[DanBlake]

Just took a quick look at this. If you are located in the 'mining valley' of Washington where power is ~2c/kwh you are still getting healthy profits.

A computer with 7 GTX 1070 graphics cards should produce ~230 mh/s and draw 1 kw. This would cost approximately $30/month in power factoring in kw demand + cooling.

The above setup will currently generate $385/month in ETH.

So basically for miners who are in the right spot with the right facility, this is still profitable. The question is of course for how long. You also need to factor in the cost of equipment, datacenter, employees and difficulty/price.

But even if you dont have a facility in washington and just mine from your apartment, your power cost would probably be $100 a month. So its still 'profitable', just not nearly as much as it was in the run up.

Cliffnotes: 'professional' miners dont care. Even with the 'crash' today, they are making more per day than they were before the entire run up. For instance the 'worst' time for mining was December 2016 where you would only make $7.50 a day gross in ETH.

[BoiledCabbage]

Call me naive, but am I the only one who looks at mining as one of the worst inventions for consuming energy possible?

Almost all of society functions on energy, some of the largest breakthroughs in society have been on sudden abundance of cheap energy and the machines, vehicles products they can create.

Entire economies can be crippled by rising costs in energy (oil shocks of the 70s) and boom by sudden drops in cost of energy.

So we've created an "industry" where you are essentially paid by comverting energy to waste. Paid to perform extremely intense difficult (ie wasteful) operations to back a useful technology (digital currency).

Assuming it catches on, energy will never be cheap, there will always be a higher floor now due to options for "mining". As we get better at it and it becomes less wasteful, the digital system will simply raise the reward so people are incentivised to once again waste it.

Ignore the short term for the moment, and which ever currency you're backing. We've created a long term societal motivation/reward to harvest every joule produce by the sun and use it to calculate hashes. I'm not talking about the next decade obviously, but we have incentivised that behavior.

If there is anything technologists should understand is that whatever your beautiful perfect technology is, it will instead be used based on whatever has been incentivised.

Regardless of the technology it powers, this is a terrible societal incentive - and one that will be around a lot longer than people are considering.

[nodamage]

A cryptocurrency enthusiast would tell you that the energy consumed is not "wasted" because it is necessary to secure the block chain.

Personally I find this argument unconvincing and somewhat tautological. 1) Surely we can come up with a less wasteful solution to this and 2) why should we assume that securing the block chain is a actually good use for the energy spent?

Especially if you consider that the process of "mining" is literally computing hashes over and over again until you find the right nonce that meets some arbitrary criteria, it's hard to see this process as anything except wasting massive amounts of energy.

[angus-prune]

It is worth comparing this against the cost of minting money.

A brief search showed that the US mint used 694,462.4 gigajoules in 2011[1] (~0.2Twh). The US is ~25% of world GDP, so lets times this by 4 for a naive cost of minting across the whole world.

This gives a cost of ~1Twh annually for minting the entire world's supply of money, with a total Gross world product of ~$80 trillion[2].

Bitcoin is using ~1.3Twh[3] for a market cap of ~$35 billion[4].

Overall this means that bitcoins are ~3000x less efficient than physical money.

After writing this I've realised I should perhaps limit to the cash based economy.

The amount of US dollars in circulation is ~$1.3 trillion[5].

This gives a slightly better ratio of 222x less efficient.

Bitcoin is 222 times less efficient than a physical system of metal coins and paper/fabric/plastic.

[1] http://www.designlife-cycle.com/us-penny/ [2] https://en.wikipedia.org/wiki/Gross_world_product [3] http://blog.zorinaq.com/bitcoin-mining-is-not-wasteful/ [4] https://coinmarketcap.com/ [5] https://www.newyorkfed.org/aboutthefed/fedpoint/fed01.html

[whiskers08xmt]

This is not entirely true, as cash requires energy after it's been minted. Money transport, cashiers and gold old fashioned vaults all require energy. With that being said, PoW is definitely wasteful, and for crypto-currency to be taken serious on the world stage, another form of consensus algorithm is required. Personally I'm partial to Proof of Stake[0].

[0]https://en.wikipedia.org/wiki/Proof-of-stake#Criticism

[omellet]

Don't cryptocurrency transactions take a lot more time & energy than, say, a credit card transaction?

[whiskers08xmt]

Depends on the cryptocurrency in question, but yes, a cryptocurrency working under a Proof-of-Work scheme like bitcoin, will use more energy than credit card systems. I'm not sure it'll ever be possible to outperform centralized, digital systems on an energy usage level. What I was critical of, was the comparison to physical money. Energy usage isn't really that important a metric for payment systems though, price is probably more relevant, where some innovations in cryptocurrency might come to be competitive with credit cards.

[lawless123]

Yes, but it sounds cooler!

[Sleeep]

Your calculations are off by a large magnitude because you cited only energy consumption by the US Mint the US Mint only produces coins. Federal Reserve Notes (paper money) are printed by the US Bureau of Engraving and Printing.

And as sibling comments say, cash uses energy after it's produced.

[maxerickson]

Doesn't bitcoin compete more with electronic money than it does with cash?

That it is person to person makes it like cash, but moving a dollar with bitcoin isn't quite as convenient as moving a dollar with cash, whereas moving $5000 with bitcoin is vaguely comparable to moving $5000 using a bank transfer (the value has to be present at the bank/in bitcoin, etc).

[h1d]

Please add all the energy banks, vaults, transportation and payment channels use where Bitcoin does it by itself.

[Consultant32452]

US dollars are primarily created by banks using the fractional reserve system. So a better estimate, though harder to find, might be the total energy (including human calories) to create a loan.

[cglace]

Why? It's going to cost energy to create a loan regardless of the fiat it's denominated in.

[Consultant32452]

Perhaps this is ignorance on my part, but my limited understanding of bitcoin suggests that new bitcoins cannot be created by creating a loan. It would be impossible to use a fractional reserve system with bitcoins by definition. This distinction is important since over 90% of US currency is created via loan using this fractional reserve mechanism.

[colordrops]

Does this account for the energy that goes into the mining, and transportation, and processing of the ore before it gets to the mint?

[Houshalter]

Gold might be a better comparison. A ton of energy is spent to extract relatively small amounts of gold.

[dublinben]

But gold isn't widely used as a currency. Which actually makes it the perfect comparison for cryptocurrencies, I guess.

[mrb]

1) No there is no other way. (An alternative, proof-of-stake, is still an active research area. Even Ethereum abandoned the idea of completely switching away from proof-of-work because they realized PoS isn't completely workable.)

2) Because a permission-less censorship-resistant decentralized financial system has the potential to truly improve society, hence worth spending energy on it.

I have presented multiple arguments why (Bitcoin) mining is not wasteful here: http://blog.zorinaq.com/bitcoin-mining-is-not-wasteful/

[dvfjsdhgfv]

I read your arguments and unfortunately most don't hold water.

> 1. Miners currently use approximately only 0.0012% of the energy consumed by the world.

This doesn't change the fact that you're using energy. The second argument is basically the same, just extrapolating into the future.

> 3. Mining would be a waste if there was another more efficient way to implement a Bitcoin-like currency without proof-of-work.

This is a logical fallacy. Mining means wasting energy regardless if there are other ways of producing digital currencies or not.

Now, this one is actually interesting:

> 4. Bitcoin is already a net benefit to the economy. Venture capitalists invested more than $1 billion into at least 729 Bitcoin companies which created thousands of jobs. You may disregard the first three arguments, but the bottom line is that spending an estimated 150 megawatt in a system that so far created thousands of jobs is a valuable economic move, not a waste.

This example is a bit special because the jobs haven't been created by BCs themselves but by VCs spending their own money into BC-related companies. I just hope most of these aren't BC mines, otherwise we'd have a vicious circle here...

>5. The energy cost per transaction is currently declining thanks to the transaction rate increasing faster than the network's energy consumption.

Again: this kind of argument is of "it's not as bad as it sounds" type, and does nothing to refute the claim that mining is wasteful.

[shoo]

> 4. Bitcoin is already a net benefit to the economy. Venture capitalists invested more than $1 billion into at least 729 Bitcoin companies which created thousands of jobs. You may disregard the first three arguments, but the bottom line is that spending an estimated 150 megawatt in a system that so far created thousands of jobs is a valuable economic move, not a waste.

We cannot decide if an activity is valuable or not based solely on if it creates jobs.

e.g. suppose the government spends $1b on some entirely useless activity, such as employing people to dig holes and fill them in again. From the metrics of job creation and GDP this program can be regarded as a success. But there is a big opportunity cost in that resources and peoples' time has been spent on something utterly pointless when they could have been focused on a less useless or indeed a genuinely useful activity.

The world still has many genuine problems that still need to be solved. The time and resources could be directed to: maintain infrastructure, better educate people, roll out family planning programs, eradicate disease vectors, draft and enforce environmental regulation, build low cost housing, refit systems to improve energy efficiency, ...

[mrb]

"We cannot decide if an activity is valuable or not based solely on if it creates jobs."

You are right. However Bitcoin is not pointless and comparable to digging holes and filling them up. The simple fact that this financial network is censorship-resistant is a huge benefit to society, already positively impacting many people.

[a_imho]

1. Miners currently use approximately only 0.0012% of the energy consumed by the world.

Actually if that number is accurate, is quite shocking.

[notahacker]

As a back-of-the-envelope calculation, cryptocurrencies are [over]valued with a market cap of ~$70bn (from https://coinmarketcap.com/) of which BTC itself constitutes just under half.

World broad money supply is estimated around $80 trillion (not including many other stores of wealth and financial instruments which are not money). So the first conclusion one could draw is that cryptocurrency is also a surprisingly high proportion of currencies in circulation, at around 0.1% (treating cryptocurrencies as equivalent to broad money and exchange values as accurate)

A second would be that cryptocurrencies which apparently use 1% of their capital value in energy per annum to function even at low transaction volumes are unlikely to be a viable long term solution to the problem of exchange.

[Sholmesy]

Agreed. This is a non-trival amount of energy. Given the infancy of the technology, it is likely to grow.

Edit: non-significant -> non-trivial

[aerique]

I assume you mean shockingly high?

[mrb]

1. See last paragraph in my post's argument 1. The point I raise is that it makes no sense for people to be so outraged and vocal about miner's energy use and apparently don't care about much bigger energy users/wasters by multiple orders of magnitude... At the very least this makes critics hypocrites as they themselves waste energy and don't to care about it (eg. driving a car with low fuel efficiency.)

3. It's not a fallacy. If X is worthwhile to society, and if the only one way to obtain X is to spend energy doing Y, and if alternative uses of this energy are not clearly superior (for example doing Z), then Y is not an energy waste. X = permission-less censorship-resistant decentralized financial network. Y = mining. Z = for example building desalination plants to provide clean water to third-world countries.

4. You seem to recognize the validity of this argument :) See the reference I give in the post: https://venturescannerinsights.wordpress.com/2015/09/04/the-... these are not mining companies.

5. This argument shows that even if you are unconvinced by arguments 1-4 and holds the view that mining is wasteful, you should at least recognize that it is becoming less and less wasteful over time.

[Jabanga]

>Even Ethereum abandoned the idea of completely switching away from proof-of-work because they realized PoS isn't completely workable

No it didn't. I follow Ethereum development closely and I know for a fact that's absolutely untrue. I'm not sure why you're misleading people about this.

[mrb]

You are right, I should have said "temporarily" not "completely". They abandoned the idea of switching to 100% PoS right away, and instead are trying to work on a hybrid PoS/PoW. Source: https://twitter.com/vitalikbuterin/status/851503370250661889 (100% PoS remains the ideal goal, but they have been unable to work it out yet)

[gus_massa]

I think they didn't abandon the idea, but until they have a working prototype it's just a idea. Nobody is sure if it's possible to use in the real word, were many people will try to steal a few million dollars if they can.

[fghtr]

But you both did not give any links...

[ckastner]

> 2) Because a permission-less censorship-resistant decentralized financial system has the potential to truly improve society, hence worth spending energy on it.

The other side of the coin: a permission-less censorship-resistant decentralized financial system also re-introduces a substantial set of significant problems that societies have already long solved for themselves.

[TeMPOraL]

In the words of Scott Alexander:

"People are using the contingent stupidity of our current government to replace lots of human interaction with mechanisms that cannot be coordinated even in principle. I totally understand why all these things are good right now when most of what our government does is stupid and unnecessary. But there is going to come a time when – after one too many bioweapon or nanotech or nuclear incidents – we, as a civilization, are going to wish we hadn’t established untraceable and unstoppable ways of selling products."

http://slatestarcodex.com/2014/07/30/meditations-on-moloch/

[CyberDildonics]

Predicting a future of devestation. You would think we would be past buying into that ruse by now.

[lightedman]

You haven't done your research, because ETH did not abandon PoS, and permission-less distributed databases have existed since the 70s and failed then for the same reason all these will fail now - you simply cannot scale and keep it distributed, it all ends up getting centralized to a few key points because of the cost to maintain the exponentially-growing distributed infrastructure.

[legulere]

How are bitcoin and ethereum censorship-less and permission-less? Most cubans for instance don't have access to the internet without a permission. China could shut down the bitcoin easily by simply blocking bitcoin traffic.

[CyberDildonics]

Is a USB drive really plug and play if it doesn't even work on my 20 year old flip phone?

Why would you conflate two separate systems like that?

[walterstucco]

> a permission-less censorship-resistant decentralized financial system has the potential to truly improve society

It is dependent on a huge amount of cheap energy

I wouldn't call it censorship resistant yet

Raise the cost of the energy and only the ones with deep pockets can decide what is true

[chillingeffect]

This comment led me to compare crypto mining with the American Gold Rush in the 1800s.

An intriguing difference might at least some miners struck it big due after finding huge geographical deposits of gold. In cryptocoins, though, the flatter distribution of the space means people tend to get out what they put in.

[trophycase]

They most definitely have not abandoned Proof of Stake

[Nursie]

>> No there is no other way.

There are various other ways, however many of them ditch the decentralised-trust nature of cryptocurrency. If this is not something important to you (which it isn't to that many people) then crypto-currency represents a huge waste of power for no appreciable gain.

[DiThi]

PoS is perfectly workable, but one wouldn't switch overnight for a myriad of reasons. Ethereum will eventually be PoS for sure.

There's also systems without "mining" like iota, which has every sender verify other two transactions, so it's very scalable and has no fees. The drawback is that having no mining, all tokens were created on genesis.

[dvfjsdhgfv]

It would be true in an ideal world, but in practice governments of different countries can and do regulate what can and cannot be done with bitcoins. Most of this is based on fear of losing control.

[TeMPOraL]

> in practice governments of different countries can and do regulate what can and cannot be done with bitcoins. Most of this is based on fear of losing control.

Most of the time people seem to say as if that was a bad thing. But consider what happens when government does ultimately lose control. You get Somalia.

As it is today, it seems governments worldwide agree cryptocurrencies are overhyped. They all seem to be saying basically: "it's cool you're having fun with new Internet points; just don't forget to fill in your tax form".

[KVFinn]

>A cryptocurrency enthusiast would tell you that the energy consumed is not "wasted" because it is necessary to secure the block chain.

Note that this is zero-sum.

If a device is invented tomorrow that is 100x more efficient per dollar/watt to mine, the cost to 'secure the blockchain' remains the same because people will just buy 100x more computing power for the same investment and the difficulty ramps up to compensate.

[IanCal]

This is not correct unless you factor in the hardware cost.

If you changed graphics cards to use 0.1% of their current power, then miners would not immediately buy 1000 times more graphics cards.

[keredson]

you're right about the detail of his point (people won't buy 100x just because a 100x increase in efficiency), but the point he was trying to make is correct. given a fixed global investment, a 100x increase in efficiency won't reduce costs. it will just increase the difficulty level to compensate.

[IanCal]

But it does have an impact on the energy use. If the energy efficiency goes up 100x and people aren't using 100x as many machines then the overall energy usage drops.

[jstanley]

> 1) Surely we can come up with a less wasteful solution to this

10 years ago people would have told you there is no possible solution to this.

Proof-of-work cryptocurrency is a tremendous breakthrough. It will certainly be surpassed one day, but it's currently our best available option for a system of money.

[canoebuilder]

What? Why isn't everyone using "our best available option"?

What are the benefits of thousands of different systems keeping a record of every single transaction, and how is this not a hindrance to widespread use of our best available option?

Crypto-currencies have utility and uses but they are rather niche and specialized. Crypto-currencies aren't a mass market system. And they're not structured in a way that they can be.

[ebcode]

One benefit of having a distributed ledger (your thousands of different systems) is that it decentralizes trust. With fiat currency, you have to trust the government (or private company, in the case of the U.S.A.) that issues it, not to print so much that they debase it. I have no trust in my government, or fellow private corporations, given what I've been through. Bitcoin, on the other hand, well, at least I can read the source code and compile the executable myself.

Given a choice between trusting a person/corporation and trusting a mathematical equation, I would rather place my trust in the mathematics.

The internet is a mass market system, and it was designed with biological growth in mind. Bitcoin is built on top of the internet. So I don't see why you would think it couldn't see widespread adoption. It's just another program.

[kahnpro]

It's incredibly foolish to think you're only trusting mathematics.

All you've done is shift your trust to the authoring process of the Bitcoin clients and to the majority of the miners. Authoring and distributing the clients, and determining the majority of clients and the majority of mining power, is still a social and political problem driven by human greed.

[nodamage]

> Given a choice between trusting a person/corporation and trusting a mathematical equation, I would rather place my trust in the mathematics.

This comes across as incredibly naive. Just consider the whole Segwit2x controversy and think about exactly who controls the future of Bitcoin. You've just shifted your trust to a different group of people.

[tigershark]

Trusting bitcoin you are trusting the Chinese miners that have the majority of the hashing power. Bitcoin won't EVER be widespread if his whole network can process a mere 7 transactions per second (Wasting a huge amount of energy).

[colordrops]

> 1) Surely we can come up with a less wasteful solution to this and

No one has found a solution yet that is as secure as proof of work, so this point is moot until then.

> 2) why should we assume that securing the block chain is a actually good use for the energy spent?

Who is this "we" you speak of? The energy is being paid for and the market demand demonstrates that it is a good use.

[MereInterest]

> Who is this "we" you speak of? The energy is being paid for and the market demand demonstrates that it is a good use.

This argument is nothing but an appeal to the majority. I could equivalently say that the existence of spam emails shows that spam emails are good for society. After all, spammers have paid for the electricity and bandwidth to send spam, and the market demand demonstrates that it is a good use.

Market demand is a useful tool for measuring the current state, not for predicting an ideal state.

[TeMPOraL]

> No one has found a solution yet that is as secure as proof of work, so this point is moot until then.

The point isn't moot. The point is that maybe we should stop pushing this tech forward until a less wasteful securing process can be developed.

> Who is this "we" you speak of?

This comes up often, and the answer is usually the same: that "we" are the people who consider the society they live in as something to contribute to and grow, and not as an exploitable resource to parasite on.

> The energy is being paid for and the market demand demonstrates that it is a good use.

Market demand only demonstrates that there are some people who are willing to pay money for this. Not that it's good or useful (see the spam example of sibling's comment). This is true especially if the thing is paying for itself.

[nodamage]

"We" refers to society that you wish to convince to adopt Bitcoin as a currency. As has already been mentioned, the mere fact that people are willing to spend electricity on mining does not inherently demonstrate that it is a good use of such electricity.

Again, I find this defense to be incredibly tautological:

"This approach to securing a distributed ledger sure seems to waste a lot of electricity."

"But it's not wasteful because it secures the ledger..."

[dalore]

You could in theory create a centralised electronic currency with all the hashes computed. But then how would you distribute it? Chicken and the egg. How do you get people to use your currency if you have all of it? Do you just give it out to people? But what system. If you're giving it away it has no value. It cost you nothing to create it. etc.

Mining also solves the distribution and creation of value problem. People won't use your currency unless it has value. To have value it needs to have a market to use it. To have a market you need people using it. Mining creates an incentive to create it since it has some value. And mining means the currency is distributed to people who create value for the system.

[akhatri_aus]

As with anything based on the enforcement/legal sector there is no economic productivity.

[freddref]

We might not be able to avoid proof of work, but we could make the work useful in some other way.

Primecoin is an example of this, where the mining process produces prime numbers which can be useful for.. something?

[keredson]

I've had this thought too, but the counter argument is gold. Spending millions in fuel to dig up chunks of gold, just to go "re-bury" them in a bank vault somewhere is an equally appalling, use of the world's energy resources, but that hasn't stopped it from being a profitable venture for thousands of years.

[Fr0styMatt88]

> Especially if you consider that the process of "mining" is literally computing hashes over and over again until you find the right nonce that meets some arbitrary criteria, it's hard to see this process as anything except wasting massive amounts of energy.

I think this wording obscures why the mining process gives the network trustworthiness.

You are changing the nonce over and over again until the HASH of the block meets some criteria. For example, your hash is below some value (it starts with a certain number of zeroes). This is important.

It's important because you can't 'fake' the work. The only physically possible way to come up with a block that has a conforming hash is to change the nonce and run the hashing algorithm and repeat this until your hash matches the criteria. There's no way I can know the nonce I need to use without running the hashing algorithm. Hence, if I present a block that hashes to a conforming value (you don't just take my word for it, you run the hashing algorithm on the block yourself to independently verify the work), you know that I've invested time and electricity (tangible, real-world things) in the creation of that conforming block.

Alone this doesn't seem like much, but the time and energy limitations it imposes gives a blockchain properties that make data more 'immutable' the longer it is stored in the chain by making it computationally more expensive to change data further back in time from the most recent block in the chain and makes it computationally expensive for a single bad actor to change past data that nodes in the network have 'agreed' upon. This computational expense is important when combined with the other rules that a node uses when receiving a new block and trying to independently determine that the block is valid and should be treated by that node as 'truth'.

The trustworthiness of the blockchain is an emergent network effect. It's not provided or enabled by one single thing, but by many things all contributing at the same time. Proof-of-work is just one small piece of a much larger system. For example, you also need to consider what other rules make proof-of-work useful in the system, such as the rule that a node must accept a chain with more work in it as more truthful than one with less work (hopefully I'm wording that somewhat right!).

Saying 'the miners provide security by hashing' is a massive oversimplification of what's going on. It's very much a sum-of-all-the-parts thing.

I found this helpful in particularly driving this process as a whole home: https://anders.com/blockchain/

I'm still really getting my head around blockchain itself, so if I'm way off base on anything in my understanding here I'm happy to be corrected :)

As I understand it, the challenging question is - is there something else we can use that's as independently verifiable as proof-of-work that's more energy efficient? What independent verification do we give up if we do choose something that's less energy consuming (but maybe less 'provable')?

[nodamage]

> You are changing the nonce over and over again until the HASH of the block meets some criteria. For example, your hash is below some value (it starts with a certain number of zeroes). This is important.

Yes I am aware of this, I simplified a bit so the sentence would read less awkwardly.

[my_ghola]

There was a coin called Primecoin that looked for prime numbers rather than useless hashes. It also went up in the last few months and is going down now like the rest.

[meric]

We could be mining gold instead with the energy.

[Capt-RogerOver]

If you knew the technical details of blockchains, you would not think that. And yes, the process IS much more than wasing massive amounts of energy.

> 1) Surely we can come up with a less wasteful solution to this

You are welcome to try. If you do come up with a less waisful solution you will be able to create your own altcoin and will make billions worth of dollars on it. Some very smart people have tried in the past and are still trying.

[cwyers]

There's already a less wasteful solution, it's called fiat money.

[colordrops]

Fiat money doesn't serve the same purpose as Bitcoin and in some ways is the opposite. Why are there so many trite unthoughtful comments being made about cryotocurrencies? I'm guessing people are turned off by all the scam coins and are throwing the baby out with the bathwater.

[cwyers]

I'm turned off by watching a bunch of people who haven't learned basic lessons about banking trying to reinvent it, badly. Anyone who spent a few days learning about how banks do security could have understood that what these exchanges like Mt. Gox were doing was frightfully vulnerable, and anyone unwilling to spend at least a few days learning about banking security has no business running a currency exchange. But it's all handwaved away because fiat money is evil and cryptocurrency makes you feel like the hero in a Neal Stephenson novel so screw the man. So now we have a bunch of programmers and enthusiasts learning how the financial system works from first principles and making boneheaded mistakes as a result. And they have enough money swirling around thanks to Chinese capital flight that they can screw with the pricing for an entire market like GPUs. And then when you point out how much electricity this whole scheme wastes they say "well banks use electricity" as if that's the same thing. It's heedless, irresponsible and discouraging.

[monort]

Bank Of America Tower requires around 8 megawatt. And it's only one skyscraper.

[root_axis]

So what? Bank of America is a company that provides banking services as a business model, they are not a necessary aspect of fiat currency any more than coinbase is a necessity of bitcoin.

[celticninja]

That's because you are not actually considering the electricity usage of fiat money at all.

[cwyers]

Banks aren't setting themselves on fire in the course of operation:

http://gizmodo.com/a-huge-fire-took-out-a-bitcoin-mining-ope...

Because banks aren't setting themselves up in places based on the electricty cost. And shoving rackmount servers in close proximity and underventilating them. Even huge fintech operations aren't chasing cheap power at the exclusion of other concerns. Banks do not use the kind of electricity that cryptocurrency uses, when you control for the volume of usage of each.

[walterstucco]

You're right

Excel consumes a lot of energy

[nodamage]

> If you knew the technical details of blockchains, you would not think that. And yes, the process IS much more than wasing massive amounts of energy.

Feel free to elaborate...

[stOneskull]

proof of stake

[phillc73]

> ... mining as one of the worst inventions for consuming energy possible?

It depends on what you're "mining." I've been mining Gridcoin[1] for a while, which basically pays crypto-currency for working on BOINC projects. I feel like I am at least contributing to something worthwhile, rather than simply wasting energy.

[1] https://www.gridcoin.us/

[erikpukinskis]

> So we've created an "industry" where you are essentially paid by comverting energy to waste.

But with Ethereum, it's not waste. The miners are providing storage and compute, a la Amazon Web Services. It's not like Bitcoin, where the hashing is just there to verify the blockchain.

In general, it's a valid concern... Do we really need to compute these things over and over to secure a fixed set of computations?

But in reality, the cryptocomputers that make more efficient use of miners will create more value and associated currencies will be more valuable. Bitcoin is essentially a proof of concept and therefore totally unoptimized.

Eventually the cryptographically verified computing market and the trustful computing market will find the sweet spots between efficiency and redundancy. I'm not generally trusting of the free market, but in this case the free market seems perfectly capable of solving this problem.

[sharpneli]

The proof of work part is separate from running the contracts. So it still stands that the mining is only because it allows us to measure in a provable way that someone has thrown X amount of Joules in the gutters.

[pheldagryph]

Surely some of it is a waste. Even for the latest, hottest cryptocurrency. The question is: how much energy is wasted versus the alternative, and does the utility justify it?

Depending on speculatively large future-values of cryptocurrencies is probably not the right way to evaluate the externalities, here.

[verytrivial]

> Call me naive

I'd call you principled. The people scrambling for free money are treating the cost to the rest of the world as an externality. They'll chime in here to say "But what about other terrible system X?", but that, I'm afraid, does not absolve them. You're all terrible!

[conanbatt]

If they are paying for the electricity, they are absolutely responsible for what they are consuming.

[jstanley]

Until we come up with a viable way to secure cryptocurrency without proof of work, such incentives are necessary.

If good people weren't incentivised to spend energy on securing the network, it would be easier for bad people to spend energy on cheating the network.

[dogecoinbase]

Until we come up with a viable way to secure cryptocurrency without proof of work, such incentives are necessary.

This assumes, of course, that cryptocurrency itself is necessary.

[em3rgent0rdr]

"Until we come up with a viable way to secure cryptocurrency without proof of work" -> Until we come up with a cheaper way to trust centralized entities in power.

[austenallred]

Consider that the alternative is all of the banking infrastructure that exists.

The energy that banks and their employees consume is orders of magnitude less efficient.

[pjc50]

Bitcoin does not replace more than a small corner of banking. It doesn't even include any loans.

[lawless123]

lol not sure lending would work if the currency is always increasing in value, why lend money when it will be worth more if you keep it?

[pjc50]

Sure, but that implies that all business and real estate transactions are going to remain using normal non-deflating money.

[arve0]

I would like to see any numbers on that claim. For example expenses / turnover or profit.

Edit: or cost per transaction.

[Nursie]

For the volume of transactions they handle, it's really not.

[nodamage]

This seems like an absurd false equivalence given that the banking infrastructure provides a much larger scope of services than Bitcoin offers.

[RexetBlell]

Proof of work (mining) is needed to be able to come to consensus on the next block in the chain. A better and more efficient algorithm is being developed. Ethereum is switching from proof of work to proof of stake in the next year or two, so there will not be any mining any more, which should reduce the cost of deciding what the next block is. Read more about it here: https://github.com/ethereum/wiki/wiki/Proof-of-Stake-FAQ

[narrator]

Proof of stake comes with its own set of problems.

[my_ghola]

I think mining should not be a way to profit. The rewards are so mining doesn't cost anything to miners. Bitcoin/Ethereum users should be compelled to mine because it means more security that no one controls the blockchain.

I wonder if there's an algorithm that could deter people from building mining farms?

[manquer]

Not that difficult.. you can always choose problems with constant difficulty level or gradually increasing ones adjusted for expected increase in computing power.

However all cryptocurrencies are pyramid schemes . I.e. earlier you get in more you make. That's fastest way to grow and make it viral. However like any pyramid scheme it is unsustainable after some point

This is not necessarily the price point we think it should be valued at though. I and am sure many ppl never expected bitcoin to have mcap of 30$ billion . As difficult it is think today It may even grow another say 30x again to 900 billion or even more. However it will crash eventually like all pyramid schemes when growth flattens and new money - in the form of money spent in mining - stops coming in at the expected rate.

Another way to look at it, for mining to be even marginally profitable price needs to keep increasing for bit coin style algorithms. Either ppl will abandon the coin or price __has__ to increase

[amalag]

Yeah it is not going anywhere and no large institutions will ever "mine" to protect their money or a blockchain. Currency mining is going to be part of a tech-anarchy-utopia dream.

Consensus ledgers are where real money is going to be transacted on.

[mejin]

isn't one of the main reasons that people like ethereum is because instead of solving random hashes, miners run programs that people want solved? or maybe I'm confusing it with another thing

[my_ghola]

They do but in order to run those programs people pay with Ether, the Ethereum coin. And for mining Ether you solve the random hashes.

[jasonlingx]

Energy used = mining reward * (1 - mining profit margin) / unit cost of energy

Where mining reward = (mining block reward + transaction costs + mining premium)

I use "mining premium" to refer to the premium people are willing to pay to mine, i.e. as a hobby, research, or to evade taxes or capital controls. I think when considering efficiency, we can disregard these costs - i.e. energy spent evading taxes is really an inefficiency of taxes, not bitcoin.

The mining block reward will become 0 in the long run.

So really:

Energy used = transaction costs * (1 - mining profit margin) / unit cost of energy

So basically energy usage is being driven by the transaction costs the market is willing to pay, which will adjust accordingly as people are willing to move transactions off-chain.

Looks pretty efficient to me.

[BoiledCabbage]

This formula which I haven't gone through in detail enough to see if I completely agree, directly support the point of waste.

As society lowers the cost of energy, we would expect even more energy to be wasted with all other variables remaining equal.

Halve the cost of energy and now you've doubled the amount of energy spent.

If the cost all energy worldwide improved dropped by a factor of 100, society should explode with new technology. Instead this formula says we'll just spend 100x as much of it mining.

At least with mining gold as technology improves, you gain efficiency in gold extracted per joule. Crypto mining works exactly reverse. The better technology gets, the more wasteful it has to become to compensate.

[jasonlingx]

Would you then say that all that energy from the millions of stars burning in our night sky is wasted?

[h1d]

What else do you suggest to waste to keep a competition? The effort must be entirely conflicting against anyone's interest besides the mining or else it will be an unfair competition.

[tomjen3]

The energy isn't wasted, it is converted. The closest analogy I can come up with is that gold used to back currency isn't wasted because it is stored in an underground vault rather than being used for jewelry or as a component in electronics.

If you are creating a currency, then it has to be backed by something: gold, silver the cost of electricity, people's faith that the currency is worth something. A currency not backed by something will keep deteriorating until it is worthless (assuming a free market, the dollar doesn't collapse because only the us mint can make them).

[Nursie]

>> The energy isn't wasted, it is converted.

The vast majority goes on hashing as fast as possible and getting the wrong answer many, many times... that seems pretty wasteful.

[kortex]

If I pan some dirt and it does not contain gold, I got the "answer" wrong. If I dig a hole and it contains no oil, I got the answer wrong. Seems pretty wasteful.

Scarecity is driven by an inability to readily harvest/create something.

[Nursie]

So what?

If you pan for gold and don't find any, that energy is wasted, not converted. So what? The initial claim wasn't "This facet of mining cryptocurrency is like gold" it was "The energy used is converted". It's not. It's burned.

>> Scarecity is driven by an inability to readily harvest/create something.

Cryptocurrency scarcity is artificial, and not driven by anything other than predefined scarcity curves. Even if two guys with CPU-only miners were the entire BTC network, scarcity would be the same, but energy output would be vastly lower.

The busy-work that is done for (say) BTC is only necessary because of the aversion to central currency authority in certain subsections of society. It's an artifact of that ideology.

[fulafel]

Gold takes untold amounts of natural resources, energy, capital and human labor to dig up just to sit in a vault. That seems the definition of waste.

[walterstucco]

It fits more with the definition of treasure

[leroman]

Don't know how it compares but for "classic" finance to work it needs many workers who use energy, who need buildings to cost energy, who need cooling etc.. The work being done by the miners is not apples to apples comparison with OracleDB sitting behind IBM Websphere server (or what ever) inside a banks data center / cloud. I do hope that in the future somebody comes up with a better scheme..

[dx034]

[1] has a chart comparing Bitcoin energy consumption to that of the Visa payment network. Bitcoin is magnitudes higher.

[1] http://digiconomist.net/bitcoin-energy-consumption

[acchow]

This is only Visa's datacenter costs. What about all the employees working at banks to support all these credit card holders? What about all the physical snail mail being carried around by actual vehicles and driven by humans? What about the entire debt collection system?

[TeMPOraL]

All you've mentioned is secondary infrastructure that facilitates the use of financial system. If ever (God forbid) Bitcoin volume rises to the scale of Visa, it'll have all those costs too. Including the snail-mail shipping of wallet hashes.

[walterstucco]

Programmers are humans too

[dvfjsdhgfv]

You're not naive but realistic.

[pfooti]

But that's 385 / month revenue? On a setup that's what, $3400 to buy, for seven 1070s? Seems like asking for 10 months to cover capex is a bit thin, given the price instability.

[gwern]

Miners shouldn't be meaningfully exposed since they can sell off their ETH as soon as they mine it (deposit it to an exchange address and sell off every few minutes via an API, or something like that). And while ETH might crash enough to make it no longer profitable and risking your capex, you can switch your GPU to mine whatever is most profitable, so you're more diversified than it looks and insulated from the instability.

[TylerE]

If Eth crashes (more) and bitcoin continues to drop, no crypto is going to be worth anything.

[AngryData]

Bit coin isn't dropping, yeah it had a small peak the last month but it is still at $2300 which was only first achieved in in May.

[flashman]

Resale value on the GPUs is also worth monitoring. While it remains high enough, you can discount much of the capital cost because you'll get a high fraction back if you decide to sell the cards.

[stouset]

I doubt resale values on GPUs running at 100% for months on end is going to be high.

[snuxoll]

Miners don't want to push their GPU's that hard, a hardcore gamer may overvolt their card to get a more stable overclock but a miner will undervolt it to run it cooler and increase the life of the card.

I'd rather buy a used Fury X off a miner than a gamer.

[h1d]

Gamers might play 3 hours a day but miners run 24...

[peteretep]

You are much more optimistic that me if you think they'll come with a usage disclosure. There's a reason cars have immutable(ish) odometers.

[DanBlake]

High risk, high reward. If you buy the average piece of real estate and rent it out, you are looking at 30+ years to cover the cost.

You can also lose it all (housing bubble) or quadruple your investment (investing in SF 10 years ago)

[creyes]

also weird analogy cause the houses value will (theoretically) appreciate... a GPU prolly not so much.

[ucho]

Buying ASICs is high risk and (probably) high reward - in event of crash you are left with worthless machines. GPUs will still have some value, even after all crypto currencies drop.

[numbsafari]

Which is why you use leverage...

[raverbashing]

Barring natural disasters and other extreme events, your real estate risks losing (let's say) 50% of value if you made a good purchase

The intrinsic value of 1 BTC is ZERO

[Zpalmtree]

But you bought the GPU not the BTC?

Even if BTC becomes worthless, you still have the GPU to sell

[raverbashing]

True, that applies for investing in mining

(there's also the electricity you used, but if you live in a cold place you saved on heating so that could be something not lost as well)

[mschuster91]

> On a setup that's what, $3400 to buy, for seven 1070s?

Well you can always resell them on the used market after stopping the mining operation and get a decent chunk of the money back...

[beedogs]

Or just, you know, buy ETH with money right now.

[jameskegel]

That would have got you in trouble a week ago.

[justonepost]

GPUs don't depreciate that fast.

[lossolo]

You forgot to tell people about risk of rising difficulty. It's not 385$ a month forever, if difficulty will go up (as it steady goes up every month) you will mine less.

And you can't get 230 MH/s with 7x GTX 1070 with 1kw of power draw. More like 215 MH/s and with around 1.2 kw. I know because I have mining rigs with 7x GTX 1070s.

[JohnTHaller]

People constantly seem to get into mining by looking at the calculator at CryptoCompare without reading the last bit which states that "Block reward is fixed at 5 ETH and future block reward reductions are not taken into account."

Using CryptoCompare and assuming your 230MH/s of mining ability will net you $470 a month (at current ETH of $165.47) people assume they make $5,640 a year. That sounds great!

If, however, you use something like MyCryptoBuddy's calculator, which takes into account pool fees and difficulty increase, you realize you're only going to make $1,553 in a year. Which is a rather paltry return if you shell out for 7 GTX 1070s.

[bogomipz]

>" If you are located in the 'mining valley' of Washington where power is ~2c/kwh you are still getting healthy profits"

Can you clarify what "mining valley" you are referring to? Is this Chelan County?

If so I thought there was talk about a moratorium on new high load customers but maybe that never came to pass? See:

http://www.coindesk.com/bitcoin-miners-public-utility-spar-o...

[nulagrithom]

Not the parent commenter but yes, Chelan and Douglas counties have fantastically inexpensive kwh due to their hydroelectric plants.

The dams are owned by a public entity and the cheap power and plants are meant to be a boon to the local economy. Last I heard (directly from a city council member last year) there is a moratorium on high load customers due in part to all the mining operations that sprung up. Realistically, for the Public Utility District it's been much more profitable to sell the power to other states than it is to allow more mining operations to eat the cheap power.

There were high hopes for the data centers that sprouted up in neighboring Grant county, but they didn't bring anywhere near the number of local jobs that were expected (leaving a bunch of empty houses in Quincy). That in combination with the closure of the Alcoa aluminum plant (which purchased the _entirety_ of one of the smaller dam's output) and you shouldn't expect the moratorium to be lifted any time soon.

[bogomipz]

Thanks for detailed response. So it sounds like the power arbitrage game is done in Washington(at least for now.) Although I imagine this is good news for those miners who are grandfathered in.

[marblar]

It seems like if you were located in a "mining valley" the last thing you would want to do is have it mentioned in a top comment on Hacker News.

[ikeboy]

Average price paid for electricity in many regions is 20 cents/kwh. Just using your numbers that translates into $300/month in costs for under $400 in ETH.

At the national average it would be $200/month by rough math.

[bdcravens]

In Texas, 8-12 cents is more likely.

[nine_k]

In downtown Manhattan, maybe. In most other places near a power station, it's not.

[fapjacks]

Southern California and its local electrical monopolies, for another.

[kobeya]

2 cents, not 20.

[ikeboy]

It depends on region, some are 20 cents, national average is 13.7 according to https://www.bls.gov/regions/new-york-new-jersey/news-release...

[kobeya]

But the regions where people are hosting miners are between 2-4c per kWh. Those are the competition, and those are the people buying (and selling) these.

[ikeboy]

I was responding to

"But even if you dont have a facility in washington and just mine from your apartment, your power cost would probably be $100 a month."

Even if they're only paying the national average, it's still off by a factor of 2.

[zanny]

2 cents is an abnormality. Europeans pay around 20 cents per kwh on average.

[dgacmu]

2 cents is what you're competing against -- very large mining farms located near cheap, under-utilized hydro plants. Many in China, due to the existence of many such plants. [0] Those are the mining operations that can remain (marginally) profitable when the small home miners are forced out because gross revenue is less than power costs.

[0] http://www.afr.com/technology/lisa-kangding-story-20160706-g...

[kobeya]

It's 3-4 cents in China. 2.3cents in Quebec. 2cents in Washington State (which is what the GP was referencing).

[nodesocket]

Sorry noob question, but if you don't pay for power at your apartment could you run a rig for just the cap x to get started?

[zachlatta]

Yes, but this is how you get evicted.

[jasonzemos]

And I can attest to learning this lesson personally in 2012.

[nodesocket]

lol, sorry, but how long until your landlord contacted you? What was the conversation like? I'd would have said, sorry running servers for my internet business.

[twblalock]

> I'd would have said, sorry running servers for my internet business.

Why do you think that would have helped? The landlord cares about the fact that he/she is paying more money because power consumption is high, not about why power consumption is high.

[iak8god]

The landlord will very much care if the reason you're using so much power in your residence is that you're running an energy-intensive business there. In many states this would give the landlord really solid grounds to toss you out immediately, especially compared to the situation where you're just suddenly using more power because you just like to set your AC to 60F in the summer, for instance.

[Mtinie]

Do many people live in units where electricity is provide as a gratis benefit? That was never my experience in California, Washington, D.C., nor in Virginia.

[jwhitlark]

And your landlord would have said they're renting you living space, not office space. Of course, you could always offer to put a meter on the servers, and pay them the difference.

It's somewhat disheartening that you've founded two businesses, yet you're looking to the letter instead of the spirit of the agreement.

[pcwalton]

And then you'd get evicted, and rightly so. You are in no way entitled to have your landlord fund your cryptocurrency speculation.

[mcintyre1994]

I've lived in a few house shares where power/bills are all included in rent, but the contract forbids running a business from the premises. I assume they wouldn't particularly care if I had a small software business or whatever, but if the servers were in the house and they were getting billed they most definitely would.

[bdcravens]

A lease can have a clause forbidding commercial use of the property.

[yclopediaen]

You'd also get monitored for a being a likey weed grower.

[AngryData]

They might think you are growing weed and evict you if you draw too much.

[herpdorplarp]

But that's still break even in about 10x July 2017. How many times will difficulty double until July 2018?

[blairanderson]

Where exactly is the mining valley? Very few articles showing up about it...

[nulagrithom]

Wenatchee Valley. KWH are somewhere just above 2 cents. Lots of electric plants down the Columbia river, and Wenatchee has great 1 GBPS connectivity throughout.

[grogenaut]

Where do I check my raw kwh price? just on my bill? I'm in WA as well.

[Sleeep]

It will be on your bill.

It should also be published on your utility company website as well. At least my utility company posts rates on the front page.

[TokenDiversity]

What's the breakdown between power drawn by card and cooling? You seem to be suggesting they are equal...

(1kw would cost $14.4 at 2c/kwh) and you say $30.

I don't know much about stuff but i didn't know cooling is that expensive.