Bitcoin mining and energy consumption
blog.bitcoin.org.hk
blog.bitcoin.org.hk
Running a Bitcoin miner on commodity hardware now is pointless which seems to go against the spirit of Bitcoin when it started. Was this predicted at the inception of Bitcoin? It's interesting how economics impacts the security of the protocol like this.
Proof of stake is meant to consume less resources but then the power is then handed to people with the most money? Coming up with a way to have a decentralised currency where everyone involved gets a fair say without consuming too many resources is a super interesting problem.
The advantage of low electricity cost is of course something that cannot be eliminated without eliminating Proof of Work entirely.
Is it possible to create ASIC resistant algorithms? Litecoin was meant to be resistant by using scrypt but there's ASICs out for that now I believe.
Could you not create a protocol that constantly changes its hashing algorithm to combat ASICs?
> The advantage of low electricity cost is of course something that cannot be eliminated without eliminating Proof of Work entirely.
Any views on what the ideal system would be? I've heard of using a combination of proof of stake and proof of work. Proof of stake will mean people with more money get more power and proof of work means people with lower electricity costs get more power so neither seems ideal. I think ideally each person in the world would get an equal vote in how the protocol progresses but it's not obvious how you could enforce that.
Making the PoW require tons of memory to solve efficiently seems the best way to attain ASIC resistance. Monero's PoW requires only 8x more than scrypt's 128KB, but Ethereum requires over 1000x more than Monero's 2MB. None of these can be instantly verified though which is a desirable property of PoWs.
Modern asymmetric PoWs like Grin and aeternity's Cuckoo Cycle, and Zcash and Bitcoin Gold's Equihash combine large memory requirements (144MB-2.2GB) with instant verifiability.
It might be interesting to combines all of the popular hash functions in a sequence. You could probably mix in some non-cryptographic hash functions in the middle as long as you use secure functions at the beginning and end of the sequence. e.g. SHA256 => MD5 => MurmurHash => BLAKE-256
You could even write an algorithm that uses the previous block hash to determine the next order of the hash function sequence.
After posting that comment, I discovered a far better proof-of-work called the Cuckoo Cycle:
* Blog post: http://cryptorials.io/beyond-hashcash-proof-work-theres-mini... * Whitepaper: https://eprint.iacr.org/2014/059.pdf
It's really a genius solution. You have to fill up 2GB of RAM with random bits in order to find a specific cycle in an enormous graph. This means that even even a very slow CPU would quickly saturate the DRAM memory bandwidth, and it doesn't matter if you use a CPU, GPU, FPGA, or even an ASIC.
This means that a mobile phone with 2GB of RAM can mine just as fast as any server with 2GB of RAM. If you want to increase your mining capacity, you'll have to buy a lot of RAM (or a lot of cheap mobile phones.)
Since it is constrained by memory bandwidth, you spend most of the time just waiting for bits to arrive. This means that it uses much less electricity than other PoWs. The proof is also instantly verifiable.
There are a few cryptocurrencies that are using the Cuckoo Cycle proof-of-work such as aeternity [1], and Mimblewimble / grin [2].
Frankly, it will be interesting to see Monero's distribution. Because I was there when a glut of Cryptonight coins were announced. The selling point was CPU mining was back.
Quite a lot of people abused the AWS free credits to get a leg up and mine tons of coins.
But the most interesting was the guy who worked on a CUDA implementation to the Monero's mining algorithm ie a GPU miner. For a long, long time this mining software remained private. I believe it was only after the BTC crash to 300 and hence, Monero's to lower $1 levels that this got released publicly.
[1]: https://www.stellar.org/papers/stellar-consensus-protocol.pd...
I'm surprised nobody has caught on to the fact that the present fiat system is in essence a Proof-of-Stake one.
> Coming up with a way to have a decentralised currency where everyone involved gets a fair say without consuming too many resources is a super interesting problem.
Indeed. I almost wonder if it can be proven that thermodynamics is in fact the only possible way to verifiably make something take considerable time and energy, i.e. actually cost in resources that we have no way of faking, especially time.
That was what I was meaning really. If the people that have large amounts of money (whether that's in owning coins with proof of stake or owning hardware with proof of work/space) have a disproportionally large say in what happens it's not much of an improvement to the current systems we all live in.
This seems intuitively correct. So much so that I'm sure there's a paper that either confirms or denies that this is the case (perhaps in information theory). I also wonder how quantum computing affects this idea.
GRANTED: more severe opportunities for fraud, unfairness and bad acting are available to Wall Street
NOTE: I find bitcoin fascinating and I really want it to work, but it has seriously flaws that aren’t being seriously acknowledged by the people willing to buy in at $15,000+ per BTC. Not trying to be a naysayer to crypto currency
[0]: https://bitcoin.stackexchange.com/questions/658/what-can-an-...
Denial of service. Should the +51% choose, they can simply ignore certain transactions and those will never end up on the chain.
Our fiat private finance system does concentrate weatlh, but DOS attacks are only possible by subverting the legal regime.
It was not predicted in the original Bitcoin whitepaper. Yes, it discussed the theoretical feasibility of a 50+% attack to double-spend but it didn't explicitly predict a "consolidation" to specialized hardware miners that leaves out the miners at home using regular computers.
>It's interesting how economics impacts the security of the protocol like this.
Every decentralized protocol suffers from unintended centralization caused by economics. The same thing happened to other protocols like NNTP (Usenet), SMTP, HTTP+HTML, and Git.
The underlying issue is that technical protocols still have to be realized on real hardware like cpus + harddrives + network bandwidth and consume real costs like human labor. You can decentralize a protocol specification but you can't decentralize the amount of money different entities are willing to spend on that protocol. Each protocol whether NNTP/SMTP/Git/bitcoin does not come with a $1 million grant for homeowners to spend at Newegg to keep protocols decentralized.
That's why protocols consolidate towards big players in a power law distribution.
> Every decentralized protocol suffers from unintended centralization caused by economics.
> That's why protocols consolidate towards big players in a power law distribution.
Hmm, given the amount of foresight in the original whitepaper I'm surprised the current situation wasn't predicted especially if it's common with other protocols that are intended to be decentralized.
At inception? No. But Satoshi was made aware of this problem. Here's a long passage from Nathaniel Popper's Digital Gold:
Laszlo’s CPU had been winning, at most, one block of 50 Bitcoins each day, of the approximately 140 blocks that were released daily. Once Laszlo got his GPU card hooked in he began winning one or two blocks an hour, and occasionally more. On May 17 he won twenty-eight blocks; these wins gave him fourteen hundred new coins that day.
Satoshi knew someone would eventually spot this opportunity as Bitcoin became more successful and was not surprised when Laszlo e-mailed him about his project. But in responding to Laszlo, Satoshi was clearly torn. If one person was taking all the coins, there would be less of an incentive for new people to join in.
“I don’t mean to sound like a socialist,” Satoshi wrote back. “I don’t care if wealth is concentrated, but for now, we get more growth by giving that money to 100% of the people than giving it to 20%.”
As a result, Satoshi asked Laszlo to go easy with the “highpowered hashing,” the term coined to refer to the process of plugging an input into a hash function and seeing what it spit out.
But Satoshi also recognized that having more computing power on the network made the network stronger as long as the people with the power, like Laszlo, wanted to see Bitcoin succeed.
Not sure if I entirely agree with/understand this. Even if they do want Bitcoin to succeed, how different is it from the current capitalist system if 40% of the bitcoins are owned by 1% of the population?
Doesn't the (economic) power that Bitcoin takes away from governments end up in the hands of a select few? We are then back to square one.
1%? According to a Bloomberg article[1] it's 1000 people owning 40% of all bitcoins. That's about 0.000014286% of the world population. I doubt any of the are involved in mining either now either, more likely just early adopters.
[1]: https://www.bloomberg.com/news/articles/2017-12-08/the-bitco...
http://nakamotoinstitute.org/bitcoin/#selection-37.4-37.173
Most people trading BTC atm are not trading BTC, but trading ownership of BTC using a traditional database.
I think the obvious solution is to do useful work rather than throwing away work on crypto. Have miners solve real problems and give them credit for getting the right answer.
This is what Ethereum is supposed to do. I don't think it's a great solution. It's still extremely wasteful since every node runs every computation. Even something like TrueBit, which is designed to be scalable and efficient, is still far less efficient than normal cloud architectures. It still throws away the vast majority of the work in the interest of safety and incentives. These systems are good if you want trustworthy computation, but not if you want high-performance and power-efficient computation.
I'm trying to design a system where something like 95% of the work goes to useful computation, but I have nothing anywhere near shippable. It's a hard problem.
Keep in mind that the parent stated multiple conditions in the problem and your "obvious solution" only addresses the "too many resources" component.
Doing "useful work" is obviously better but still doesn't address the "fair say in decentralization".
>I'm trying to design a system where something like 95% of the work goes to useful computation, [...] It's a hard problem.
And to make it an even harder problem, also try to design a cryptocurrency system where Bill Gates' billions to buy supercomputer hardware has no advantage over a typical homeowner with a cheap computer. Maintaining _economic_ decentralization is very hard. I'm not aware of any decentralized protocol that has solved it. Heck, most whitepapers about decentralized protocols don't even explicitly discuss it.
Bitcoin is entirely owned by people with ASICs. The barriers to entry are enormous. It's totally pointless for me to mine Bitcoin on my laptop, or even on my Beowulf cluster. But if mining involves normal computations like you'd want to run on a normal cloud, then a laptop is capable of doing useful work and the barrier to entry is low.
Making normal computers useful is a huge improvement over Bitcoin and is the motivvation behind currencies like Monero.
No, the "low barrier to entry" because of laptop use instead of ASIC is orthogonal to volume of computation/diskspace/specialization. That does not solve the "spend more money on computing resources" that inevitably leads to centralization. Making useful computations that's resistant to ASICs is an improvement but does not prevent economic centralization. Therefore, Monero doesn't solve it either.
You can see real-world evidence of this where SMTP email protocol and Git protocol do not run on specialized ASICs and yet, they still got economic consolidation.
This is an aspirational wish but not reality at the moment. Nobody including BOINC or any other decentralized grid computing endeavor has proven that economically. For your cause & effect of "less centralization" to happen, homeowners cpu would have be cost competitive with cloud services. Selling "idle" time on home cpus isn't cost effective. Most realistic computation workloads also require large datasets (diskspace) in addition to cpu cycles. People don't pay only for the Amazon cpu clock cycles; it's also S3 data storage and network bandwidth.
Likewise, Filecoin's whitepaper for decentralized disk storage is also aspirational and economically unproven at this time.
>A system where anyone can mine productively will result in less centralization than we see with current cryptocurrencies.
Probably not. The "mine productively" in your solution has 2 components: (1) the useful computation and (2) the cryptocurrency computation.
For a thought experiment, let's assume you invent the 95% useful computation algorithm that you targeted. The 5% computation will still be have to be done on hardware where different entities can spend vastly different amounts of money. (Pay more for more powerful computers or pay less for watts of energy because China computer is near hydroelectric dam, etc.)
The 95% of useful computation (say brute force protein folding or DNA cancer analysis) can be an absolute gain for society. However, the 5% computation's "productive value" is _relative_ to all other participants mining that cryptocurrency. Since others can work through that 5% computation much more efficiently (both in pure hardware cycles and/or energy efficiency), the 5% crypto work on the homeowner's laptop loses value in relation to everybody else. Whatever crypto coins the laptop successfully mines, others have mined even more coins rendering the laptop coins to be worth less because of the relative time & energy it took on that modest laptop cpu.
The 95%/5% ratio that's resistant to ASICs does not stop the inevitable trends towards centralization into major entities holding disproportionate power and/or coins in the system. The 5% "non-useful" computation can still be optimized by people spending more money than others. Since spending is not decentralized, the resulting rewards from the 5% computation will also not be decentralized.
No shit. If it existed, it would be pointless to work on it.
Is it costly just because it was newly created though, and won't this cost come down? You can also mine in countries with high electricity if you build your own renewable energy generator.
In terms of decentralisation, the nodes are still decentralised and can at least detect when miners try to use their dominance. They can only do it after the fact, and it might be messy, but it makes performing the attack fairly worthless as you can no longer mine on the network anymore.
We are all spending time talking about it as a collective effort to determine whether or not it should be adopted in order to optimize the transfer of value. Ultimately it could flush out the middlemen who just seek rent by being a financial gatekeeper. Bitcoin isn't entirely different from that.
Many folks see it as a slot-machine. Buy, sit on it, cash out back to federal reserve notes. Profit. The question here is - what will those proceeds be spent on? If people are still conditioned from an early age to seek out excessive hedonistic experiences, what are we gaining by implementing bitcoin?
Part of me sees bitcoin explanations as a Rorschach test - you can observe some character traits of the person through their explanation or prediction of bitcoin.
I am wary of the entire thing because we are starting to spend more time talking about money. Or making money on money. It just has nothing to do with what we actually DO with our time/resources. But that is content for a different thread.
I am not a financial expert. Just working-class trying to invest surplus cash. I want to use simple terms and concepts and condense the complex discussion.
The losses due to counterfeiting.
I think "embodied energy" is the technical term for that.
You are very kind to point out that the pursuit of money is not an end-goal at all, and it would be great if a lot of these young bucks who made it big on bitcoin were investing their earnings into furthering the globe, instead of just living lavishly.
How's that not a simple concept?
> Buy, sit on it, cash out back to federal reserve notes. Profit.
If someone wins, someone else has to lose. And no, it's not the credit card companies. It's the last one holding the potato.
You (and many others) might need to look up the concept of "mutual benefit"
You can also have positive discussions about money, for example: inequality, altruism, basic income.
I've been thinking about a hypothetical "altruistic" cryptocurrency that could have some rules to combat inequality. It might even be possible to provide a basic income to everyone based on a web of trust, although there's probably a million ways to cheat that kind of system.
Here's an interesting simulation that shows what happens when every person gives a dollar to another random person: https://en.yaronshemesh.com/inequality/
So the algorithm wouldn't do that. But it could have some built-in taxation rules for every transaction, and the taxes would be distributed fairly. A portion could also go towards "effective altruism", such as charities listed on givewell.org.
I just realized that taxes are actually a form of forced altruism. The money goes towards good things (schools, hospitals, roads), but you get in trouble if you don't "donate" the appropriate amount.
It's actually very possible that taxes could be enforced with a blockchain contract. It would be nice if all of humanity could somehow agree on all of our necessary shared expenses, and then encode those laws into some software that governs a blockchain. We wouldn't need the IRS anymore, and nobody would need to donate any money. It could just be implemented as a "sales tax" on any transaction. Perhaps the United Nations might be able to manage the distribution, but I actually agree with Donald Trump when he says the UN is plagued with mismanagement and bureaucracy [1]. So I don't know, maybe we need to reinvent politics at the same time. (Although I wouldn't want a democracy where every single person can vote on every issue.)
Imagine if no-one accepted any currency unless it was backed by a global blockchain. Companies could no longer hide money offshore to avoid (or evade) taxes. All of their corporate income taxes would be calculated and paid out automatically by the global network, and no-one would accept their payments if they don't follow all of the rules.
[1] http://www.scmp.com/news/world/united-states-canada/article/...
Like war (on drugs, iraq, afghanistan), NSA, CIA, etc? And you have also not much choice, if you don't support the spending on the current education system for example.
Thats why it is not a donation. So in my Utopian future taxes would be indeed voluntarily and directly bound - so if there is no money for certain things, then apparently nobody wants them. I suppose most people favor education and streets over sustaining a global invading force, but I might be wrong. Anyway, still a long way to go for this ...
On a per-transaction basis? I don't even need to work out the math to tell me it isn't even close. Bitcoin can at most handle 4 transactions per second. By virtue of its very design it will never get much more than that. Our existing financial institutions process tens or even hundreds of thousands of transactions every second. Even if the entire worldwide financial industry consumed as much power as bitcoin, on a per-transaction basis (or even a per-dollar/euro/yen basis) our existing system is much, much, much, much, much more efficient.
Just to again emphasise, Bitcoin (and hence the blockchain) by design cannot scale much more than it already has. It is simply impossible without giving up all of Bitcoin's attributes that people consider important (decentralization, trustless trust, etc...)
> spend more time talking about money. Or making money on
> money. It just has nothing to do with what we actually DO
> with our time/resources.
It's funny you say this because back before I got into bitcoin I used to see the entire finance/banking industry this way. Now that I'm a bitcoin user/developer, I think this technology will make the finance industry diminish to the levels of where other industries are, because it's basically removing the middle men. It's actually doing the opposite of what you think is doing.
If blockchains and distributed computing does present a new leap forward in how we communicate, this electricity will not be wasted.
Yes.
Cryptocurrencies might be able to serve as a stop-gap before we figure out how to get there. It keeps everyone honest. In the future, it might even replace the entire tax code. There would be no more fines or punishments, and companies would no longer be able to hide their profits offshore. Taxes could just be automatically collected from each transaction, and managed by some global organization (maybe the UN). The world could even agree on a "basic income" contract that solves inequality.
These are all hypothetical examples, but I think this would be a worthwhile use of electricity. Especially if it's all cheap and renewable energy.
Which isn't, you'll notice, necessarily a counter argument to yours.
Vs. Bitcoin, which will still require miners to expend the same amount of CPU cycles to keep the network working so you can transfer your Bitcoins to others, even after all of the Bitcoins are mined and the blocks provide no new Bitcoins into the market.
Also yes.
Roughly 2/3rd of electricity in China is produced from burning coal or gas, which, upon combustion, emit their own weight in CO2, basically. [2]
From this two facts, one can conclude that Bitcoin has a pretty huge carbon footprint overall. This is not the impression that I got from reading that article.
[1] https://bit-media.org/bitcoin/where-in-the-world-are-bitcoin...
[2] https://en.wikipedia.org/wiki/Electricity_sector_in_China
Fossil fuels emit much more than their weight in CO2 when burned. A CO2 molecule produced by burning typically includes two oxygen atoms from the atmosphere.
Some of those plants are not fully stretched, because there's almost no civilization nearby, so delivering to the next city comes with a huge loss in the energy network.
So miners can get cheap energy from those plants - energy that would otherwise be unused or lost anyway.
Not sure how much this saves in the end, but this shows that you can't simply convert energy to CO2 emission without considering which types of plants are used and where these are located.
It misses the fact that China didn't build power plants out where it's too far away from civilisation to be useful.
Transmission losses are also nowhere near the suggested values. 1.5% per 200 miles is today's standard. Meaning you could cross all of China with a loss of:
east-west axis: 2200 miles
0.985^(2200 miles / 200miles) = 0.846 = 15.4% loss
As to miners actually physically located in China, many are located in China’s province of Sichuan specifically due to cheap hydropower, not coal.
What a bunch of crap. Bitcoin also needs servers (i.e exchanges, uses computers, routers, switches etc. )
Assuming all flights are independently profitable, the cheapest ticket I've seen from London to San Francisco is £261 return, which sets an upper limit on fossil fuel emissions of about 6 barrels of oil equivalent. Me plus luggage is about 100kg, which at current prices is worth £3.292.122,21
Utility: a transaction costs $10, right? It's useless.
Security: Have you followed this? The actual recommendation is to print your private keys and stick them in a vault.
Trust: It's funny that, fundamentally, people started doubting the Federal Reserve when the gold standard was abolished, because suddenly "value of money is just fiction". Yet here they are, imbuing some far more ephemeral nothingness with the same sort of faith-based value.
Most people are quite happy with trust-based currencies which don't have the proof of work overhead. Cryptocurrencies have their utility, but not for the whole population.
So people mine bitcoin, which is really just validating the transactions on the network, and they get paid in bitcoin for doing so. However, they will receive fewer and fewer bitcoin over time because of the fixed supply. At some point doesn't mining become unprofitable, causing the network to crash. And if the price drops doesn't this exacerbate this problem? Would love some clarity here.
This would actually cause the price of Bitcoin to increase, so that it keeps up with the cost of mining it. The miners won't sell their Bitcoin for less than the cost of electricity.
At some point, large mining operations will run their own power stations, and they'll just run on solar / hydro / geothermal power. They might even sell any surplus energy. Other miners might start paying for their electricity with Bitcoin, so the price of electricity might become directly related to the cost of mining Bitcoin.
Probably won't be profitable to have a whole warehouse full of nodes but I'm sure there will still be folks willing to make a couple bucks to keep the network up and running.
>The energy that Bitcoin consumes in a year would only last the U.S. for 19 hours.
That doesn't sounds small to me, given the relative utility of 320 million people against that of Bitcoin.
Not to mention that "consuming" energy is somewhat a fiction: it's more so changing the entropy of a system. True consumption of energy, at the extreme, would imply sending energy into deep space (any direction other than the sun really) where it can never be recovered.
Is the problem that such entropy changes usually (always?) result in at least some heat loss? And the the planet is warming, so heat itself is seen as a negative?
To make my point a different way, if a mining rig were 100% solar and 100% efficient (both fantasy, of course), then it could consume unlimited solar energy with no ill effects. Thus basing our judgment on energy consumption instead of waste heat and energy source issues seems misguided.
Is it just too complicating to discuss this in terms of CO2 emissions and waste heat, instead of Watts alone?
Fun Fact: The article says Bitcoin mining consumes ~1.1 GW of power. There is currently a massive political struggle over the construction of "Site C", a 1.1 GW Hydroelectric dam in British Columbia, Canada. Construction on that dam began last year, but public opposition caused a new provincial government to review the dam construction. The dam will flood 93 square kilometers of land but produce 5.1 GWh of low-carbon electricity annually, about 53% of the articles assumed Bitcoin mining consumption of 9.6 GWh (which is = 1.1 GW x 8760 hours/year). Whether construction of the dam will be cancelled or resumed, the decision will be controversial! (The decision will be announced before the end of the year).
In summary, bitcoin mining from renewable energy sources is desirable, but bitcoin mining (and all other forms of computing) will always have some negative impact. That is something we need to be aware of.
But, it’s true that mining using a raspberry pi would not be very useful.
That’s not how halving works.
When you consume more energy than your house does over 1 week to make a single transaction: it’s not efficient.
When it takes hours or days to settle: it’s not fast.
So what is bitcoin, anyway? From what I can tell a way for people to get rich.
Most of these comparisons sounded pretty identical to the ones from the "If I had the intention to lobby for a ban of Bitcoin mining, I would use references like the one below" list to me.
https://digiconomist.net/bitcoin-energy-consumption#assumpti...
Also, their yearly estimate was only 26 TWh a month ago. If we project this into the next year, then we'll end with 130 TWh at the end of 2018 - or roughly 40% of what the UK consumes.
Contrast that with the all the office buildings (e.g. Manhattan) and computer systems required by banks, ongoing. The banking industry has google-size datacenters as well, they just don't like to talk about it. As well as the energy to print the paper/plastic cash and all the statements, etc. It is very difficult to estimate the true energy consumed by fiat, while bitcoin energy is in the hashrate, and that's pretty much it.
An elliptic curve key, which is what possession of bitcoin is, is just a large number. To send a bitcoin you need to create a "transaction", and in fact the only aspect of generating a bitcoin transaction that requires energy is the SHA calculation, not something you can do in pencil, the EC math is actually simple enough that a computer is not required.
Also your argument that cash does not require energy does not apply to the cashless society we have become today, when was the last time you actually used cash? Credit cards are useless without electricity.
For example here is the famous 37 BTC ($500,000+) donation to Andreas Antonopoulos yesterday, this couldn't be done with cash, at least not as seamlessly, inexpensively and quickly:
https://blockchain.info/tx/e68ebe16e6aaa08e3f9d271ba78aaeb4c...
Lots of "users" ( not investors ) interested in Bitcoins are actually .onion sites or other shady / crime organisations.
They actually don't care for human life and I doubt they will care for electricity consumption.