Bitcoin mining uses more energy than Ecuador – but there’s a fix | Hacker News Reader

Bitcoin mining uses more energy than Ecuador – but there’s a fix

newscientist.com

30 pointsrbanffy65 comments

From the point of view of economic and societal impact, the ridiculously high and growing energy consumption of Bitcoin mining makes no sense. Smart people everyday wonder, why are all these miners around the planet burning up all this energy every instant of the day? Is this really necessary?

From a marketing standpoint, however, I think proof-of-work is a brilliant feature of Bitcoin. As mining becomes harder and costlier, requiring ever more specialized knowledge and energy, Bitcoin becomes harder and harder to obtain, and therefore looks more "valuable" to many human beings.

Bitcoin's wasteful energy expenditure reminds me of the hypothesized handicap principle in natural selection: Many species spend a disproportionate amount of energy on "wasteful" features and traits (such as colorful peacock tails) to signal their worth and status to potential sexual partners: https://en.wikipedia.org/wiki/Handicap_principle . For the species as a whole, the expenditure is wasteful, but from the point of view of individual members of the species, the expenditure is not wasteful if it leads to mating!

From the Bitcoin network's standpoint, the energy expenditure is not wasteful if it leads to greater perceived "value" and adoption (compared to other networks).

I have found this article’s quality to be extremely low. The “fix” mentioned in the title is proof-of-work, with an explanation that spans one full sentence. According to the author, the annual electricity consumption of bitcoin mining is estimated to be 27.03 TWh. To compare, we have a very scientific “that’s still peanuts compared to the energy use of the internet”. Recall that the bitcoin network processes approximately 3 transactions per second when running in full gear.

I dug up this report [0], which seems to be one of the most recent estimates of datacenter energy usage (US only unfortunately). The estimated energy usage for datacenters in the US is 70 TWh in 2014, and estimated to be 73 TWh in 2020.

So bitcoin uses approximately ⅓ of the energy the Internet uses in the US. That is far from “peanuts”.

0: https://eta.lbl.gov/sites/all/files/publications/lbnl-100577...

> that’s still peanuts compared to the energy use of the internet

This is terrible reasoning. The internet services something like 3 billion users across millions of websites and applications, bitcoin is a niche toy that runs on the internet and services a few million users by the most optimistic projections.

> the estimated energy usage for datacenters in the US is 70 TWh in 2014

So bitcoin manages to consume 1/3rd the energy of all US data centers to verify a handful of transactions that could be performed on a raspberry pi in a trusted environment.

Based on my energy costs of around $0.20/kWh, the electric bill for bitcoin would be around $5.5B. Where as the current value of all bitcoin is around $126.5B.

At least it is worth more than it costs? I feel the internet provides more value per kWh than bitcoin does. At least, that is, until it doubles and triples a few more times.

Most people who are new to cryptocurrency try to "fix" Bitcoin at least once. After a while some come to realize that Bitcoin's proof-of-work algorithm is actually a feature and not a bug.

It's a terrible feature.

Yup. And that feature is why we need to get rid of it.

Bitcoin asks you to pay a humongous upkeep for a trusted system. I am not convinced the benefits of such a system outweigh the energy waste it requires.

I'm not sure there's anything we can do about it. Proof of Stake is extremely difficult and potentially impossible. Mostly because of the problem that there's no economic incentive to mine a single chain: self-interested miners will mine every fork and there won't be consensus.

Have it be state-mandated, against the law to not use the decidedly best option? No economic incentive needed then at all.

So bitcoin, which still has almost no practical use for anyone in the real world, consumes as much energy as a country with a GDP of $98B and population of $16M.

It used to be that this was acceptable because bitcoin was going to solve our inequality problems and make money more egalitarian and all internet transactions smoother (or something). Yet, with a price of $7K per coin I still struggle to find anyone I know who has ever made a real purchase with it.

So what exactly are we getting from this energy demand? Does there exist a pie chart for where all the miners source their energy and maybe a metric for tons of CO2 per coin/transaction?

Some people think they’ll be rich, that’s enough for them, and they either don’t care about, or rationalize away the rest.

I have purchased hotel rooms and computer parts with Bitcoin.

And realize these energy costs can only be afforded because of the speculation that these crypto-assets will gain mass legitimacy, and therefore cause a (unreasonable) wealth distribution weighted towards the earliest adopters - along with the increased perceived value of the crypto-assets.

You can spend fractions of a Bitcoin, so it doesn't matter what the price of a full one is. ~15,000 satoshis (a satoshi is the lowest it can be split, 1/100,000,000 BTC) is currently equivalent to 1USD.

No, it is not a feature. It is an implementation of the feature - to have secure and zero trust transactions.

These articles about how much energy Bitcoin uses never mention the value it provides - it's a secure way to make financial transactions internationally with zero trust, relatively (compared to most other current methods) low fees and high speed. There are other ways, like the banking system, but they have their own costs associated with the security of the transactions, and their guarantees are backed by government, which don't always agree with you.

Maybe this kind of system is worth burning a lot of electricity for. It seems like the market is saying that it is.

Exactly. This is the crucial point all of these people seem to miss.

There is all kinds of other egregious waste (electrical and otherwise) going on in the world, how is this any worse than say millions of computer's online/burning power to play DoTA for instance (sorry MOBA fans, nothing personal. using it bc it has an enormous player base). It's a weird comparison but it gets the point across.

Plus, relatively speaking.. an entire year to power the country of Ecuador is the same amount of energy it takes to power the Bitcoin blockchain for a year doesn't seem that "wasteful" considering how many lives Bitcoin has changed & the miracle of being able to send up to $x,xxx,xxx USD (or any other currency) around the world, confirmed in less than 10 minutes for a paltry fee of ~$5.

I'm not sure why but there is a clear bias on HN regarding most Bitcoin articles.

This is just my opinion; take it with a grain of salt.

Can you talk more about the lives Bitcoin has changed? Are you referring to people using it for black market transactions or are you referring to speculators?

By the way you asked the question I can already sense your clear bias. I'll answer you however.

Countries where the currency is inflating at enormous rates, countries where the banking system is corrupt/out of date/not functioning, Bitcoin has changed these people's lives. Venezuela would be a good example from this year: https://www.forbes.com/sites/realspin/2017/02/03/why-venezue...

This is a myth. I can tell you from first-hand experience that bitcoin is not changing lives in Venezuela (at least not on a noteworthy scale). Practically nobody accepts bitcoin in exchange for food and services and many of the rural regions that are hit the hardest by the failing economy don't even have reliable access to internet. This idea that bitcoin is somehow aiding the impoverished is the height of technocratic arrogance.

I don't see proof anyone has been helped. The article points out that Venezuelan Bitcoin users rose to 85,000 users in 2016 and that a travel agency in Venezuela only accepts Bitcoin. The article also says Bitcoin is useful for buying stuff from online retailers.

I believe the larger systematic problem with corrupt and non functioning economic instruments is not their corruptability but rather the structures that promote and rely upon on corruption which Bitcoin doesn't appear to resolve.

> the miracle of being able to send up to $x,xxx,xxx USD (or any other currency) around the world, confirmed in less than 10 minutes for a paltry fee of ~$5.

That's a miracle?

I've ordered physical products from all over the world - China, USA, Europe, India - and the payment confirmation always a) took just a few seconds and b) cost me just a few cents.

I also rarely bring cash with me, and pay most things with debit card. After punching in the PIN, the confirmation takes about 2 seconds to be printed. I'm happy, the seller is happy, we all carry on with our lives.

I don't know what it looks like on the receiving side, but I routinely buy games from GOG and Steam and they immediately let me know that "payment has been confirmed" after a purchase is made, so I am assuming that they see the confirmation of the transaction on their side immediately.

The only miracle I see here is how somehow $5 and 10 minutes per transfer is considered a great leap forward in efficiency.

Certainly bitcoin brings a lot to the table, but efficiency, so far, isn't it.

Well the market also says it's worth burning a lot of fossil fuels to move around and heat our houses. The market is just not great at factoring in negative externalities.

As for the other things, most people don't use Bitcoin as a secure way to make financial transactions at all. In fact it seems to be mainly used for speculation. And when it does get used for transactions, it's mainly for crime. Turns out most people are actually fine trusting an institution like a bank when making transactions.

Bitcoin is a really cool solution looking for a problem to solve it seems.

I'm not sure what your point is. Are you saying that we should not heat our houses or move around?

All I'm saying is that just because the market might prefer to burn lots of coal to do so, that might not actually have the best outcome for society.

let's be honest, mostly it only serves to buy guns, pills & dodge taxes

Just to note, financial transactions from bank account to bank account are slow and expensive compared to other methods. Even with BTC -> BTC transactions, there are much faster options, and if you claim they are cheap, you must take into account that there is an implicit cost of holding an volatile asset (even if lately that volatility has been beneficial for the asset holder).

>>> Maybe this kind of system is worth burning a lot of electricity for. It seems like the market is saying that it is.

But it seems to me people aren't using Bitcoin for that, they're using it more as an investment.

Whatever sets the value of Bitcoin, it's not because it's a competitor to Western Union and the likes.

Why is this a problem?

We do all sorts of relatively useless stuff that burns electricity. How many gigawatts are wasted in internet forums? Call of Duty? Lighting empty rooms?

Why is bitcoin worse?

If not worse, at least just as bad, which makes it, on itself, bad. Two wrongs don't make a right.

On top of that, although not everyone may agree: one may argue that internet forums and Call of Duty are not a waste. At least not compared to bitcoin mining.

Bitcoin mining requires more energy if more people join in (difficulty increases), while the outcome basically stays the same. Furthermore one could say it's not doing anything, only preventing something (too many blocks being created).

Why? Mining allows you to certify transactions and create a tangible thing. How is that worse than electricity from credit card terminals, etc?

We should educate everyone in reducing energy waste.

Energy efficiency is being improved in manufacturing, most appliances we use in our homes, etc. Should we not demand the same of Bitcoin just because other parts of the society uses >0 energy?

"Bitcoin - not worse than lighting empty rooms"

lol proof of stake, that's basically giving all the power to banks and governments to control the transactions which is exactly what we have now.

In a sense, if the growing power requirements for cryptocurrency mining box out players who can't afford the compute and the energy, wouldn't that give all the power to the monied players anyway?

Indeed, anyone who can afford and dumps the most computational power and energy into the system. Imagine the waste of resources if there was a war/battle against bad actor(s) for control.

Not to mention that there is a central figure on ethereum to talk about (and hence to be controlled) which I think is a bigger threat in the longer term. I think Satoshi disappearing was what makes btc stronger.

> that's basically giving all the power to banks and governments to control the transactions which is exactly what we have now.

Sorry for a possibly stupid question, but... what, exactly, is wrong with the current control governments and banks have on your transactions?

If you don't mind the power relationship between governments/banks and society there's a cheaper way than cryptocurrency called using banks. However many seeking cryptocurrencies are doing so to either escape regulation, or to escape capitalist centralized power structures.

Maybe instead of a linear chain we need a hierarchical chain, with distributed verifications

As in: you send a transaction, there's a low effort PoW needed for that to be accepted by the network

Intermediate nodes bundle up some transactions and create a hash using a higher difficulty PoW. Those are compensated with a small fee

Higher level nodes bundle those in bigger blocks with a high-effort PoW and higher fee.

This way the transaction cost is spread more evenly

Bitcoin and all it's altcoins that connect to it could work like this if 'atomic swaps' took off.

https://ark.io/ is a coin with proof of stake with delegates, who mine. You vote is proportional to your money, but you get back a portion of the mining bonus from the delegates. A bit more complicated than BTC but that's ok for me for not burning so much energy.

What I don't understand about the energy use is, what percentage of that is mining, and what percentage is to facilitate transactions?

If 90% of that energy is going into mining, won't Bitcoins energy usage really come down once all the coins are mined? Or once the value of the coins is much less than the cost of the energy required to mine them?

It seems like there are some natural checks here.

Mining and facilitating transactions, by which I'm assuming you're meaning certifying that transactions are legit, are the same thing.

Here's a short video by 3Blue1Brown on how cryptocurrencies work:

https://www.youtube.com/watch?v=bBC-nXj3Ng4

Mining is transactions.

Can we have something like a person or democratic coin?

Everyone gets one vote for being a human alone.

Not for having more money or more computing power.

How do you propose one proves that they are (a unique) human?

Web of trust?

Absolute proof this way isn't possible, but you could render the cost of fraud very very high.

> the cost of fraud very very high.

Just high enough for lots of people and their friends to conspire against the rest of the network?

I'm afraid web-of-trust alone is not enough to make a safe and useful cryptocoin. It would also require significant additional changes from current popular cryptocoin designs.

>Just high enough for lots of people and their friends to conspire against the rest of the network?

That depends upon what they conspire to do, surely?

As far as I can see there are three attacks:

* Creation of fictional people

* Conspiracy to "unperson" a real person

* Identity theft

There are many defenses that could be put in place against all of these using a web of trust system.

The PGP version is a little rudimentary to protect against them, but it's still a good foundation.

good question.

proof of fingerprint?

proof of face?

You will find that designing such a system requires distinguishing between humans and computers on a computer network. This is a lot harder than it sounds. For something when there's very little at stake, we use CAPTCHAs. But what if there were a cryptocoin's transactions or governance at stake?

It's not impossible but it's very, very difficult and would suffer slow transactions, high fees and probably nonzero trust. The downsides are so great as to not make it worth the effort.

I have a piece of A3 paper stuck to my whiteboard here that tried to flesh out how something very much like you describe might work.

Unfortunately, it was the product of a very late night booze and hashish session so is pretty much indecypherable.

A pity, as I'm sure there solution to all of the world's problems is in there, somewhere...

If a different cryptocurrency can make Proof Of Stake work _and_ persude BitCoin miners that switching to Proof Of Stake is better for them than not, then Bitcoin might switch.

But those are both pretty high hurdles to clear.

The GDP of Ecuador in 2014 (all time high) was $102.29B.

instead of eliminating PoW we can improve on the efficiency of the miners. Its possible to use the emitted heat from the ASICs/GPUs to warm up households for example. Imagine if this were a city/country-wide thing: you could have an economically self sustaining heating system.

Is Ecuador known for using a lot of energy?

No, it's a small country too, only 15 million people, and it uses a lot less energy than the US per capita.

I prefer units of power of a burning Library of Congress.

Everyone is happy to sentationalize the amount of energy that the bitcoin network uses, but no one ever mentions how much energy use is acceptable.

Nor does anyone seem to attempt a thorough cost/benefit analysis. How much energy from financial infrastructure could bitcoin free up?

Edit: vague comparisons like these really bug me, because while they make problems seem large by comparing them to A WHOLE COUNTRY, they actually offer little in the way of information. Most of us know nothing about Ecuador's energy use in comparison to any other country or industry, except perhaps that it is relatively small. But I guess it gets you the clicks...

with you on that one... a better comparison would be to see how bitcoin stacks up against the power dedicated to porn... or spam

Porn and spam don't have an incentive structure to redistribute wealth weighted towards the earliest adopters - they're otherwise not really comparable unless you narrow in on one point, such as energy use.

Ecuador is about 66th percentile by economy size, with a GDP of about $100bil.

So Ecuador is producing $100bil in value, which is similar to the total volume of transactions on the Bitcoin network.

So using less energy, Ecuador produces more economic value than Bitcoin can even move.

That sounds like Bitcoin is wildly inefficient, in that it can't transmit stored value for less than that value costs to generate in power:

It would be more efficient to use the power to create more wealth than transmit it with Bitcoin!

But you're ignoring the fact that bitcoin is about more than simply moving money. There is value in security and decentralization.

Further, GDP ignores consumption, it is gross product, not net product. Further indication to me that comparing bitcoin to power consumption of a small country is inappropriate. Apples to oranges.

Edit: after reading a couple articles on wikipedia, I'm confused. Gross implies product before subtraction of consumption (I.E. revenue), while wikipedia claims that GDP factors in so called intermediate consumption, which would make it more like a net value (I.e. profit). Do we have any economists browsing? Genuinely curious.

Their 'fix' is proof of stake, which is unproven, though I think can work technically. Whether it works economically and leads to more or less decentralization remains to be seen.

Either way, bitcoin isn't going to switch to proof of stake, so the headline is still nonsense from any angle.

Edit: I see downvotes but not replies - I'm not sure which part of this is controversial.

Reply on news.ycombinator.com