MS: Bitcoin mining uses as much electricity as 1M US homes

https://twitter.com/FatTailCapital/status/899714838796148745

by pulisse • 9 years ago

79 95 9 years ago

twitter.com

cbanek2• 9 years ago

Is that more or less than the amount of power used to grow marijuana indoors?

https://www.theguardian.com/us-news/2016/feb/27/marijuana-in...

dajohnson89• 9 years ago

Consider also that residential power consumption is a minority percentage of the total (vs manufacturing and commercial).

cbanek2• 9 years ago

True, although I imagine the legal dispensary grows are all on commercial rates, and those are pretty big.

cgb223• 9 years ago

I wonder how much that contributes to something like global warming

It would be pretty strange if bitcoin had a noticeable effect

alexkcd3• 9 years ago

Proof of work systems are, at the core, a race towards ever greater energy consumption. They're an environmental disaster waiting to happen. Surprised how little attention this gets.

I would argue that the benefit of decentralization is not worth the price.

sbenitoj8• 9 years ago

What would you say is the value of decentralization?

otheotheothe2• 9 years ago

Freedom?

On some stuff you cannot put a price tag - like mastercard ads say, some stuff is priceless.

alexkcd3• 9 years ago

I'm not advocating for decentralization. Just offering that up as a common argument that PoW advocates use.

I think that in reality, PoW systems become more centralized over time. You can already see that with Bitcoin. Eventually the gatekeepers become those with the largest energy resources available to them, and "we the people" will be in the noise when it comes to hashing power.

I'd rather take energy efficient centralization than faux decentralization that also happens to destroy the environment.

Sir_Substance• 9 years ago

Same as the value of free speech. Civil forfeiture is a great example of a chilling effect based on seizure of property. There's nothing illegal about buying a car with cash, but you might not want to because if a US cop stops you and searches your car, he gets to just keep your money, so people think "eh, maybe not, wouldn't want to be mistaken for a drug dealer".

It's not healthy for a society when people start self-censoring their actions like that. If nothing else, it places banks in a privileged position where they can get away with more predatory behavior, because people are more likely to see them as a necessity.

It's for this reason that anonymous transactions are valuable. You really should be able to buy legal marijuana or books about firearm construction without worrying that people might judge you. Neither is illegal, if you want one (or both) you should have the freedom to purchase them. However, many people would choose not to buy them, simply because they know the bank might see it on their balance sheet if they ever have to call the support line, and they're worried about follow-on consequences to their insurance premiums or their job if knowledge about that purchase ever circulates. That's a chilling effect in action.

weavejester• 9 years ago

PoW isn't the only path to decentralisation. A proof of stake system is also possible; or a hybrid approach, where PoW is used to bootstrap the currency, and PoS incrementally takes over after a certain hash-rate.

sbenitoj8• 9 years ago

Maybe, hopefully they figure it out, Vitalik Buterin et al are certainly working on it -- but there is no proof yet that a Proof of Stake system can function at scale like a Proof of Work system can.

geewee• 9 years ago

It really seems like a race to the bottom. We produce specialized hardware, stick 'em in giant mining farms where they devour power - for what? Crunching arbitrary numbers. It just seems incredibly wasteful.

twblalock2• 9 years ago

Bitcoin advocates ought to understand that the whole mining thing, and the increase on the rate of difficulty in mining, and the eventual limit on the number of bitcoins that can exist, are just as artificial as anything that happens with fiat currency, and controlled by a much smaller number of unaccountable people.

AlexCoventry3• 9 years ago

Compared to the military and law enforcement which back a regular currency, it's incredibly efficient. Especially when you count the human cost of conventional power politics in terms of oppressed minorities and corruption.

urza5• 9 years ago

Common misconception. I let the great Andreas to educate you: https://youtu.be/rsLrJp6cLf4

bdcravens3• 9 years ago

You're assuming people know or care who Andreas is. Many aren't glued to /r/Bitcoin. A common misconception is that the various Bitcoin heroes are well known.

lsd5you• 9 years ago

Not convinced. He seems to be preying on willing believers like an MLM salesperson. Mixing truth with omission.

SkyMarshal• 9 years ago

Clean energy production is becoming increasingly viable and cost competitive. CA already produces surpluses of solar power at times [1]. A significant portion of electricity consumed by Bitcoin is already clean energy - Chinese hydropower, Icelandic geothermal power, etc.

The hand-wringing over Bitcoin's energy consumption, and suggestions of sacrificing decentralization for it, I'm not sure are justified just yet.

[1] http://247wallst.com/energy-economy/2017/07/03/why-californi...

MagnumOpus• 9 years ago

significant portion of electricity consumed by Bitcoin is already clean energy

Electricity is very fungible, and hydro locations are limited. This means that when the (very limited) hydro energy is converted into heat to mine bits because it is easier to colocate some ASICs next to the dam than an aluminium smelter - then the smelter will run on slightly-pricier coal instead. The marginal carbon emissions can still be blamed on the coin mining.

noonespecial• 9 years ago

Some delightful trains of thought along these lines have an answer to the Fermi Paradox as their last stop...

mathgenius• 9 years ago

I'm not even convinced it _is_ decentralized. Seems like mining is clustered around a few big operations (mostly in china) that have access to cheap electricity.

sbenitoj8• 9 years ago

Even if it's not perfectly decentralized, the fact that users can fork and create their own cryptocurrency acts as a fail safe. Miners with > 51% mining power would only be destroying their own long term profitability if they destroy the network.

Synaesthesia• 9 years ago

It’s but a tiny fraction of the environmental damage the US mulitary creates, with it’s reliance on oil.

__warlord__• 9 years ago

I know this proof of work is not efficient (energy wise) but I think is worth the price to get rid of centralized entities. I hope with time we will find more efficient ways to handle the decentralization but for the moment is a good start.

anandsuresh3• 9 years ago

The proof-of-work algorithm is an expensive proposition. I wonder if there are ways to utilize all that power and compute to perform useful work... like reCAPTCHA. The desired characteristics would include to ability to dynamically modify the difficulty of the problem.

hamandcheese• 9 years ago

I really like this idea. Perhaps transaction fees could be paid by those needing compute, and the result might be a globally distributed cryptocurrency AND a huge pool of (hopefully) cheap compute resources.

anandsuresh3• 9 years ago

Exactly! I don't expect this to work for every case, but if we can re-think even a small subset of socially-relevant computation problems as a PoW scheme, that could be a game-changer!

otheotheothe2• 9 years ago

It absolutely is USELESFULL work, its protecting a 40 BIO USD Marketcap, financial freedom, censorship resistant payments...

Recaptcha and co are gimmicks compared to this.

anandsuresh3• 9 years ago

I wouldn't go so far as to call reCaptcha a gimmick. It is an excellent example of how a slight modification/re-mapping of the original problem-space can solve more socially/commercially useful problems.

PoW has proven itself in the field for years, while PoS is still untested at scale. Being able to piggyback scientific/medical/environmental computation over a PoW scheme would provide significant value as opposed to a fairly useless compute problem such as the double-SHA256.

The key here is to be able to re-interpret/re-state a problem in terms of the PoW algorithm with the following characteristics: - Expensive to calculate, cheap to validate - Dynamically-adjustable difficulty

An ideal implementation of this would be similar to an OS running an idle process (double SHA256), and switching to running user-processes (medical/scientific computation) when needed/instructed.

kalleboo2• 9 years ago

I think there are a few that have tried. E.g. Primecoin calculates prime number chains https://en.wikipedia.org/wiki/Primecoin Looks like there are some protein folding coins as well - curecoin and FoldingCoin that contribute to folding@home

ATsch• 9 years ago

one such example would be https://en.m.wikipedia.org/wiki/Primecoin

The big issue with making a POW is not making something hard to calculate, but making something hard to calculate and easy to verify. As far as I can tell, useful tasks that satisfy those conditions are very hard to find.

dplarson2• 9 years ago

I'm having trouble finding the source of the figure shown, but I did find a page with similar information ("Bitcoin Energy Consumption Index"): https://digiconomist.net/bitcoin-energy-consumption

mrb• 9 years ago

I have produced what is to my knowledge the most precise estimate of Bitcoin's energy consumption: http://blog.zorinaq.com/bitcoin-electricity-consumption/ That other index ("Bitcoin Energy Consumption Index") is famously flawed, and its author is strangely defensive. I wrote a critic: http://blog.zorinaq.com/serious-faults-in-beci/

The chart https://twitter.com/FatTailCapital/status/899714838796148745 was produced as described in the footnote (mining hashrate × efficiency linearly declining from 1.5 to 0.2 J/GH), which IMHO fairly represents the "upper bound" of the energy consumption. The real consumption is likely lower. For example my study estimates the current efficiency is between 0.100 and 0.195 J/GH.

To give you an idea, "1M US homes continuously" is roughly the annual electricity consumption of decorative Christmas lights in the US.

dplarson2• 9 years ago

Thanks for the info and context!

0xbear3• 9 years ago

I only mine during late fall and winter, when heat is not wasted. I wonder if they took that into account.

czechdeveloper• 9 years ago

Most mining are centralized factories where such thing is not even possible, so I believe it would not even make sense to put your case into account.

kalleboo2• 9 years ago

Well, it could be possible (district heating is a thing in many places)

pkulak2• 9 years ago

Resistance heating is still pretty wasteful.

hueving• 9 years ago

How so? Do you just mean electric heating in general?

colordrops2• 9 years ago

Isn't a lot of mining done in places where electricity is cheap and/or clean, such as hydroelectric in China and geothermal in Scandinavia?

zakk3• 9 years ago

57.2% of Chinese energy is from coal.

https://en.wikipedia.org/wiki/Electricity_sector_in_China

Also it doesn't seem to me that Scandinavian countries are particularly active in mining. Understandably, since electric energy there isn't cheap.

urza5• 9 years ago

Yes it is. Chinese mines are mostly hydro powered. On iceland is running huge mining operarion on thermals.

zakk3• 9 years ago

No, more than half of Chinese energy is produced from coal.

https://en.wikipedia.org/wiki/Electricity_sector_in_China

I did some research, and it doesn't seem that the mining farm in Iceland is even comparable to Chinese ones in terms of hashrate. Do you have any source?

analog31• 9 years ago

It displaces other uses of the energy that could also be moved to the same location. My thought is to first make sure the coin mining isn't subsidized, and perhaps to tax it.

kamaal2• 9 years ago

>Isn't a lot of mining done in places where electricity is cheap

I think the argument is you can put that energy to better use than mining bitcoins.

coryl2• 9 years ago

Whats the alternative cost of printing, storing, protecting, and maintaining an equivalent cash value? Is that a fair comparison?

bdcravens3• 9 years ago

Then you'd have to factor in cost of running exchanges as well as storing for those who maintain their own wallet.

coryl2• 9 years ago

Regardless, an economy running on say $100b of cryptocurrency is more cost efficient than one running on cash, is it not?

I mean, as a matter of principle, the transaction fees of a bitcoin transaction are always cheaper than any credit card swipe, ATM withdrawal, or wire transfer.

krfsm• 9 years ago

If you don't include the cost of mining hardware, you're probably not making a fair comparison.

(Then you have the philosophical question about what is a reasonable comparison target.)

modeless4• 9 years ago

I am always disappointed when this discussion comes up, because nobody considers what will certainly happen in the future. This situation, where a large percentage of Bitcoin's value is spent on mining, is temporary.

Bitcoin's mining reward halves every couple of years. When the Bitcoin block reward becomes negligible, as it certainly will in within a few decades, the only thing funding Bitcoin miners will be transaction fees, so the entire cost of Bitcoin mining will be borne directly by users of the network.

The amount of money spent on electricity for mining is exactly the same as the (perceived) value of the mining rewards. Therefore, in the future the amount of money spent on electricity for mining will certainly go down as a percentage of Bitcoin's value.

If Bitcoin becomes very popular and valuable, transaction fees may be significant in absolute terms, but they will always be a very small percentage of the total, and likely smaller than the percentage of economic output consumed by the traditional financial system (in the future when the block reward is small).

beebmam• 9 years ago

I think that's one of the biggest flaws of bitcoin and will eventually lead to its collapse.

I simply don't want to invest in mining considering how much people have already mined. There's deflationary pressure for many reasons, but definitely because of the rewards being cut in half every couple of years. And currency deflation is highly unappealing for those who don't have large amounts of that currency.

Psychologically this is bad design, and it turns a lot people off from it. It makes it feel like a scam. That's good enough to piss enough people off about it to never use it, including me.

ringaroundthetx3• 9 years ago

It makes it feel like a scam.

Under what definition?

Is this definition a standard that couldn't be applied to other currency systems or asset classes that you respect?

There's a reason I ask so specifically

modeless4• 9 years ago

And currency deflation is highly unappealing for those who don't have large amounts of that currency.

For an individual investor making a decision about which currency to hold, all else being equal, deflation is always better than inflation. Not sure what you're trying to say exactly.

abhisharm• 9 years ago

You're describing a cognitive bias wherein you would be turned off from a monetarily beneficial decision because of the foregone gains you would've had had you made the decision to enter the game earlier. This won't lead to the collapse of bitcoin, it just means that those who will profit from mining now and in the future are those who won't fall prey to that sort of thinking.

kevinwang• 9 years ago

Bitcoin's mining reward halves every couple of years. When the Bitcoin block reward becomes negligible, as it certainly will in within a few decades, the only thing funding Bitcoin miners will be transaction fees, so the entire cost of Bitcoin mining will be borne directly by users of the network.

I don't think this logic is sound. The cost of bitcoin mining is already borne by users of the network (aka miners)? What does the diminishing block reward have to do with anything? The end result is that if you do calculations, and perform proof-of-work, you receive bitcoin. It shouldn't matter to the miners if it comes from block reward or transaction fees.

kobeya• 9 years ago

"Users" is a much bigger set of people than miners (a supermajority of which would fit on a couple of couches on a stage).

modeless4• 9 years ago

It shouldn't matter to the miners if it comes from block reward or transaction fees.

The point is that transaction fees will never be as high as current block rewards. The cost of bitcoin mining is currently subsidized by inflation of the Bitcoin money supply. That inflation will end, and it will not be fully replaced by transaction fees.

alexkcd3• 9 years ago

So long as mining remains profitable, as it has to be in order for anyone to spend resources on mining, the energy race will continue. As mining profits decrease over time, so too will the pool of miners who are able to mine profitably.

The network will, as a result, centralize over time to a select few that have access to cheap and plentiful energy. These "big energy" :) companies will control most of the hashing power.

You can already trace this with bitcoin by looking at who the early miners were versus who the miners are now.

twblalock2• 9 years ago

Bitcoin's mining reward halves every couple of years. When the Bitcoin block reward becomes negligible, as it certainly will in within a few decades, the only thing funding Bitcoin miners will be transaction fees, so the entire cost of Bitcoin mining will be borne directly by users of the network.

Just let everyone mine as much as they want. Don't limit the rate or the amount. All of a sudden, miners will have to self-regulate their output lest the currency become hyperinflated.

You might not like that outcome, but if you say that is going to be the endgame anyway after a few decades, why not let it happen now?

modeless4• 9 years ago

If you're asking why Bitcoin mining had to be heavily subsidized at the start, the reason is that Bitcoin had little value and little usage, so transaction fees were generally zero and a large mining incentive was necessary. Obviously that isn't true anymore, but the exact time inflation should end could not be foreseen in advance, and the inflation schedule was fixed at the start.

If you're asking why the inflation schedule had to be fixed in advance, that's because otherwise there is no trustless way to ensure it stays in check. The genius of Bitcoin is the way it uses incentives to enforce rules in a distributed system made up of selfish and even malicious participants. As you say, miners would have to self-regulate against their own self interest, and that would not be stable.

linschn• 9 years ago

But the point is the rate is limited (1 block per 10 minutes) and the amount is also limited (by the value assigned to the transaction by the users). There will be no runaway mining, and the miners do not need to 'self-regulate'.

billytetrud2• 9 years ago

Wow, what a terrible idea. Do you not know how inflation works? If a bank prints money for itself, do you think it worries how inflation is affecting its holdings? No, they have more holdings because they're the infiltrators. They have more monetary even as they devalue the currency. This is why Bitcoin is necessary. This is where bitcoin shines. Deflationary currencies are the future. Don't be that idiot that pretends temporary deflation events are the same as a currency that is intentionally deflationary.

kamaal2• 9 years ago

>If Bitcoin becomes very popular and valuable

For this to happen, you need this to be a fiat currency.

Other wise the rest of the world has no obligation to trade in a currency where there already exist people who own bulk of the wealth and others enter poor by default. The non-bitcoin crowd can go to a new currency, which doesn't have mining kind of a setting.

At that point in time bitcoin will lose its value rapidly and all the money spent mining and acquiring bitcoins would have gone for nothing.

There is reason why currencies have to be backed by governments with massive militaries. If they didn't the poor can simply walk out the current scheme, leaving the existing ecosystem in tatters.

Fiat currency is a necessary evil. People who come later, care nothing for people with prior advantage. Bitcoin's goals are too ideal to work in the real world.

skrebbel• 9 years ago

As usual in any environmental debate, all sides choose deceiving stats. In most western countries, homes only account for about a quarter of total energy consumption [0].

So while people reading this headline would think "wow, that's the total energy use of a town like San Francisco!", in reality, it's more like the total energy usage of a town like Fremont. Less spectacular for sure.

Remember to use this! If you're pro-environment, always express your energy consumption numbers in terms of "homes", for a 4x more impressive result. After all, who care about the truth when the planet is going to shit (note: the anti-environment folks, or whatever they call themselves, do the exact same thing but I can't come up with a nice example right now).

This also means that if you see someone express energy usage in terms of N homes, you know they're ok with twisting the numbers to make a point. Read the rest with a grain of salt.

[0] http://ec.europa.eu/eurostat/web/products-eurostat-news/-/DD...

choonway• 9 years ago

I wonder what would happen when quantum computers manage to break Bitcoin.

pmuk• 9 years ago

Please correct me if I'm wrong but does this mean it takes around 3-4 households worth of daily energy use to confirm a bitcoin single transaction?

https://blockchain.info/charts

westurner• 9 years ago

So, clean energy incentives.

That means 1.2% of the Sahara desert is sufficient to cover all of the energy needs of the world in solar energy.

https://www.forbes.com/sites/quora/2016/09/22/we-could-power...

Nearly all other animals on the planet survive entirely on solar energy.

https://en.wikipedia.org/wiki/Solar_energy

billytetrud2• 9 years ago

This isn't at all true. 1.4 terawatts would only power about 230,000 homes. And other sources have estimated bitcoin's energy cost at significantly lower: 330 megawatts: https://securitygladiators.com/2017/03/15/bitcoin-uses-energ... .

Also think about the cost of normal money. Just printing currency costs the US over $700 million per year: https://www.federalreserve.gov/faqs/currency_12771.htm . This is not to mention the rest of the infrastructure.

Last point, Bitcoin mining won't get much more energy intensive. Why? Because the amount of energy used in mining is directly proportional to the value of mining a block. As soon as segwit is activated tomorrow, the value of mining will drop. As soon as the lightning network funnels 99% of transactions off the blockchain, the value of mining will drop. As soon as bigger block sizes are allowed, the value of mining will drop. Bitcoin doesn't have to be any more "wasteful" than is needed to secure the blockchain.

urza5• 9 years ago

I wonder how much energy is consumed by wallstreet and all banks worldvide, including the work wasted by all the employees, money printing facilities, and all this other things required for fiat money to have any value. I bet you bitcoin minig is drop in the ocean compared to it.

MS: Bitcoin mining uses as much electricity as 1M US homes

Related Stories