Photos: Life inside of China’s massive and remote Bitcoin mines | Hacker News Reader

Photos: Life inside of China’s massive and remote Bitcoin mines

qz.com

180 pointsurahara183 comments

If you have even the slightest concern about global warming or the environment overall, you should consider Bitcoin a horrible mistake we should be trying to eradicate.

It's a currency backed by willingness to waste electricity doing useless work. And worse, as it gets more popular, the willingness and price will continue to drive higher.

Hopefully China recognizes that this is in odds with their efforts to clean up their environment, and crushes Bitcoin fast before the damage is too catastrophic.

I think it's not entirely valid to say it's useless. Perhaps overly energy intensive, but the whole point of mining is to verify the blockchain, it's not a mindless calculation (though it is artificially inefficient). This is supposedly what the Ethereum proof of stake will fix. The problem is that you have two competing goals: you want fast hashes to keep transactions quick, but since you incentivise hashing, you don't want people to be able to exploit it to print money.

Currently, we have around 6M Thash/s. If you pretend that it's all using an Antminer S9 (1.3kW/14Thash) you end up with around half a gigawatt of power running the Bitcoin network. So over a year that's about 4.4TWh per year. In reality, few people have S9's and it's more like 15TWh per year. China alone uses around 4000TWh per year for comparison.

That is a staggering amount of power, but the global power consumption is perhaps 1/1000 of China's (or the US's for that matter) annual energy usage.

What is the equivalent for, say, VISA or Mastercard processing? Somewhere there are a bunch of enormous datacentres which are processing transactions. I would imagine a lot less (my back of the envelope suggests around 1GWh/year).

EDIT: Actually they have VISA's consumption here, so I wasn't far off: http://digiconomist.net/bitcoin-energy-consumption. I estimated each data centre (x2 with a back up each) to use 8MW of power - they have 2x diesel generators quoted at 4MW. That works out to around 0.25GWh of power per datacentre per year. I assumed that there's probably more overhead elsewhere in the system (i.e. POS readers, etc.) and doubled it. I don't know if their analysis accounts for the fact that each centre has a backup running simultaneously.

Also because people seem to be getting the wrong end of the stick here: where did I say that Bitcoin is comparable to VISA? I'm putting the numbers there to show how massively disparate the two networks are. Partly out of curiosity, and partly because you can't just quote one number without a reference point and I think a large finance company is a good place to look. Nor am I saying that bitcoin's inefficiency is somehow justified - it uses more energy than a small country.

First -- that's even worse because VISA powers probably 1/3 of all modern commerce. Bitcoin is a hobby of the .01%.

And even if that wasn't the case, it's not comparable because bitcoin is _fundamentally backed_ by the willingness to waste electricity. This is very important, and I feel like people gloss over this.

If VISA figured out how to run their datacenters without electricity, they would do it in a heartbeat, and it would have no impact on the amount of processing being done -- they'd just do the work cheaper, and it would be great.

If bitcoin mining got more energy efficient, THERE WOULD BE MORE MINERS BECAUSE COSTS DROP. The amount of money miners are willing to waste on hardware and electricity is fundamentally tied to the price, and as the costs go down, it's easier to fight for the fixed-sum payout. This will never change. As bitcoin gets more popular, a proportionally large amount of electricity is WORTH BEING WASTED.

It is a fundamentally broken model, if you are trying not to rape the environment.

This is assuming the value that the VISA system provides is the same as what Bitcoin's does. Bitcoin is being used more as a store of value these days. The ratio of the value it provides to electricity requirements is much, much greater when taking this into account.

What value are you talking about exactly?

Newer designs offer the same service with improvements like scrypt and other algorithms to prevent ASIC attacks which just centralize control to hardware designers/producers, like these warehouses in China.

Exchanges can and have artificially inflated prices:

http://www.coindesk.com/bot-named-willy-did-mt-goxs-automate...

Came here to say this.

A more accurate comparison would include costs related to storing and securing wealth (banks, guards, etc)

You're off your rocker - a "hobby of the .01%". If you can't see the potential of a decentralized cryptographically secure store-of-value/currency then you are incredibly short sighted.

The point is that almost nobody uses bitcoin, and it still uses as much electricity as VISA (or more). If it became the currency of the masses, we'd spend half the world's electricity farming blocks.

> If it became the currency of the masses, we'd spend half the world's electricity farming blocks.

Yeah, some rough math on current market cap suggests not quite half but probably more than a tenth.

But the problem is saying the 0.01%. Bitcoin is certainly used by a small fraction of people, but it's not at all a rich-person thing.

i think the electricity use is a bigger problem.

And traditional currencies have been cutting down trees for hundreds of years to process them into paper. Isn't that incredibly inefficient? Think about all the energy that is wasted to get the heavy metals out of the ground to produce coins?

That's not even including the fact that every retail store I've worked at never turn of their POS terminals or the 24hr ATM's. The gross waste of having to build buildings to store money in?

Is the traditional banking system really any more efficient?

Not to mention its not exactly wasting electricity at all its just inefficient by design for security purposes.

> Bitcoin is a hobby of the .01%.

It's still relatively easily accessible to the middle class. It also allows the middle class in regimes with unstable currencies to more easily transfer and store value.

I think you're reading too much into my post. I was simply curious to see what the actual numbers were. There's no point saying "Bitcoin is bad." without any justification: so I justified it - it's insanely inefficient. And you need that extra data point (VISA) to see just how inefficient it is. My favourite horrifying fact is that if Bitcoin were a country ranked by energy consumption it would come in at around North Korea.

>Bitcoin is a hobby of the .01%

Do you mean of the .01% of the richest people of the population? If so, you have no clue what you're talking about.

You could argue that energy is being wasted in traditional currency systems. Printing currency, ATM machines, cash registers, transporting cash in armored cars. This energy doesn't serve any purpose besides maintaining the currency itself, similar to bitcoin.

There are upcoming optimizations that will allow the bitcoin protocol, via layer 2 solutions, to run at the same capacity as visa.

Yes, it really is possibly that bitcoin will have a higher capacity for less energy costs.

Bitcoin feels like digging a mine with plastic spoons. Technically speaking the energy is not useless, since it does go into digging. But at the same time there are far more efficient ways to do it, so "useless" is a pretty decent way to describe the vast excess.

The comparison with VISA is bonkers once you account for the different sizes involved. Bitcoin uses 26,000 times more energy per transaction. If energy per transaction remained constant while Bitcoin scaled to the size of VISA (a dubious assumption, to be sure) then Bitcoin's total power consumption would be 390,000TWh/year, several times more energy than is consumed by human civilization worldwide at present.

Edit: it sure would be nice if we could just price the externalities into energy costs, then all of this would just work itself out and we wouldn't have to worry about it.

All the electricity spent isn't to do the transactions, it is actually a security cost to prevent 51% attacks on the blockchain. Amassing the amount of miners and electricity to do that attack becomes immense, hence the security. Hopefully in the future Ethereum's proof of stake implementation can solve the problem without wasting electricity.

The purpose of that security is to protect the integrity of the transactions. Without transactions, the network is pointless. The energy used for mining should be charged to the transactions, just like the energy VISA uses to light its offices would be.

> The energy used for mining should be charged to the transactions

I mean, it basically is.

The built-in rewards drop every couple years, so the only reason people will continue to mine is for transaction fees.

What are you suggesting should be different?

I'm suggesting that a different security scheme should be used, one which doesn't depend on wasting a shitload of electricity.

As I mentioned in my edit, an easy way (in terms of coming up with the solution, not implementing it) would be to price the externalities into the cost of energy. Then the electricity would only be used if it was actually worth it.

Not wasting electricity is a good idea.

But I don't understand how transaction fees relate to it. That just means a different person pays to waste electricity.

And pricing in externalities won't actually stop people from wasting power, so you'd better have a good system set up to use that money to counteract the externalities.

I'm not talking about explicit transaction fees, I'm just talking about the energy use of the system on a per-transaction basis. Looking at the absolute energy consumed by Bitcoin isn't enough by itself, because you also need to account for how big Bitcoin is.

Pricing in externalities will absolutely stop people from wasting power pointlessly. People will only use power if the benefit is greater than the cost. This includes "waste" if the effort of stopping that waste is greater than the cost of the waste, but if externalities are priced in appropriately then that means there's actually a net benefit to that. It's only a problem if the cost to the world is higher but the person paying the bill doesn't see that.

In terms of bitcoin, if the electricity to mine one costs X on the miner's power bill, and that bitcoin can be sold for 2X, then the miner will keep on doing it even if the total cost of that electricity including damage to the planet and other such things is 3X. If you price in the externalities so that the power bill actually says 3X, then the miner will stop. You don't actually need to use the extra money to counteract the externalities.

But as some miners stop, the efficiency of mining goes up again.

In a general sense, the amount of money spent on mining will stay constant. If you make electricity 3x as expensive, then maybe 1/3 as much electricity will be wasted, but it's still fundamentally a waste.

If you don't use that extra money to undo the externalities, then the world is still getting more polluted just so bitcoin can spin in place.

This isn't something like a power plant where charging $X per ton of CO2 released will make them cut emissions to 0. You would need a per-bitcoin fee to stop waste. Charging per-kWh has a limited effect, because people will always buy [bitcoin price] * [bitcoins per day] worth of electricity.

If pricing in externalities increases the price goes up by a factor of three and bitcoin mining continues at 1/3rd the energy usage then it's no longer a waste. As long as you've done it correctly so that the price is accurate, then people will only use that energy if the benefit it creates is greater.

Using energy for some benefit is fine, and bitcoin clearly has some benefit. There's only a problem when the benefit is lower than the cost of the energy, and that only happens when the energy's price doesn't reflect its cost.

But bitcoin could be sustained on a fraction of that much energy. There is no benefit to mining beyond the point that prevents 51% attacks. After that it's all tragedy of the commons waste.

That isn't how a blockchain works. The decentralization comes with penalties, such as if 51% of the participants say something is true it is. You have to provide some way to insure that doesn't happen due to a malicious attack. Proof of work systems do that buy making the cost and resources to achieve 51% astronomical.

I don't understand what you think I'm getting wrong here. All of what you say seems completely consistent with the point I'm making.

Yes we know. We are questioning whether the orders of magnitude increase in energy usage is worth the decentralization.

Unless there is a world government that dictates "no" I don't see why arguing with the market that clearly is saying "yes" is worthwhile. Even if externalities like "damage to the planet" were properly priced in by each country and energy source chain that still wouldn't guarantee the energy usage would go down, it just might be slightly less profitable. (From my standpoint there's currently no such thing as energy waste through human activity since the vast, vast majority of local energy in our solar system is not being used for any purpose, "wasteful" purpose or no.)

The whole idea is that if the price of electricity reflects its full cost, including externalities, then you'll no longer waste it. If people keep mining bitcoins in that environment, then it's because bitcoin mining actually provides more value than the cost of the energy it consumes.

If the fact that we're eight light-minutes away from a zero-maintenance 400YW fusion reactor makes it hard for you to consider human energy usage to be waste, just think of it in terms of CO2 pointlessly added to the atmosphere instead.

It doesn't help to think of CO2 pointlessly added either since I don't think mining is pointless. Now if they kept their existing setup but tweaked it so they still do all the computation but never submit any successful blocks, or just kept buying hardware and setting it on fire, then I guess I'd agree that seems pointless and wasteful (even if inconsequential compared to the giant ball of fire that's currently severely underutilized), but only because it's entirely and obviously unprofitable. Since "waste" by other metrics is so subjective I try to only consider the best proxy I can think of, which is cost and benefit figures. I don't put efficient LEDs in my house because I buy into the environmentalist argument of trying to live while using less energy personally, or live while "wasting" as little as I can, but because it saves me a significant amount of money. Similarly I don't stress out about "waste" if I go to sleep but forgot to turn some lights off because although I still have to pay for it, it's not very much. Even when the cost is high, like buying a lemon car, I don't feel bad about waste having to get rid of it and get a different one but about the extra cost.

The only way to affect miners is probably with economic consequences, rather than environmentalist philosophy. Maybe less CO2 would be added if the mining costs were higher, maybe not, and in any case I still have no idea how to price in the rise in CO2 due to a ton of SHA hardware cranking away, independent of whether it's powered by solar panels or water or coal. You'd have to have some sort of sensor on each unit (and the units involved in its manufacture -- put one over my mouth while you're at it) to capture its CO2 output at each moment of time (since it might not always be on due to repairs or whatever), and then you'd have to price the cost of each bit of output somehow, and who should pay it at what times.

To price the CO2 emissions of miners, you don't try to track the miners themselves. You tax CO2 emissions at the source (i.e. the coal-fired power plant the miners are plugged in to), and then the miners pay for it through higher electricity costs. If they use solar or wind instead then their electricity prices wouldn't reflect that tax. If they use a mix, the price will partially reflect that tax.

If you can price in the consequences at their origin then the rest just works itself out, whether it's LEDs or cars or bitcoin mining.

The data center doesn't consume the max output of each diesel, that would be an idiotic design.

Also, Visa handles orders of magnitude more transactions per second so the energy difference is staggering. Bitcoin users are like regular people driving around in coal fired steam locomotives.

Right, but this is a Fermi problem. We have no idea what the true output of each generator is, or that there are only two per centre. All we know is anecdotal evidence from reporters who visited the sites and whatever VISA will publicly admit. As an order of magnitude estimate, it's good enough.

And yes - it does highlight the problem with bitcoin well: 3 orders of magnitude less electricity to process 3 orders of magnitude more transactions.

Bitcoin mining is wasteful compared to other mechanisms (e.g. Stellar SCP) that can secure a blockchain without mining.

I couldn't find anything but I wonder how global energy consumption changes on track with the predicted bitcoin processing growth? are we talking 1% of global power consumption in 10 years or 10%?

It's possible to estimate an upper bound based on the dollar value of the BTC miners receive as revenue from inflation (subsidy) and transaction fees. The inflation rate in Bitcoin drops over time; transaction fees will likely increase, although not in proportion to the total economic value of transactions (and other uses) secured by Bitcoin (bitcoin is not "backed by energy", it is secured by energy expenditure).

Currently miner revenue is less than $7 million/day. It's a bit tricky to convert that into, say, CO2 emitted as miners tend to use the cheapest possible sources of energy, which are stranded renewable resources like hydroelectricity dams built in places distant from consumers (apparently China among others have built a lot of uneconomic dams for various political reasons).

The PoW security model doesn't require the amount of energy destroyed to increase indefinitely as a percentage of total value secured. Rather, the amount of energy destroyed by PoW simply has to be more than the highest funded attacker can afford. That's easy to do, as attackers who would attack via PoW are few and far between - as your attacker gets more powerful (governments) there are much cheaper attacks that can be done than naively attacking PoW directly.

In any case, I wouldn't be at all surprised if the amount of CO2 emitted by Bitcoin mining operations is dwarfed by the amount of CO2 emitted by plane travel associated with people attending Bitcoin conferences...

> Rather, the amount of energy destroyed by PoW simply has to be more than the highest funded attacker can afford.

Miners don't measure or care for how much an attacker can afford. They will simply continue adding mining hardware and increase electricity usage as long as they can turn a profit, which depends mostly on:

* the marginal cost of electricity

* the amortization cost of the hardware

* the price of BTC

* the difficulty

If you calculate power usage for the global hash rate as if it were produced by, on average, 2-generation old mining hardware @0.51 W/Gh, you end up with 0.017% of average global electricity production (3x more than last year).

If you extrapolate an unconstrained tripling of production every year, you will (to my surprise) arrive at your 10% estimate.

However, if we assume the last year's additional capacity was on modern hardware (5x more efficient), miners are presently using 0.0005% of global production, and after 10 years of unconstrained annual tripling would only account for 0.295%

Further advancements in efficiency or a slowing of the growth rate from 300% annually seem like likely diminishing factors.

In the end it doesn't matter how efficient the mining ends up being. All that matters is the price of electricity. Assuming efficiency gains level out and thus the cycle of buying new hardware to keep up approaches infinity length, the only major cost left is electricity use.

So in that case the energy used is dependent on (transaction fees * bitcoin price) / (energy price)

So to get a % of all energy usage, you need to know the future energy usage, the future energy price, the future bitcoin price and the future amount that will be spend on transaction fees.

I'm sure the future energy usage and price have good forecasts available. Then you're left with the bitcoin price and fees (both which are dependent on popularity (which in turn is dependent on price and fees)), but good luck forecasting that.

Ironically the best way to get bitcoin to switch away from proof of work would be to silently acquire 51% share of the mining pool and start voting for proof of stake instead of proof of work.

That's not exactly how it works as there's no voting per se. That would basically lead to a fork.

Mining Bitcoin is completely analogous to mining for gold. Was all of the effort and energy put into gold mining "wasted"? Perhaps, but the economic liquidity it provides has value in and of itself. I'd argue it's quite high, given that money supplies are only going to tighten in the future.

The article said it's largely run on hydroelectric power, which would seem to mean it's not hurting the environment that much.

China isn't trying to clean the environment.

The Paris Climate Agreement lets China get away with polluting more since they're considered a developing country. Not only that, they get paid by the first world countries who signed the agreement in the hopes that they will be developed faster. You have to be an absolute idiot to believe that a country with such an awful track of human rights violations would sign it out of the goodness of their heart.

Do you not think that mining metal, the minting process etc... is not waste?

Who said anything about metal? Physical cash is dying.

It is still used and being produced. Cash will not being dying anytime soon. People are using Bitcoin for many reasons, one of them being privacy, cash provides privacy.

It is a myth that Bitcoin provides privacy. Bitcoin is pseudonymous. With external knowledge, it is possible to detemine the identity of users. And you have to give your real-world information if you want to convert Bitcoins to $$.

There are a few Bitcoin "ATM"'s out there that dispense cash for BTC or let you buy BTC for cash. The latest one I read about took your picture and required an ID, but some out there probably don't, and a motivated individual could steal an ID and just not show the camera their face.

I really think it's amazing that we have this authoritarian China that is hellbent on destroying the environment in the world, destroying manufacturing and jobs and society in other democratic countries, and guess what, we keep buying crap that don't last long from China.

But it's work to prevent armored cars (often diesel) transporting cash. Or potentially large portions of banking and minting systems which as a whole expends plenty of energy. You can't just look at the cost without weighing the benefit.

You could say you don't think that is likely to happen, but spending energy to attempt to save more energy down the road isn't a clear mistake– it would just be a failed ambitious attempt to reduce waste.

It cost money to run a finance system.

What you are arguing, effectively, is that if you care about stopping global warming, then lets cut the electricity to every single bank in the world.

Some finance systems cause global warming because of the air conditioning systems running in their buildings. And others, like bitcoin, cause it via mining.

How does your categorical criticism not also apply to things like barbie-doll manufacturing facilities?

Also, it does not take into account the potential for value being removed from dollars (denominator and subsidizing currency of big oil, big war, big agra, etc.) and being given irrevocably to the proletariat who are actually concerned about climate change.

> irrevocably to the proletariat

Yes, a currency that is so hard to understand that even seasoned technocrats can't figure it out is perfect for the proletariat.

Cryptocurrencies are yet another example of wealth transfer from educated elite to another educated elite. There is nothing "democratic" about it.

It requires a tech expertise which 95% of the planet doesn't have to even accept/make a payment. Forget about "mining" it - 99.5% of the world's population doesn't have the knowledge or the resources for it.

If you can't figure it out, you're not really a seasoned technocrat, are you?

Sure, it requires some tech expertise right now. But so do ALL emerging technologies. Bitcoin will probably not be the dominant cryptocurrency forever. I foresee whichever one is able to be quickly and secured transferred and stored on smartphones changing the global economy. Think about how many people in developing countries have phones and need a way to securely transact and store currency.

Computers at some point was technology that 95% of the planet doesn't understand.

Most people still don't understand computers, but they're easy to use and relatively predictable, although security is still a big problem.

The thing about currency is that if people lose their money they might suffer and die. So it should be easy enough to understand that they can't easily be ripped off, and transparent enough that nobody can accumulate system-changing quantities without that being pretty obvious to everyone else. I've been interested in Bitcoin for years but understanding the Bitcoin ecosystem seems like a full-time job. You can't have an economy entirely based on financialization and trading, and if the people who specialize in handling the currency are sure to end up with the largest amounts of it then it's hardly better than what it's replacing.

I'm not saying there should be no rewards for specializing in currency work, but I'd like to see some sort of self-limiting mechanism. I'm reminded of a humorous Stanislaw Lem story in which a wicked king is tricked into replacing all the currency in the kingdom with uranium coins, so that his insatiable desire to accumulate as much as possible in the royal treasury eventually leads to criticality and his subsequent destruction.

Do you consider accessing Facebook as 'understanding computers'? If not, this number hasn't changed that much...

Do you consider having a wallet as 'understanding bitcoin'? If not, this number hasn't changed that much...

We can do this all day. You don't need to understand anything other than downloading a wallet, storing your coins there, and being able to send them anywhere instantly.

Except bitcoins are not instant, and aren't even guaranteed to go through.

The transactions are also non-refundable and pseudo anonymous, unlike the payment methods in your wallet. This is not a feature.

Making analogies with computers and smart phones and wallets doesn't mean Bitcoin is guaranteed to be successful or mass adopted.

being given irrevocably to the proletariat

Really? Where is this mass distribution taking place? I don't think most of the proletariat has ASICs and my attempts to mine bitcoin on a CPU some years ago were fruitless. I think I have half a BTC somewhere on my hard drive but but I don't have any Bitcoin software installed right now because having an active wallet seems to require handling massive blockchain files, and I really don't enough about security to be sure it isn't some kind of explosive payload.

I guess I could dig up an older computer, put a minimal operating system on it, and dedicate that to being my bitcoin wallet. Or my Dogecoin wallet. Or my Ethereum wallet. There's >100 cryptocurrencies now and I have no way of knowing which ones are subject to Tulip mania and which (if any) are being created in the public interest; they all seem to be their own raison d'etre (rather than having positive externalities that I'd want to invest in), they all seem to be speculative, they all seem to reward hoarding.

At least Ethereum seems aimed at offering utility beyond simply being currency, but so much of the hype is speculative that it's drowning out more proletarian considerations like real-world utility. I'm not trying to be snarky here, I'd love to hear which cryptocurrency you think is best for regular people and why.

By 'regular people' I mean people who don't have lots of capital to invest and don't want to restructure their lives around mining or something. I'd be particularly interested to hear of any currency that was designed to limit concentrations of value and instead prioritized transaction velocity.

the distribution today is evident in redditors being made into millionaires, and in venezuelans buying basic goods and health care.

the traditional government control on value is only starting to erode though, and who knows what the future looks like.

The idea that this will somehow wrest control from governments is a libertarian fever dream, and will be as long as governments have a monopoly on force. At best, governments will allow it to continue to operate in parallel, albeit subject to regulations. At least for citizens of developed nations with heavily regulated financial industries. I don't know if it will be more true for Venezuelans.

Wrong. Bitcoin is a black swan. For example, me and my friends use it to pay each other for various thing because it's just so easy/painless. There is NOTHING out there that is as easy and painless and cheap (pre fucked up BTC fees, we're using LTC for now to transfer between one another).

How is the government going to stop me from using Bitcoin? Buy bitcoins locally, store them in your wallet, memorize your passphrase, and delete your wallet. Now what? How are you going to know I have coins? All you can surmise is I bought 'something' for $X which could have easily been chips I bought with cash.

Used to work for a popular bitcoin wallet. LOTS of Vene users. Even talked to some personally who used ETH/BTC to pay for goods.

Just to speak of the "fiat" alternatives here:

Alipay and WeChat are free and convenient in China, relying on simple QR code scans to pay contacts. They are free. Other countries have similar things (LINE?)

PayPal is so close to being this too but it's just so clunky and relies on input of your friends' email addresses.

If I remember correctly Visa tried doing their own thing in the UK some time ago (Visa V.me) but it didn't really catch for whatever reason.

Apparently you haven't used Venmo, or any of the other tools for paying your friends using government backed money conveniently, for free.

I have. I can't stand that a third party 'owns' my money and the fact that it's not based off anything. It could be gone tomorrow. You can say the same for BTC, but that's only if someone steals my coins. You also failed to address the use case in my OP.

"Really? Where is this mass distribution taking place?"

It happens at the transaction level.

You now don't have to pay 30 dollars for a wire transfer. You can send money anywhere in the world, for 1-30cents, depending on your cryptocurrency of choice.

Plus whatever fee (often % based) you're going to pay if you need to convert from your cryptocurrency to another one or to your local currency, but that still may be less than other methods depending on where you live and what those methods are. Here in the US I can't even pay my landlord online without a $3 fee each time (or more if I want it to go through in less than a few days) because we bank at different banks and neither of us is willing to set up an account at the other's, so a check it is, but even that I have to pay the cost of the stamp.

Barbie doll manufacturers don't want to use electricity, and barbie doll manufacturing isn't zero sum. It's fundamentally different.

If someone learned how to make a barbie doll which used 1% of the current electricity usage, a couple extra dolls would get created, but electricity usage would drop a lot.

If someone learned how to make bitcoins with 1% of the current electricity usage, _miners would just mine 100x what they do now_. Fundamentally, the amount of electricity that will be wasted on bitcoin mining is driven by the price. So as it gets more efficient, it becomes worth it to waste more to match the price in the race for mined blocks.

It's horrible, and it's incredibly dishonest to compare it to any other kind of manufacturing.

I see what you mean, but in reality miners are constrained by mining hardware production speeds, and mining hardware pricing which charges a 10x more for a 10x more efficient machine. It is almost always a margin call whether buying a new machine will pay for itself, so the 100x elastic response you are describing seems unreasonable.

There are also 2 factors to control the growth of miners, reward halving and the end of this weird period of high fee transactions. After transaction fees drop down, a lot of older mining hardware will be unprofitable and come offline.

And how would that in any way refute the point that Bitcoin itself wastes energy?

It's a familiar pattern by now - someone criticizes Bitcoin, and the reply is that some other thing is bad too. You are essentially changing the subject, not addressing the criticism. Are you hoping I'll get distracted, or something? Surely you can't be hoping I'll suddenly start thinking Bitcoin is wonderfully energy-efficient, for example.

Reminding me that Ted Bundy was really bad, is not going to convince me that Jeffrey Dahmer was a great guy. (I almost went "full Godwin" here with you-know-who's next-door neighbor as the example, so, you're welcome.)

Are there any rhetoric experts out there who can properly categorize this "argumentation technique?" The best I can do is some combination of

https://en.wikipedia.org/wiki/Evasion_(ethics)

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

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

Anyway I call it "A full and glorious admission that you can't argue the point on the merits whatsoever."

How and why would Bitcoin remove value from real money?

Basically, because one is being printed faster over time and and the other is being printed slower over time.

Also boom/bust cycles and the increasing mainstream sentiment of crypto as a safe-haven asset.

I read an argument once that an inflationary currency encourages spending and discourages hoarding, and it follows that the reverse would be true for a deflationary store of value like Bitcoin.

And are you saying boom/bust cycles are a good thing? For pretty much any average user of money, they're not.

This is completely irrelevant to the article, where they use hydroelectric power to mine bitcoins. But even if they used nonrenewable soures, the negative impact on the environment shouldn't be faulted to bitcoin. It's the energy sector's fault for prioritizing harmful energy to renewable energy.

And even so, using energy to mine bitcoins isn't wasting the energy, it's turning the energy to real value. Just because it's not physical currency doesn't mean it isn't real - it has advantages (and disadvantages) to physical money. The electricity/algorithms cost create these features. And when countries ban cryptocurrencies, they would only do so on the guise of global warming, but more accurate and unspoken reasons would be that it gives too much power (anonymity) to the people.

Electricity is fungible. The clean hydropower would be used somewhere else if not wasted on BC.

Electricity is not perfectly fungible because of infrastructure costs and transmission losses.

The nature of mining as a competition for the cheapest electricity means that miners will naturally move towards areas where the electricity is not in demand.

They are using electricity to do computation. Even if, hypothetically, the electricity could not be transported elsewhere (and that's not realistic), if hypothetically manufacturing could not be set up there (and that's exactly what normally happens), the electricity could be used to do computation which is _actually useful_, which is why companies set up datacenters in the middle of nowhere with cheap electricity.

Minus the transportation costs.

The original piece in ChinaFile has more pictures from Liu and a similar amount of context.

http://www.chinafile.com/multimedia/photo-gallery/inside-wor...

I am unable to identify where the photos are in this page. Have they been taken down? I've tried multiple browsers.

They are still right there, front and center. Do you have a work filter that might be blocking them?

Interesting. I am using my home network, and after connecting with a VPN, I am able to see the images.

Yeah, that's certainly weird.

My work has websense and I've seen it before that it silently doesn't load some content on a page when that content was loaded from another domain (which was blocked) which can be confusing, which is why I suggested that. But upon further investigation it looks like this page is loading images from it's own domain.

good find, thanks for taking the time

Thanks!

Vice produced a video called "Life Inside a Chinese Bitcoin Mine" which complements these photos with a bit of an eerie audio/visual experience.

The noise of these machines is incredibly loud and annoying even in the edited video and the people running these things essentially live in squalor.

https://video.vice.com/en_us/video/life-inside-a-chinese-bit...

Non-Flash: https://www.youtube.com/watch?v=K8kua5B5K3I

This is the first article I've read that really made me feel that we live in a dystopia, instead of it being something on the horizon.

These photos are straight out of a cyberpunk comic.

You should see how most of the rest of China lives

I think it's the combination of high-end tech with third-world everything else that gives off that Blade Runner vide.

Such an incredible waste of power and hardware, for a payment system that does something like 3 transactions per second globally.

Off the top of my head idea:

What about a system that uses a bit of real-world, legal responsibility to generate consensus. Something a bit like ICANN and the domain name system.

Take 10 or so trusted people or organistions worldwide, get them to each run a node, each of which coordinate via the Raft protocol or whatever. A distributed ledger is agreed on by each of those 10 nodes.

This should use far less power and be able to handle more transactions per sec than bitcoin.

Or we could all switch to Ethereum, which will have Proof of Stake (low electricity costs) instead of Proof of Work (insanely high electricity costs) mining soon.

High electricity costs needed to rewrite the blockchain is a feature of proof-of-work, not a bug.

We could all switch to proof-of-stake, but then anyone would be able to create a valid-looking blockchain in a few minutes, and the only way to get the right chain would be to connect to nodes you know or trust, at which point a blockchain becomes meaningless in the first place.

The challenge is about coming to agreement. Reaching agreement on information which determines how much money people own is hard. Proof-of-work reaches consensus by having a vote -- on which transaction history is correct -- constitute something that cannot be forged: energy.

POS networks already exist. No need to way 3+ years while the Ethereum team reinvents the wheel.

See PeerCoin ( https://peercoin.net/ ) which has been around for quite some time and has a POS/POW combination.

Bitcoin is not a payment system. It's essentially a digital commodity with built-in ownership transfer. Bitcoin payments can be cleared off-chain, like VISA payments can be cleared without using SWIFT.

Most importantly, though, it's decentralized. It's simple to implement a fast, centralized currency and payment system. But Bitcoin's main attraction is that it values decentralization greater than efficiency and throughput, thus making it, primarily, a store of value, as opposed to a means of payment.

https://bitcoin.org/bitcoin.pdf makes it clear that Bitcoin is definitely a payment system, a distributed one. But it's not only a payment system, and using it like one right now is pretty dumb because of the absurdly high transaction fees, but high fees are entirely a temporary artifact as the system scales. (Edit: removed hostility.)

Bitcoin will never work as a consumer-to-merchant payment system. Even with 1GB blocks (1000 times today), it would max out at ~7,000 TPS, where VISA does 200,000 during peak. We need something off-chain to be able to scale to +100k TPS (unless you consider 100 GB blocks realistic).

I believe you're off by 2 orders of magnitude for Visa's seen-peak and about 1 order off for their network's designed max peak. Unless you've got sources more recent than https://en.bitcoin.it/wiki/Scalability uses; maybe you're confusing the number with the network handling all transactions world-wide which Visa doesn't do?

100 GB blocks is politically unrealistic in the foreseeable future but technologically not really, especially because network and storage continue to get better and cheaper even if CPUs are stalling. Nodes don't need to store the full chain, and 10 years of that size of block would only be about 50 PB. Gigabit networks are only becoming more and more common, a 1 Gb connection will let you get 75 GB over 10 minutes (and using the math from the wiki you only need 0.76 Gbps to handle 200k tps). Also not every node needs to see every transaction, the effective bandwidth can be much lower, just like the effective storage.

And as the wiki points out, PayPal has a much lower TPS than Visa or others but is still widely used as a direct consumer-to-merchant or friend-to-friend payment system. In addition, the better straw argument against Bitcoin as a payment system has always been "who is going to wait 10 minutes on average, sometimes much longer, especially if you want the gold-standard 6 confirms, to buy a stupid pack of gum?"

> Bitcoin payments can be cleared off-chain, like VISA payments can be cleared without using SWIFT.

The whole point of SEPA transfers was to move all transfers to be cleared directly via the main ledger, in a matter of seconds, across different countries and currencies.

If Bitcoin is inferior to an existing payment system used by 500 million people, then it’s worthless.

Bitcoin is global and decentralized while SEPA is EU-only and centralized. The two are not in competition; neither is an alternative to the other.

> If Bitcoin is inferior to an existing payment system used by 500 million people, then it’s worthless.

If Bitcoin is strictly inferior to an existing payment system used by 500 million, then it's worthless.

IMO the major value proposition of bitcoin is that no one can exert influence to invalidate tx or blacklist certain users. Payment systems that have this property will always have some utility for organizations that nation state actors are trying to stop, e.g. ransomware, any darknet market provider, political dissidents, freedom fighters/terrorists, people trying to participate in a market in a controlled economy etc.

Efficiency was and never will be the primary goal of bitcoin. It is meant to provide you with permissionless ownership. In a permissionless system if what you own_ is worth X dollars, than it will take some amount, say Y, dollars to secure this ownership. In the case of bitcoin we do not have a way to arbitrarily stake y dollars in the system (see proposed proof of stake in ethereum). Instead you can indirectly assume that Y is supplied by the miner investment in hardware and electricity. By stressing known hardware limitations, we can force miners to act more and more and more competitively as more total value is held on the network.

So the increasing cost has nothing to do with use (aka transactions). It is a consequence of providing security to a growing amount of value on the network. There is simply no better way to do this currently assuming you want a permissionless architecture. It is step 1 towards creating an open finance system. Other steps like proof of stake will eventually solve this problem.

> Such an incredible waste of power and hardware, for a payment system that does something like 3 transactions per second globally.

Bitcoin does a few transactions per second on-chain, however those few transactions per second support many more transactions per second on systems built on top of Bitcoin in a similar way to how the tx-per-second Canada's inter-bank transfer system does is quite low, but those settlement transactions support orders of magnitude more transactions per second on systems built on top of that settlement system.

An example of exactly what you proposed is Blockstream's Liquid sidechain technology, which maintains a high-speed ledger secured via multi-party signing between various exchanges to allow high speed, highly efficient, off-chain transactions backed by on-chain Bitcoins.

So, we should use a system that’s inferior to actual simple wire transfers, which are less than 20 seconds, free transactions, directly via the ledger?

I’m sorry, but I don’t see the value in Bitcoin if it is worse than SEPA transfers.

It is not worse.... That the point. There are significant advantages over a wire transfer.

With bitcoin you can send money to anywhere in the world for 20cents. Thats cheaper than a 30$ wire transfer.

Also, the censorship resistant part of bitcoin is really important, and is being used by the Venezuelans right now.

And there are technologies being built on top of it that will make it even better.

This is like 1994 internet right now. Even though it works, and has significant benefits, there are still a whole lot more stuff to come that is being worked on.

That's the thing, I can send money with bitcoin for 20ct, or with a SEPA transfer for free. 0ct. And it's faster.

Looks like current fees are closer to 1 USD than 0.2 USD, but that's not the point, the fees are not a core feature of bitcoin. It's the censorship/manipulation bit that's the core feature IMO.

There's no need to trust neither the processor or the issuer, only the protocol & distribution of hashing power.

Bitcoin feels like one of the biggest scams around. It feels like it's propped entirely by speculators at this point.

> Take 10 or so trusted people or organistions worldwide, get them to each run a node, each of which coordinate via the Raft protocol or whatever. A distributed ledger is agreed on by each of those 10 nodes.

then I'd argue it is still centralized - you have this 10 nodes acting as the traditional bank.

check out bitshares, steemit and eventually EOS. Their DPoS model uses exactly that -- users vote on the witnesses that will produce blocks. Every witness gets a shot at producing the next block. If the witness fails to produce a block, they will be voted out by the users.

Scales to 100k tx/sec, supposedly.

I wonder if it would be possible for 90% of users to come to agreement on assigning the balance of the wealthiest account to themselves. Without proof-of-work, there'd be no way to prove this has happened, since it would be easy to create a new blockchain copy which just assigns the coins from the wealthiest account to all other accounts.

Don't you lose decentralization then?

Nothing is as decentralized as proof-of-work when it comes to cryptocurrencies. It ensures you can know what the right chain is as long as only a single node you connect to tells the truth.

Where were you 8 years ago?

I understand the need for a distributed ledger, but it still seems absurd to me. All of this power is just getting put into a slow append-only database.

It's definitely a huge waste and a major impetus for moving to a Proof of Stake, Useful Proof of Work, or other model for cryptocurrency security.

It's a bit of a tragedy of the commons, not sure who will be the third party to impose a cost onto cryptocurrencies that consume excess energy. So far bitcoin has maintained it's lead as the largest cryptocurrency essentially entirely due to network effect. For most major complaints against bitcoin there is an example of a crypto currency that doesn't have those 'failings'.

E.g.

Deflationary good: siacoin

Long tx times: Ethereum

Proof of work consumes needless energy: Peercoin

ASICs shift the power balance: Primecoin

Not anonymous enough: Monero

A singular append only blockchain is not enough: Byteball (it's simply a DAG of transactions)

The problem is that there is no centralized player that is currently exerting a cost, above mining costs, on the individuals to move them to a crypto currency that doesn't have whatever negative property is being rallyed around at the moment.

0.0000001% of the power runs a slow append only database. 99.99999999% runs a security system for it.

That is just waiting for some type of disruption. Not that I can predict what that disruption would be but things this inefficient can not be an optimal end state for this type of technology.

It's true, work on other systems like proof-of-stake instead of proof-of-work are in active development. It might be too hard to backport to bitcoin specifically, but other cryptocurrencies don't have to be this bad in the future.

Bitcoin was the first practical secure distributed ledger and cryptocurrency. The first of something is never the most efficient or capable. I'm sure there are better ways.

If you read about this in a science fiction book 20 years ago you'd think the author didn't know anything about computers or programming.

I was thinking something along those lines but struggling to put it into words. This sounds right.

Just, picking my jaw back up off the floor. The conditions there are shockingly unsafe (pun intended) -- all that standing water on the floor? And it's a firetrap -- all that trash lying around, no fire supression. In fact the opposite -- those box fans that will very effectively oxygenate the flames. Disaster there is almost certain.

Funny. I thought it was actually pretty organised and neat considering this is rural China and an operation aimed at doing things @ least cost.

In the quest for cheap power & better density, you'd think they would run direct DC to everything, instead of a bunch of ATX power supplies

Thick copper wires for DC probably cost more than energy converted to heat on that power supplies.

That's up for debate. Facebook and Google use DC in their datacenters to save costs. Azure has a requirement for optional DC power from their vendors and might be using it actively. I think AWS has looked into using DC power as well but isn't using it afaik.

Do they use higher voltage DC and buck it down, or is it all 12V, 5V, etc. ?

Good god this is so cyberpunk

Crazier than neuromancer

Truth is stranger than fiction

Does anybody know why the bitcoin mine has customers? If you can mine bitcoin, why do you need to sell it to anybody other than on the open market? I could see needing investment to build the infrastructure, but those would be investors not customers.

Is there value in buying direct from the miner or selling direct to a customer and not on an exchange? Maybe avoiding transaction fees?

Exchange transaction fees might be one thing (cut out the middle man) but if you buy directly from a miner it won't influence the price as it would if you buy on a public exchange.

If you buy a big amount this is significant.

Such OTC (over the counter) transactions are also anonymous compared to exchange buys.

Can you say more about how buying at an exchange influeces the price and why that's bad?

The price we usually see for any traded asset is a weighted average of public exchanges. It's a somewhat artificial number insofar as it only tells you that somebody in the past was willing to pay that much for the asset. There's no guarantee that you'll find a seller in the future for that price.

Consider if you wanted to buy a large amount of Bitcoins, like 10000. If you look at exchanges, you will see that the sell and buy offers are public and you can only buy as much Bitcoin right now as are offered on the orderbook.

If there are more than 10000 BTC for selling, you can just drop your money on it. But those Bitcoins will be with different price tags. The cheapest offer will sell 1000 for $2400, the next 1000 for $2450, etc. If you blindly buy, your average buy price might be much more than the average price was!

Now that you cleared a portion of the orderbook, you have cause a price imbalance. On this exchange A, the cheapest sell offer might be now $3000 but on other exchanges B, it is still around $2400. Now bots and people will scramble to take advantage of that price imbalance. You've create a volatility wave that affects all public exchanges. And you never know how people will react to that. Will the price keep increasing as people think "the price is skyrocketing" or will people start selling massively because "oh, better cash out before the price drops again".

So, you bought an asset which you payed more for than the market thought it was worth before you bought it and you are uncertain what price it will settle on afterwards.

Also, you might want to buy more in the future, so you don't want to let people know that you are interested in acquiring it. If you buy publicly, people will notice.

If you do an OTC, that is much more hidden and you don't create volatility, especially if your OTC agents get them from miners directly and are not accumulating them from exchanges in small amounts themselves.

Thank you! Great reply.

Why not move to proof of stake? Proof of Work mining used in bitcon to secure the network as originally proposed in Nakomoto's paper has huge energy requirements. The result is large energy hungry mines as shown by the article. In addition, there is a centralization of governance by the miners.

One such example of a long running proof of stake is Peercoin. Peercoin is a legit energy-efficient contender to Bitcoin. It's one of the first publicly known implementations of proof of stake has been around for close to 5 years now.

Peercoin is energy-efficient and runs on raspberry pis. It's been running successfully for years now.

https://peercoin.net/

There are also academic proof of stake chains which virtually eliminate the chance of forks, such as algorand.

https://arxiv.org/abs/1607.01341

> Why not move to proof of stake?

Because it doesn't make sense. If a chain, on which a token exists, needs securing, you can't use that chain's token to secure itself.

> Mining has huge energy requirements as shown by the article.

Mining has no requirements at all. People just follow the best chain, and miners can sell the bitcoins they earn by extending the best chain. It's entirely voluntary.

> Why not move to proof of stake?

>> Because it doesn't make sense. If a chain, on which a token exists, needs securing, you can't use that chain's token to secure itself.

I mean I'm not entirely sure what this argument is based on? In addition to existing proof of stake systems which have been running for years now, there exist several papers with rigorous formal proofs. Including Turing Award winner Silvio Micali's algorand.

Refute his paper. You will became very famous overnight.

https://arxiv.org/abs/1607.01341

> Mining has huge energy requirements as shown by the article.

>> Mining has no requirements at all. People just follow the best chain, and miners can sell the bitcoins they earn by extending the best chain. It's entirely voluntary.

I assume this is trolling or extreme nitpicking. Securing the network in proof of work requires that the majority of computing power is held by honest miners.

> I mean I'm not entirely sure what this argument is based on?

Proof-of-stake (PoS) doesn't solve the fundamental problem that we need to reach consensus on one, single history of transactions. All PoS does is enable us to say whether a new block is valid, but that's not important. The important part is that everyone else in the network agree that this is the new block (and not some other valid block). Valid blockchains are easy to create, the valid blockchain isn't.

Defining a valid block as a function of some preceding blocks is simple. The challenge is to incentivize a decentralized network to come to complete agreement on one, single version of all people's balances. Collusion is a huge risk factor here (rewrite blockchain to assign the balance of the five most wealthiest accounts to everyone else), and PoS does nothing to mitigate that.

In the worst case scenario -- where an exploit spreads through a security hole in the reference implementation, and replaces the blockchain with its own copy, silently assigning the balance of an old, rarely used account to the attacker -- PoS falls on the floor, and proves unsuitable as a base for digital money.

>> In the worst case scenario -- where an exploit spreads through a security hole in the reference implementation, and replaces the blockchain with its own copy, silently assigning the balance of an old, rarely used account to the attacker -- PoS falls on the floor, and proves unsuitable as a base for digital money.

Replace exploit with malicious miner and it's not clear why proof of work is better than proof of stake.

An honest mining majority is a hard requirement of Bitcoin. But proof-of-stake doesn't solve the problem of collusion -- it's even more susceptible to it since it's so easy to recreate valid-looking chains. With Bitcoin, two or more competing chains will at least be visible (two distinct chains with a lot of work), whereas they're everywhere in PoS and the main chain is qualitatively no different from any other valid chain (of which we can create thousands quickly).

>An honest mining majority is a hard requirement of Bitcoin. But proof-of-stake doesn't solve the problem of collusion -- it's even more susceptible to it since it's so easy to recreate valid-looking chains. With Bitcoin, two or more competing chains will at least be visible (two distinct chains with a lot of work), whereas they're everywhere in PoS and the main chain is qualitatively no different from any other valid chain (of which we can create thousands quickly).

I'm not sure what math you are using to base this assumption on. A formal argument by algorand is that the probability of a fork would be the age of the universe. It is NOT straightforward or cheap to recreate valid looking chains and generate the next block creating a fork.

https://arxiv.org/pdf/1607.01341.pdf

A fork is not just created by the next block, but can be created by rewriting any number of blocks (including from the very beginning, which must be valid or it wouldn't be possible to begin a chain). Starting a new chain from scratch -- with all valid blocks -- is easy with PoS, and it would look no different from the right, valid chain. Again, the challenge is to come to agreement on the correct history, not just any valid chain. The inherent problem is that any valid chain of blocks can be produced quickly (just start from scratch), and there's no way to know you have the right chain except through majority vote, since there's nothing qualitatively separarating all the valid chains.

PoS seems simple if we assume that everyone will gladly stay on the same old chain, and only accept blocks that extend this, rather than attempt to get a new chain accepted on which they have 10x as many coins. But why would they? Why not switch to an entirely new chain, with the coins of the wealthiest 1% assigned to the remaining addresses? Surely, the person who owned those coins will protest, but no one will hear 1% in a network where majority vote is the only solution to consensus (and the 1% would have no actual proof that the chain, on which they own 1% of coins, is the right one).

At this point this is just trolling.

The proof of stake you are describing is of your own invention and doesn't address how systems like Peercoin or algorand actually function.

Starting from scratch with a new genesis block? Sure go ahead.

Rewriting any number of blocks is not straightforward due to the use of digital signatures and the cryptographic guarantees that go along with generating and choosing who gets to sign and generate the next block.

Russia: http://englishrussia.com/2017/06/26/russian-biggest-bitcoin-...

I've heard that some actually mine at a loss, since the real purpose is to get money out of China and they can charge fees for that service that compensate.

For me, the really sad fact here is that they picked that place to run their bitcoin mine because the electricity there is cheap, but what that poor place can benefit from those mines after providing them with such more affordable energy? I just saw pollution and greedy there.

If that mine wasn't there, the energy would just be shipped off onto the grid, never to be seen again.

That town does look depressing, but that's what like 90% of little western Chinese towns look like.

if that mine wasn't there -

people, actually poor people, living next to those mines won't hear those noise 24/7. they don't need to witness the greedy saw in those photos/videos. the depressing town is still going to be depressing, but at least with less electronics junks.

the issue is pretty simple here - what is the benefit for this little depressing town or the society? nothing. it is just a cancer.

The hydro plant, which predates the bitcoin mining operation, creates far more noise than some electronics churning out hashes.

Why don't you let the residents decide what they want? Maybe they want access to the internet and cell phones, which the mining operation helped provide.

what a joke. the "hydro plant" generates electricity, noisy or not, at least there is some benefit.

Internet and cell phone network is available in like 99% of Chinese towns. Do you have any evidence showing it was the mining of bitcoin that brought those thing to such small towns? If you don't, how about stop using your imagination as proof?

Is it safe to assume that this isn't quite what Satoshi had in mind back in 2009 ?

It seems that no, he expected exactly this:

At first, most users would run network nodes, but as the network grows beyond a certain point, it would be left more and more to specialists with server farms of specialized hardware.

https://webonanza.com/2015/10/02/did-satoshi-predict-pooled-...

So... We're centralizing the production of the currency that's supposed to decentralize everything?

There's a meaningful difference between centralized, decentralized, and distributed. I think you may be confusing the latter two.

Decentralized systems can still form groups and have various hubs. In such a case, the security of the network depends on how large these hubs get, not whether they exist at all.

Distributed systems ensure that most nodes are similarly sized and have similar connectivity to the network. In the case of cryptocurrency, many people imagine this as every user of the currency having a roughly equal stake in maintaining and securing the network.

Distributed systems may be ideal for some applications, but they are (likely and thus-far) not required for Bitcoin.

First, the degree of centralization is still relatively low. Individual miners remain small fractions of the network. Some mining pools have gotten quite large in the past, but miners can join and leave pools as they see fit. The risk they present is fairly limited as their network power is not controlled by a single entity.

There's also the issue of geographical centralization. This may be largely the result of ASIC development. Right now, mining profitability relies on rapidly adopting new, more efficient miners as quickly as possible. It strongly favors cheap power and short timelines. As the ASIC market matures and runs into the same hard limits that CPU manufacturing is faced with, the economics change such that more capital investment can make sense. In these cases, mining may be profitable in more geographically diverse areas.

Iceland has very cheap geothermal electricity and a very cold climate to cool the servers. How does China compete with Iceland?

I think the rapid development of ASICs is a large part of the issue. The capital lifecycle of Bitcoin mining is currently on the order of months to perhaps a couple years. That, combined with the volatility of Bitcoin price, makes the Chinese approach ideal.

As the ASIC market matures, and perhaps as the price becomes more stable, I think we'll see mining begin to occur in more areas as planning for 2+ years begins to make sense. Iceland would be one attractive area, as would some desert areas with low costs and solar energy.

Certainly difficult to predict, though.

The Tibetan plateau is huge - 25x bigger than Iceland, and at an average elevation of 14000 feet. The amount of water that comes out of the plateau and needs to make it to sea level is mind boggling. There is a lot of cheap energy to be had there.

China is kind of corrupt and under regulated. I imagine the capital to build the miners comes from some banks or nationalized corporations that are not going to get it all back while the bitcoins disappear into the organisers pockets.

Extremely cheap hydro power at around 0.01-0.02 USD/KWh. China has more hydro power than customers in a lot of areas.

Labor and regulatory costs are a huge factor as well.

Subsidized electricity? Just one data point, but my electricity bill last month in Shanghai was ~$15.

Closeness to the chip manufacturers?

Wow. So damp. Someone's going to get electrocuted. Where's Chinese OSHA?

Chabuduo http://boingboing.net/2016/10/12/chabuduo-chinas-culture-of....

Its not just the appalling dangers of water and electricity in close proximity but water on the floor indicates high humidity along with the heat causing premature failures in all the equipment. I guess those costs aren't a factor, however, when you're paying your workers pennies a day.

I'm 100% sure their workers aren't getting paid pennies per day. Why say something that's just patently untrue?

Premature failure isn't much of an issue when miners become obsolete in months. Very few miners are profitable to run after just 2 years, usually less.

What about "Bitcoin-Mining-Water-Power Plant" :) Would be cool!

Thanks China for wasting resources and being extremely pollutive mining coins for tiny profits. Anybody who complains about the Paris Agreement needs to be informed and look at the real climate offender China.

It's not china's government, it's the Governments of the world that put a premium on anonymous value. Most of the demand for that comes from the US. If you feel strongly about it, petition the USG to allow alphabay to take credit cards.

So China cannot be held responsible because their citizens are poor and economically disadvantaged? Must be the US's fault. Big bad USA. Typical.

That page has some sort of funky scrolljacking, -1 for usability. Yet another reason to disable JavaScript.

These images are even more proof that a bitcoin is worthless and the market is manipulated.

The life of a bitcoin miner worker doesn't look that much different from the coal miners ones.. at least in this case in particular..

OT/Rant: In my browser, all pictures except the hero image are extremely blurred to the point that I can't even tell what's in the image. It could be because Privacy Badger is blocking some script or stuff like that but honestly I don't have the energy to find out why right now (it's 00:45 here in NSW), I believe users should be able to see a picture without going through executing some JS, accessibility is important. This is almost as annoying as those scripts that mess with your scrolling. It is also completely unnecessary since every modern mainstream browser since probably 5 years ago implements some form of lazy loading.

It does look like they're copying Medium's fade-in code, but these fallbacks should let the page degrade gracefully. Two examples:

  <noscript>
    <img itemprop="image" src="https://qzprod.files.wordpress.com/2017/07/china_bitcoin_002-e1499795756538.jpg?quality=80&#038;strip=all&#038;w=320" alt="Inside the World of Chinese Bitcoin &#039;Mining&#039;" title="">
  </noscript>

  <img data-srcset="https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=320 320w,
  https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=640 640w,
  https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=940 940w,
  https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=1600 1600w,
  https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=3200 3200w"
  data-src="https://qzprod.files.wordpress.com/2017/07/china_bitcoin_010.jpg?quality=80&amp;strip=all&amp;w=1600"
  sizes="(max-width: 320px) 320px, (max-width: 1024px) 640px, 940px"
  class="progressive-image-large"
  alt="Inside the World of Chinese Bitcoin 'Mining'">

AFAIK that degrades gracefully only if you have JS disabled but there are many other reasons why the progressive loading script may not work as expected.

Reply on news.ycombinator.com