at home are long gone.

> There are may solutions for storing your deposits securely, hardware wallets being one.

So in addition to buying their Bitcoin from early adopters, these poor repressed people will also need to purchase an expensive (for them) hardware wallet and then learn to use it? What if that hardware wallet gets lost or stolen? What if they forget the key? What if the owner passes on and the children need to access the wealth? How does inheritance work?

> Everybody who tells me that knows very little about Bitcoin

Bitcoin users are an intersection of people who don't understand finance, economics, politics, computer science, monetary policy, legislation, regulation, socioeconomics, math, or seemingly life itself.... It's borderline willful ignorance.


> As in "not using the blockchain" otherwise known as "not using bitcoin".

You don't understand Bitcoin

> Cause last I checked the days of individuals being able to profitably mine bitcoin at home are long gone.

You don't understand Bitcoin https://powercompare.co.uk/bitcoin-electricity-cost/

> What if that hardware wallet gets lost or stolen? What if they forget the key?

You don't understand Bitcoin or the concept of backups


The world's banking system is at around 500 Billion transactions per year https://www.worldpaymentsreport.com/reports/noncash while bitcoin has a mere 200 thousand per day https://blockchain.info/charts/n-transactions Even if the banking system (not sure why you include commuters, but then we should also include bitcoin lambos) were to consume the same amount of power it would still handle 2.5 million times (edit: >6000 times) as many transactions.


Note that 200,000 transactions per day equals 63 million transactions per year. It’s hard to compare figures if the units are different.


oh, sorry. my mistake. still something like 6000 times more transactions than bitcoin.


To multiply your mistake, those Bitcoin transactions in the chart can be multiparty transactions, where, for example, a store issues all the daily refunds to its customers in a single transaction. Basically, a better metric is needed.


Also, lightning network needs to be taken into consideration. But does anyone have exact stats for that?


That's part of the reason why Bitcoin is seen as V1 tech that's being moved away from and other cryptocurrencies using variants of Proof of Stake (which consumes orders of magnitude less electricity but has other potential pitfalls) instead of Proof of Work have grown more popular. Ethereum, generally considered the #2 crypto, is planning to make the switch from PoW to PoS, for instance.

There are some people who still support large-scale adoption of PoW coins, though, and you are right that they're generally willfully ignorant of the environmental concerns. The only case I can see for those is something that aims for true trustlessness and anonymity like Monero, since most PoS methods make compromises around that.

EDIT: I stand corrected. Ethereum is still laying the groundwork for a switch from PoW to PoS and hasn't yet transitioned.


Ethereum has already switched to PoS? This is the first I've heard and I can't find any news about it.


It hasn't moved yet, op is incorrect. They just released the EIP (ethereum improvement proposal). It's going to take some time to get it to production.


Ethereum is still PoW, they released an initial PoS that runs along side the PoW, "verifying" every 100 transactions or so.

The goal is to move entirely to PoS, but things take time.


Why has it taken so long? I thought this would happen in 2017.


Because there are still issues with the idea of PoS. Major one AFAIK is that there's nothing preventing miners mining a fork in addition to the "main" chain, which destroys consensus on which chain is the main chain.


What's the cost in PoS on working on parallel chains?

Does the cost increase in proportion to chain length?


«A the cost of a single bitcoin transaction (~300kW)»

Transactions don't consume energy. This is the biggest misunderstanding floating around here. The transaction rate could halve or double tomorrow without affecting the energy consumption of miners. Furthermore, it's a misleading statement because it implies Bitcoin's utility is equated solely to its transaction rate. The social benefits we extract from Bitcoin are much more than just a raw transaction processing system. For example:

- Some Bitcoin users have become so wealthy that they are donating millions to charities, like the Pineapple Fund (https://pineapplefund.org/).

- The mining industry is pushing so hard for low electricity prices that it is starting to accelerate the research & development of renewables (which are now the cheapest source of energy), helping the industry as a whole move away from fossil fuels.

- A Venezuelan can escape the high inflation rate of his national currency by simply holding bitcoins (whether he withdraws/transacts from his "bitcoin savings" once or ten times a month doesn't really change the utility of the system for him)

None of these points are taken into account if one simply looks at the transaction rate to determine whether Bitcoin's energy use is worth it or not.


>Some Bitcoin users have become so wealthy that they are donating millions to charities

Are you serious? Like, that’s a serious argument? Cartel bosses donate money... same thing right?

>Transactions don't consume energy. This is the biggest misunderstanding floating around here.

The misunderstanding is that you think there is a difference between transaction number and miner power. We don’t care. It’s the same thing because what we’re looking at is total power used (massive) and dividing it by a transaction number which is effective output which is transactions.

Bitcoin is a resource waste, you’re defending it in part because “some people have donated to charity. Good luck with that.


«effective output which is transactions»

No. The "output" of Bitcoin is bigger than that. You focus on a single aspect of the system (processing transaction) and disregard everything else. You fail to look at the big picture, at what the system indirectly creates outside of transactions (helping people escape inflation, and yes redistributing wealth that is donated to charities, etc.) The point is: none of these benefits would exist if Bitcoin/cryptos didn't exist.

Your argument is like saying that "cars are a waste" because one aspect of their operation (engines losing energy as heat) is waste.


There can be an infinite number of off-chain transactions based on a single on-chain transaction.


> There can be an infinite number of off-chain transactions based on a single on-chain transaction.

What the fuck is the point of bitcoin if the majority of all transactions are off-chain?


The transactions are denominated in fucking Bitcoin, each party has the option to go off-chain without counterparty risk.


> The transactions are denominated in fucking Bitcoin

LOL. No transitions get denominated in BTC. Cyrpto is way to volatile for that. Anybody sane prices their stuff in USD or some other currency.

Whether bitcoiners admit it or not, the fiat value of their internet tokens are the only thing they care about. What is their BTC worth and when can they go buy their lambo (using fiat, of course...)


The lightning transactions are denominated in bitcoin.


> Transactions don't consume energy.

Makes me wonder - what is your definition of a transaction?

Traditionally a transaction is something which has been committed to the book or ledger. Transactions cannot exist out in the wild without being on the books. And the ledger creation in bitcoin doesn't happen until a block is mined. Block mining consumes energy. And hence, transactions consume energy too.



> Some Bitcoin users have become so wealthy that they are donating millions to charities

Couldn't you say that about anything that ever profited anyone?

> The mining industry is pushing so hard for low electricity prices

Does it have enough money/influence to make a difference here? I don't know, I'm just asking.

> A Venezuelan can escape the high inflation rate of his national currency by simply holding bitcoins

Wouldn't a real-world currency like $ work here too? What does Bitcoin add?


Sure. Think about polling backends in mobile apps for example. Not just talking about twitter and facebook clients: The app that provides my bus tickets downloads the company's logo and a few other assets every few hours. A billion smartphone users, a few apps per phone, an asset or five per day, before you know it the numbers are big.

But I think you weren't interested in an answer like that, right? You wanted to criticise bitcoin, fairly or not.


But you can optimize those things. Google's current logo is now 305 bytes. Also there's caching. The internet is not built on proof of work.


That's a perfect example of the trend I tried to describe. "Let's download it every 12 hours but optimize that shrinking the file size to just 305 bytes and caching it on the CDN". For a logo that hasn't changed since before the app was first developed.


I'd say gif memes are probably a lot more of a waste. So much bytes stored and streamed u_u; appalling


There has been a gigantic move to offchain transactions since 2015. Myself, Ive only done 3 'onchain' transactions out of the 60ish transactions ive done.

Regarding waste of electricity? Does war count?

Prior to electricity, moving and protecting gold was a major industry.

Ive read about green solutions to this problem. You'd probably still have miners, but it seems the world is moving to offchain for most transactions.


This is "whataboutery" of the first degree - you can't say that no problems in the world should be addressed while war is still a thing, that's just an excuse for irresponsibility.


> There has been a gigantic move to offchain transactions since 2015. Myself, Ive only done 3 'onchain' transactions out of the 60ish transactions ive done.

So your solution the ills of Satoshi's Glorious Bitcoin is to.... wait for it.... not use Bitcoin?

What kind of headspace are you bitcoin shysters in where any of this makes sense? I don't understand it.


If miners were doing the mining for the transactions maybe that would matter but the $100,000 lottery is all that miners care about.

They have opinions on transactions because transactions fees make the lottery more lucrative is all (there exist evangelism in miners but they are the minority)


That's 300 kWh not kW.


A surprisingly common mistake nowadays. I've spotted a couple of times in a major Dutch newspaper over the past few years.

Perhaps it's easier to use Joule instead of kWh.


It’s not a mistake. It’s pretty common and IMO as an EE to say kW when kWh is the only unit that makes sense. It’s colloquially acceptable.


300 kW.. for how long? That's like saying it's 400 horsepower.


kWh I'd wager, it's used interchangeably here.


That would be anywhere between 9 to 90 $ per transaction at domestic rates, assuming 3 to 30 cent per kWh. JSYK


Where is here? They mean completely different things.

(we have joules for this.. but that's a lost battle)


> So de Vries calculated how much electricity bitcoin miners would have to consume for bitcoin mining to no longer be profitable. He assumed that electricity prices make up 60 percent of the cost of mining and that electricity costs an average of 5 cents per KWh. That yielded energy consumption of 7.7GW, which serves as an upper bound of the bitcoin network's energy consumption

That's not the upper bound, since there a monetization strategies that end up using other people's hardware and electricity. Websites that run JavaScript miners are ridiculously inefficient but also don't have to meet his profitability formula because the electricity is paid by someone different than the person/entity that gets the Bitcoin.


Three hundred kWh per transaction!!! That keeps my household running for 40 days!

Humanity has a pretty screwed up way of getting its global priorities right.


What do you think human's priorities should be?

Just curious. I'm not a scam-coin fan at all.


I think that securing a safe and prosperous future should be an obvious one, and one that we're royally screwing up with the amount of energy we spend.

If climate change is even a tenth as bad as it is projected to be this will have been a terribly selfish era of history. We're essentially devouring our young.


Okay, but that came as a problem long before Bitcoin arrived on the scene. In fact, bitcoin itself is a response to an overreaching, overextended global system of control led by banks and other corporations - the same ones who have ravaged the Earth and lied about it, causing the runaway global climate change we have now.

Yes, bitcoin uses way too much energy. But if we had a more fair system in the first place - that is, distribution of resources - it is unlikely anyone would have resorted to inventing bitcoin in the first place.


Well you can't change it now. There's no use crying over spilt milk. I don't see the logic in creating this wasteful system in response to an pre-existing wasteful system, especially with how dubious it is that cryptocurrency is beneficial to mankind.

Like okay Alex Jones, we got a global system of control led by globally globalist banks. I don't think bitcoin is going to change that. Call me crazy, but that seems a bit grandiose.


Well, I'm no anti-capitalist, but I'd say we have to always make sure society serves human needs without excessive waste.

A broad, catch all, empty statement...i know


Bitcoin is a ridiculous waste of energy. Can we please have more regulation of this waste? What is the real benefit of Bitcoin? It can't even be used to pay for anything.

Gold sits somewhere and can be used for manufacturing. Storage of gold uses almost no energy whatsoever. Maybe to protect and produce a safe and hire people to watch it if its a large stockpile. Extraction is also difficult and energy intensive. Best way would be to move beyond move gold and Bitcoin. Bitcoin just continues to waste more and more energy without being used for anything. Storing it and seeing if coins are actually real is too energy intensive.

Bitcoin is irresponsible and a waste. Gotta be a better way to use and create cryptocurrencies for the environment and to avoid exuberant transaction costs.


Yes but extracting bitcoin need not leave large holes in the ground nor create heavy metal runoff in the water supply. Can always use more efficient power sources to mitigate some of the effects. And, there is incentive to do so.

But agree on better ways to create cryptocurrencies. That seems a better solution.


You know what else is a huge waste of energy? Video games. They produce negative value. Should we regulate them? Should you even be allowed to play them?


This argument comes down to "we can't regulate anything if we don't regulate everything". You could use the same argument to say everyone should have nuclear weapons if they're allowed to have toothpicks. It's a slippery slope!

Given that we all accept some things should be regulated/disallowed, we can set aside the slippery slope argument and try to decide what the threshold for societal good or harm should be. Should you be allowed to dump CFCs into the atmosphere? How about lead in the drinking water? Can you defoliate the rain forest for your own amusement? Can you fire machine guns into the air over a city and let the bullets fall where they may? People disagree about where the threshold is, but they generally agree that there is some threshold.


If bitcoin was not using said electricity (if there was no demand), would the power plants be producing that electricity in the first place? You might say no and hence we burn less fossil fuel. But for renewable sources like hydro, you just produce less when the demand is less. (Based on what I read on wiki [0]).

Should we really say `don't try out new ideas because it's wasting precious electricity`? Maybe we should. But I feel like humans have enough technology under their belt that enough electricity generation isn't a dire situation. So are people worried about fossil fuel? If so maybe we should address that. If not, are we saying we need to quota electricity and decide if new techs shouldn't get a shot (if they are somewhat 'wasteful' and we don't know how useful it is)?

Side not, I wonder how much electricity gaming system uses.

[0] https://en.wikipedia.org/wiki/Hydroelectricity#Flexibility


> But I feel like humans have enough technology under their belt that enough electricity generation isn't a dire situation.

It isn't a direct situation it is just wasteful...

> Side not, I wonder how much electricity gaming system uses.

A high end system uses around 750W and assuming 3 hours of game play a day is around 800kWh a year, so enough for 2.5 Bitcoin transactions.

Besides honest question: is there anything about Bitcoin that is experimental anymore? Seems most research switched to different coins so defending Bitcoin to defend research is odd.


1. There is no proven alternative to Proof-of-Work. No, not even Proof-of-Stake: https://download.wpsoftware.net/bitcoin/pos.pdf. Most PoS fans admit this.

2. Thanks to second-layer scaling solutions, there can be arbitrarily many off-chain transactions powered by an on-chain transactions. For more information, visit http://lightning.network/

3. Bitcoin was not invented as a cheap transaction medium, although it was mistakenly marketed as such for a long time. It was invented to serve as a uncensorable sound money. The cost calculation needs to reflect that aspect, otherwise we are measuring the aerodynamics of a tank.


Oroborus proof of stake is kind of proven on paper https://eprint.iacr.org/2016/889.pdf

And bitcoin proof of work isn't that perfect. Over 50% of the hash power is with a few guys in China so they could hack the system if they wanted to.


No they can't "hack" the system - they can try do a double-spend, but it would be very expensive. The cost is proportional to chain length.

What's the cost of creating a parallel chain in "Oroborus" as a function of chain length?


Why the obsession about energy consumption from mining bitcoin? It is because many people still consider it imaginary or non-productive? It is because, unlike many other markets, it is fairly straightforward to find upper and lower bounds for consumption?


Well, if this sort of system is going to be the future of finance, isn't the environmental impact of it something we should consider? Especially, when we hear about old inefficient power pants coming back on-line to fulfill the new demand created by cryptocurrencies.

Edit: Don't the mining costs basically exponentially increase as the supply of bitcoin runs out as it's an exponential decay? Being bounded by an exponential function isn't a good sign.


In response to your edit, no. There is no set cost to mine a bitcoin. Energy is expended in trying to solve the problems the generate bitcoins. Once bitcoins run out, this is still true as instead of directly creating new bitcoins, miners will still be rewarded with the transaction fees connected to whatever block they main.

The difficulty of solving a problem to generate coins is dynamically and automatically adjusted such that it takes about 10 minutes. What this means is that if you had a total of 800 nuclear reactors dedicated to exclusively powering ASIC systems to mine Bitcoins it'd take, on average, just as long as it would if your total electricity consumption was the equivalent of one guy running a low power laptop using a hand crank or bike generator.

So all the electricity consumption represents is demand. As demand increases for bitcoin, so does this price. When the price goes up this means you can afford to pay more for electricity and still show a profit. When demand goes down for bitcoin you need to pay less and less for electricity to maintain profitability.

What will happen when no new coins are being generated is that the reward for each block mined will decrease. This will mean you receive less money for mining a block and thus can not afford to pay as much for energy and remain profitable. The net effect being that we'll see a decline in energy usage as miners running on thinner margins are pushed out of profitability.

The dynamic balancing is actually quite a clever design.


Minting of new coins is an exponential decay (hence the mining metaphor, gold becomes rarer to find as mining continues). There's also transaction fees which are based on supply/demand for blockchain operations.


I think people looking into it doesn't necessarily rise to the level of obsession. The incentive here was likely to publish something (rather than an emotional attachment to some aspect of bitcoin).

That Bitcoin are probably little more than coal-fired Beanie Babies likely does contribute to message board attention.


Beanie Babies can be reverse engineered and replicated.

Bitcoin cannot generate copycat coins(on the BTC blockchain). They are rare numbers.


That’s true. But like beanie babies they only have value people want to believe.

Bitcoin is not fiat, it’s a commodity. It’s not decentralized at all and there are a handful of people thus could destroy it at any time.

Sounds a lot like beanie babies the more I think about it.


Everything only has value that people assign it. I think paying some thousands of dollars for a handbag that has a certain label, or paying $1000 for a phone with a certain label, or so on is completely irrational. But in a market economy, I don't get to decide - the market does. And there's no irony using dollars in that example, as once again they face the same problem. The value of a dollar changes quite regularly against other currencies and it's just a market supply:demand reaction. Should the petrodollar end, I'd expect to see the dollar plummet globally. Great for our exports if our economy doesn't go down alongside the dollar!

51% attacks in Bitcoin are not a real risk for two reasons. The first is that it'd be completely illogical. If there were a group of miners able to work together to generate the resources sufficient to obtain a 51% attack on a major coin, they'd already be printing massive amounts of money completely legitimately. If they started trying to use that to do things like double spend, that'd ruin all of this because of the second point.

Bitcoin is decentralized. And if one branch becomes broken because of something like a 51% attack, it is trivial to fork it and go. Bitcoin Cash being the obvious example of this, and that was over something that lacked a unanimous consensus - nobody is going to want to stay on a fork that's been compromised.


>It’s not decentralized at all and there are a handful of people thus could destroy it at any time.

Uh you might want to check that.


The three largest pools combined hold over 50% of the total hashing power, and all are based in China.


Thats the mining pool though. You can join those pools from anywhere.


Ok, wink, because the blockchain I’ll play along and pretend there aren’t at least 12 people that could crash the entire thing one way or another if they wanted to.


They aren't even rare numbers, they are entries in a ledger.

There's no technical limitation on how many bitcoin can be produced either, only the network consensus that a limited number will be produced. It's built into the software and unlikely to change, but it isn't fundamental, it's part of the consensus.


>There's no technical limitation on how many bitcoin can be produced either, only the network consensus that a limited number will be produced.

Yeah but then you need to start saying things like:

Well Walmart MIGHT start building tanks. Its unlikely to happen, but it isnt fundamental, its part of business.


No, it's not because people consider it imaginary or non-productive: it's because they consider it insufficiently productive for what it costs the globe and that the rewards are obviously localized while the costs are global.


It's quite a gimmick imo. The amount spent has no basis in what is needed, but simply chosen by multiplication of the guesswork of a person several years ago by the chaotic swings of the market.

It's also quite inefficient and even if it was chosen with some rationale basis, the idea of burning energy for years in the hope of providing a high enough bar to prevent a single tsunami of effort over the a brief period of time, less than a few hours and only required to match it, is one of the dumbest security ideas I've ever heard.

I'm not against cryptocurrency, not do I think it will fail in concept, but I very much hope that PoS proves viable and will succeed this utter waste.


I think it's interesting because it allows you to quantify the impact of Bitcoin mining on energy consumption. What would we do with our resources if this was not a viable way to produce wealth? With the continuing decline of natural resources in the world, it's worth seeing what we as humans do with these resources from an academic perspective.

Also, I imagine a lot of people do indeed think that Bitcoin is imaginary or nonproductive. Arstechnica makes money off of readers viewing their articles, so content that validates an opinion or strikes controversy generally does well.


You’re being down voted, but this was a useful comment for me. Made me realize that there’s plenty of energy being spent to mine and handle other commodities too, e.g. gold.


Gold is generally a one-time investment though - you could smelt it into a bar at a high initial energy cost, but after that, transferring it is free. As opposed to BTC which apparently costs 300 KWh per transaction.


> you could smelt it into a bar at a high initial energy cost, but after that, transferring it is free.

How is it free after the initial smelting to transfer?

Moving physical gold around is very expensive.


Exactly. Plus you need to factor in the exploration, mining, and extraction costs too


Because it brings people together. Crypto bashers meets Sustainability preachers. Just connect two popular sub-trends and you are golden!


I have no real opinion on crypto except that like everyone else I’m mad I didn’t buy bitcoin st $5, and am far from an “environmental preacher” -

I only enjoy the irony that some people will go from posting about climate change and go right over checking their bitcoin value.


In a market system, why is this a problem? How much energy and resources are used in all manner of activities which somebody could consider wasteful?

In any event, it seems to me that bitcoin is becoming a proxy for Arthur C Clarke's "megawatt-hour" currency which he predicted would emerge by 2016 [0]. Tying a currency to a fundamental quantity (the Joule) seems like a reasonable way to overcome the deficiencies of fiat currencies.

[0] http://www.kurzweilai.net/arthur-c-clarke-offers-his-vision-...


It's a problem because market systems commonly fail at correctly gauging externalities without enforced regulation or taxation to properly tune the market to more effectively balance short term gains against long term/big picture (planetary) losses.

Something like Carbon taxes, for instance, might help the market hit a better long-term equilibrium for uses of electricity that takes long-term problems like climate change into better account.


Carbon taxes apply to power generation, not power utilization.

Put another way, why is it not a problem for millions of people to turn on tv sets and "waste" electricity watching football? What is the logical endpoint of the regulatory approach?


A) See the "something like" hedge.

B) Most taxes get "passed on" to the next buyer, eventually. A carbon tax on power generation affects power costs, which in turn affects power utilization. It's not a direct effect, but it's still more awareness of the externality in the market as a whole than without such a tax.

C) It's possible that household TV usage is a waste of power utilization, in terms of long term carbon state, and a carbon tax or something like it might encourage more people to watch football games in person or read more books. A carbon tax could encourage more household users to reexamine their power utilization. If you trust the markets to be rational (and that's a wild assumption, I think, these days), people will sort that out for themselves if the market reflected externalities better.

D) That said, the people truly fighting against options like carbon taxes are industries whose daily output far outpaces household power utilization and households may not need to change hardly that much, if industry simply got cleaner or paid more of its share of the damage it was causing.

E) Just because your hypothetical is a silly one, I think the statistics show that right now in terms of carbon output, staying at home and watching TV is actually one of the best power utilizations with regard to greenhouse gas output of an average household. Most American households' carbon output are most influenced by their a) internal combustion vehicles, b) HVAC systems, c) beef-eating habits, roughly in that order. Modern LED TVs are masters of efficiency with respect to other things in one's household.


What is the standard for virtuous use of energy?


So much outrage for bitcoin energy consumption from people on their computers, in well-lit, air-conditioned rooms.


Please somebody convert it to human deaths, for example using data from this paper about death toll of energy generation: https://sci-hub.tw/https://doi.org/10.1016/S0140-6736(07)612...


Correct me if I'm wrong, but this seems like a problem that can't be corrected, either. Decreasing energy use means doing fewer calculations (and a few people here have mentioned PoS). If I'm running a multi million dollar cryptocurrency farm, it's against my interests to support a system that's less computationally intensive. The whole point of investing in powerful GPUs is to crunch numbers harder. If the community comes along and says "we're switching to this thing that doesn't need your beefy energy-guzzling hardware" these miners are going to do whatever they can to make sure their investment is still maximally beneficial. Nobody is going to say "Well it was fun while it lasted!" and throw millions of dollars of hardware in the trash.


FYI: At Bits & Blocks Press I've collected some bitcoin myths and facts, see Best of Bitcoin Maximalist - Scammers, Morons, Clowns, Shills & BagHODLers - Inside The New New Crypto Ponzi Economics [1] or Crypto Facts - Decentralize Payments - Efficient, Low Cost, Fair, Clean - True or False? [2]

[1]: https://bitsblocks.github.io/bitcoin-maximalist [2]: https://bitsblocks.github.io/crypto-facts


It's a currency that is backed by crypto and is printed via a kWh backing. Better than gold and government if you ask me.


This is one reason why I think Nano (Nano.org) will eclipse Bitcoin for pure transfer of value transactions. Nano uses a block lattice that doesn't require mining. The transactions are instant and feeless b/c of this. There is a tiny bit of PoW on the sender's device to prevent spam transactions.


Hmm, how small is “tiny”?

I suspect it would either be too small to be a deterrent, big enough to be wasteful at scale, or both. Is there really a sweet spot that avoids both of those?


This site claims that a single Nano transaction taking 0.112 Wh vs a Bitcoin transaction taking 950kWh. (note Wh vs kWh) https://isnanogreenyet.com/

Check Nano's FAQ regarding "Is Nano vulnerable to attacks?" here https://nano.org/en/faq


Why weren't these points raised for gold & diamond mining through out human civilization. Africans have been killed ruthlessly for diamonds and our so called "smart users" of HN can't seem to see eye to eye on these points. Cause they missed out on BTC! and now they hate it.


Disclaimer: I have invested in Bitcoin, and I think that PoW is outdated. However, I also think that people that keep attacking this aspect are better off tackling other issues.

As everyone else before me said, the banks are wasting a much greater amount of electricity in order to operate. I am not talking simply about gas, lights et al. (although all banks where I live keep lights on 24/7, and in some cases even air conditioners), but also about the massive, MASSIVE amount of legacy infrastructure that keep them going. I'm talking software and hardware that are several decades old, stacked on top of each other with little to no optimization in order to support legacy features and to avoid breaking anything. I fully expect that if we even had a way to calculate how much power is wasted in order to support that kind of jurassic infrastructure for every bank, Bitcoin would still come out on top by an order of magnitude at the very least. And on top of that, we have to compound the aforementioned energy to operate the physical locations.

Yes, I know that this is pure and simple whataboutism, but seeing this kind of argument over and over and over again is tiresome, especially since the fact that we found a way to monetize power consumption could also -hopefully, I admit- have the added benefit of incentivizing renewable power sources such as solar panels.

And remember, cryptocurrencies are at their infancy, and we are already moving away from PoW. I fully expect that at some point Bitcoin will move on too, if it is not replaced by Ethereum or something much more sensible energy-wise.


You might have a point if banks provided the same service as Bitcoin, but they do multitudes more. Such as, providing a infrastructure for doing payments that is actually used by billions of people. Bitcoins are mainly used as an investment or as a store of value, for payments they are not a very good fit.


I fully expect that if we even had a way to calculate how much power is wasted in order to support that kind of jurassic infrastructure for every bank, Bitcoin would still come out on top by an order of magnitude at the very least.

If Bitcoin is around .01% as estimated, that would mean banks account for 1% of world energy usage. I’d bet against it being anything like that high.

High-frequency trading might waste a lot of energy, but that’s another silly arms race with dubious social value, exactly like Bitcoin.


> And remember, cryptocurrencies are at their infancy, and we are already moving away from PoW.

Bitcoin and the crypto "space" is 10 years old. Its only use cases so far have been money laundering, child trafficking schemes, cryptolocker scams, pump & dumps, ICO scams, murder for hire, securities fraud, exchange fraud, and don't forget massive speculation. All while converting massive amounts of non-renewable fossil fuels into heat.

Sorry dude. The market has already spoken. Crypto isn't really useful for anything above-ground real-world. We've got fiat. It ain't perfect, but it does the job.


This is why proof of stake is so needed


There's nothing inherently bad about energy consumption. The raw amount of energy consumed is a useless metric without determining how much of this energy is derived from clean/renewable sources.


You're right that's there's nothing inherently bad about energy consumption.

I would gladly use 7.7 gigawatts and more if it meant developing a cure for cancer, or exploring Mars, or anything useful really.

The crux of the issue is that Bitcoin is mostly used as a speculative asset, i.e. something with little actual usefulness except for traders basically.

Using 0.5% of the world's total energy consumption for this is an issue, imho.


If there's nothing inherently bad about energy consumption, why do you think it is an "issue" for 0.5% of the world's consumption to be used for mining bitcoin? What if 100% of bitcoin mining is powered by renewables?


"If there's nothing inherently bad about energy consumption" is assuming a lot. In theory we could produce all the energy we need with wind, tides, solar, hydro, fusion, zero-point energy, or whatever. Then there would be no issue. But we don't live in that world. We live in a world where most energy in some places is produced with enormous externalities. In other places, the externalities are fewer, but wasting this energy generating bitcoins doesn't eliminate the other uses of the energy, so the aluminum smelters or whatever that might have used the clean energy buy their energy from a different source that pollutes. And if most of the bitcoin mining is happening in China, most of it is being powered with the dirtiest energy source: coal.


You assume that a substantial portion of bitcoins are being mined in populated areas where the energy would otherwise be used for what you deem "better" purposes. How do you reach this conclusion?

The article does not make any effort to investigate the location of bitcoin mining facilities or sources of energy being used. For all we know, 99% of bitcoins may be mined using sustainable energy, off the grid. Without this information, any estimation of raw energy consumption is meaningless.


> What if 100% of bitcoin mining is powered by renewables?

Opportunity cost. What better thing could that energy have been used for? I can think of one.... not being used at all!


> I can think of one.... not being used at all!

You seem to be missing my entire point. There is nothing bad about consuming clean, renewable energy.


Bitcoin contributes zero GDP to the world. It is the equivalent of printing paper then selling it to the next idiot who thinks he can sell it to someone else for a higher price.

Whether the mining consumes renewables or non-renewables doesn't change the fact that the energy expenditure is pointless... even burning oil provides warmth... mining bitcoins?.. Nothing.


There is an argument that could be made that if that power was from a renewable source, it could instead have been used to replace non-renewable power output - all of bitcoin's power usage combined is several coal burning power plants, I'm sure.

Second there's the fact that most of that power is changed into heat. I don't think the heat emitted from anything using electricity is significant enough to have an impact on climate (outside of the cities at least), but still.


Only about 10% of energy consumption in the US comes from renewable sources... https://www.eia.gov/tools/faqs/faq.php?id=92&t=4


Not sure why this is relevant unless you think 100% of bitcoin mining takes place in the US.


Most of it comes from hydroelectric plants with very cheap prices.


Forcing other use on the same grid to come from fossil fuel plants that would not need to operate at the same capacity without the Bitcoin operation.


I don't really think so - I think China built dams in places with too much capacity and it's cheap because it's not in high demand.


> Most of it comes from hydroelectric plants with very cheap prices.

Hyrdo isn't green energy. Dams block fish migrations and fuck up the surrounding ecosystem.




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

Search: