Nobody on HN pointed out who is behind the increase... This is BitFury who just launched a 40 megawatt data center filled up with their new 16 nm chips which reportedly achieve approximately 0.06 joule per gigahash. They also use immersion cooling which gives them an insane PUE of 1.02. So the mining capacity of this DC alone is ~650 Phash/s! We saw an increase of ~200 Phash/s in the last 30 days, so presumably they are at only operating at 1/3rd of their capabilities so far:
So less than half of one datacentre entirely owned by one company is enough to commit a 51% attack. I get that they say they are going to work on distribution these chips to ensure they don't end up in that position but let's be honest they just destroyed any claim that the network is powerful enough to avoid malicious control by one miner now. They've not even raised that much money.
Plus, with BitFury online, the cost of a 51% attack just raised to $200 million.
The only reason it is (currently, for a short time) hypothetically possible to 51%-attack the network with a budget in the low hundreds of million of dollars is because most of the miners are not using such efficient 16 nm chips. As the market migrate to these last generation chips, expect the cost to increase to $1+ billion in the next year.
>People seem to forget that it is NOT in BitFury's best interest to perform a 51% attack because, bitcoin's value will plummet if they do that.
If BitFury can turn $200 million into $400 million now rather than over a couple of years, why on earth would they care about the side-effect of killing off bitcoin?
For the avoidance of doubt, I think that BitFury's best interest is BitFury; as long as they make a decent chunk of money, the end result for bitcoin is - and should be - immaterial to them.
But if that was a predictable consequence, they can't turn $200million into $400million now, because other would anticipate the future state of things and the price would plummet before a couple years. Not that I think that is at all what is going to happen.
As the mining rewards keep lowering it's getting to the point here someone is going to do this for the lulz. As to cost you can sell all your coins before doing a 51% attack. Short the market and at some point it might even be proffitable to do so.
You couldn't buy BitFury for less than 200 or 300 million dollar. After all they expect this $100 million data center to be profitable so I am sure they value their company at least 2x or 3x this.
It is profoundly unlikely a state-level actor would bother.
You can cripple Bitcoin transaction clearing with a thousand dollars to DDOS the network with spam. If only state-level actors had a thousand bucks lying around ...
Malice isn't enough. The only special power of a 50% miner is to consistently resolve double-spend attempts in its favor. If this happened, it would surely lower the value of Bitcoin. This, an attacker must be not just malicious but also irrational enough to forfeit the $800,000 in Bitcoin that it creates daily by virtue of its 50% control of the network. (6 blocks/hour x 24 hours/day x 25 bitcoin/block x $450/bitcoin x 50%)
Not true. A malicious 50% miner can cancel any transaction it wants, not just fix doublespend. Assuming the transaction was sufficiently recent (the more hash power they have the more such a transaction can be in the past).
So they can "unspend" their own bitcoins, as follows : use 51% of your hash power to create a parallel blockchain, which is not published. Include all transactions, except the one you used to buy a TV. Put something else in it's place. Keep doing this until the TV is shipped/arrived. Then publish the parallel blockchain. Boom. Unspent.
If BitFury resolved a single doublespend transaction differently from the "main" blockchain, people would(/should) realize that it now has the power to cancel transactions, and stop trusting bitcoin ASAP. At that point you can no longer trust the blockchain, since you can no longer know you're looking at the "true" blockchain, and not a fake one that has been presented to you to make you do something (e.g. pay someone real money).
In any reasonable person's version of "wait X blocks for confirmation" (currently mostly 3), X would be infinite.
The remaining American Bitcoin traders are certainly gullible enough to keep buying and trading a 51%-compromised coin. (I mean, there are people who still think Paycoin could make a comeback.) But American traders are a sideshow - all the action is in China (miners, actual traders). So the question would be: will Chinese speculators keep gambling on a 51%-compromised coin?
(And of course the MMM ponzi buyers, whose judgement is sufficiently bad that they wouldn't even understand the problem.)
Stupid question: if control of 50% of the hash power is so unimportant, why does Bitcoin rely on hash power anyway? Why not simply have a consensus system without proof-of-anything?
If you remove proof-of-work, then the question then becomes: what else do you measure consensus by? A few other crypto currencies have tried other metrics but they all aim to for the same thing: removing the need for human trust.
I think the reason is because then there would be no way for new coins to be mined and given to those who are rewarded for ensuring the security of the network.
Short first hope the exchanges don't fold and run it into the ground with double spends. Alternatively just attack it because you feel scared about it but I do t believe anyone in power is since it is not going anywhere
A real bad thing here is the asymmetry. To prevent 51%, the network has to burn enormous amounts of energy 24x7 forever while the attacker only needs his energy for the duration of the attack.
It has always been the case that a single company could 51%-attack the network. No regression here. In fact the cost has continually increased over time. So, yes, progress.
I will forever refuse to spend my 0.5 BitCoin just so one day I can say to my grandkids in a creaky old voice, "See! I mined that myself on my own computer in a couple of nights!"
How those 40 megawatts of energy consumption compare to known companies? How much energy for example consumes an average cloud storage company? Or a corporation? 40 megawatts sound a lot but it would be nice to have a reference.
After searching around, it looks about average for a large data center. This link[1] estimates Google's data centers require between 50-100MW.
50MW is enough power to supply ~14k homes from different references I've seen.
I'm curious about what normal data centers use for backup power since natural gas and diesel generators don't get much bigger than 2MW. Or maybe they just don't have backup generators.
I understand that gas turbines can produce that kind of power, in fact I used to work on some pretty large turbines and turbo generators in the 500-1000MW range. However they aren't used as backup generators where they have to be online in a minute.
These are the types of generators I was referring to, https://powergen.gepower.com/products/reciprocating-engines/...
More batteries to get over the slower generator startup time, and likely quite a few smaller generators to do the best they can to keep the batteries going if the big generator is slower than expected to startup that day.
I ran a ~10MW capable facility for a while, we had 8 2MW Cummins gensets that would run in parallel. Extra units for redundancy as sometimes units would be offline for oil changes or other routine maintenance.
They have gas capable engines (50DF series, 50cm cylinder bore) with ratings up to 975kW/cylinders and up to 18 cylinders or 17MW per genset at 60hz.
They're not the only player in that field. Fairbanks-Morse sells the Colt-Pielstick PC2.6 line with 750kW per cylinder at 600rpm and up to 16 cylinders for 11.5 MW @ 60hz. They also sell MAN 48/60 engines that go up to 21MW and MAN natural gas 51/60 G engines up to 18.5 MW.
40MW = 40,000KW
Typical power density in a colocation facility these days is approximately 5kW per full cabinet (sold as 30A @ 208V or 60A @ 120V at 80% utilization). So, approximately 8000 full cabinets at average density, which would be enough power for one of the 10 largest known facilities in the world. Most likely though, the power density is going to be much higher than typical with bitcoin mining, and could be as few as 1333 actual cabinets at the high end of density of 30kW per cabinet.
Right at the moment the current hashrate is 723 Phash/s, so BitFury hashing at 200 Phash/s rate would get ~28% [1] of the newly generated bitcoin in average, i.e. 0.277 * 144 [block/day] * 25 [BTC/block] ~= 996 [BTC/day]. (Assumes that the difficulty approximately keeps up with the hashrate, which seems to be fine to me.) That translates to more than 400,000 USD per day if BitFury manages to sell them at the current price (highly unlikely) without disrupting the market (impossible at all), but yeah, that gives the rough order of magnitude.
[1] This also means that putting 650 Phash/s gigs into the network is not as profitable as it seems, since it will double (55%) the ratio while three times more expensive (if it linearly scales).
Since the amount of Bitcoins introduced into world daily is near constant (except for the scheduled halving of the rewards every few years), SOMEBODY is selling most of the freshly mined coins without disrupting the market.
It doesn't really matter how high the difficulty is or who the miner is the coins are mined and most of them sold.
There might be a small fraction of coins which are still mined in pools using outdated miners and those people don't bother selling them coins(fractions of coins that is).
It has been accepted wisdom that most mining operations sell most/all of their mined coins.
It's a race to the bottom. It will end with the party that manages to just win that race by making the smallest profit possible at the highest efficiency of Joules per hash computed. Bitcoin is interesting for many reasons, the real-world effects of a couple of configuration settings and some cleverly picked auto-scaling parameters are immense. The number of orders of magnitude that the protocol has survived with minor tweaks to date is very impressive.
Anywhere in the USA who has to pay USA power prices is going to be at a disadvantage by that alone then. It would seem that locating the hashing equipment to the place with the cheapest power possible would be a logical step when things start to get cut that fine. Unless that could be overcome with clever usage of naturally generated local power (such as solar power).
You'd think Canada would have more major data centers. They've got a lot of extremely cheap hydro power. You could surely lay ultra-high-speed fiber along the same right-of-ways that the transmission lines use. Also, it's cold as hell more often than not, so cooling is less of an issue than putting them out in the desert somewhere.
USA (Washington state) has the cheapest power in the world (at volume -- places like Iceland are cheaper/nearly free, but with too limited capacity for large scale mining).
There have already been articles about people doing that a couple years ago. I believe they were located in a cooler climate as well, made cooling the datacenter a bit cheaper so more power was going to computing than maintaining the computing equipment.
I'll look for that article now.
EDIT: Not the article I was looking for, but it covers it and was from 2013. This guy was using the cheap Icelandic power and the cold Icelandic air to his advantage.
What you describe is price-based competition. It’s probably the most documented situation in economics, and it present in many markets. It is a race to the bottom, but rarely ever ends; when it does, it is generally by disruption (something people on HackerNews know a lot about). In the mean time, it drives technical innovation after technical innovation—but I doubt, after 40 years of Moore’s law that it will suddenly end.
I'm wondering, isn't this going to put lots of current miners out of business, as their return-on-watt used to mine bitcoins is severely cut? Doesn't this risk to make BitFury even more dominant in hashing power, until somebody else can get comparable economies of scale?
Are these systems totally self contained? Seems a bit unsafe to have fire retardant chemicals vaporizing constantly in a confined space. Has anyone given any thought to the health effects of these type of data centers?
The best way to think about the mining market is that the number of Bitcoins available to be mined each day is _constant_* no matter what the total hashing power deployed is. As a miner your share of that fixed pool of available coins is determined by your hashing power relative to the overall hashing power of all miners. So if you control 30% of the hashing power you should on average get 30% of the coins mined each day. Your profits are then determined by your costs and the value of a bitcoin when converted to the currency your costs are denominated in.
* The constant amount actually changes periodically when the coinbase reward is adjusted downward but that only happens about once every 4 years and historically exchange rate price appreciation has outstripped the reduced coinbase mining rewards. eventually the coinbase reward will go to zero and the number of coins that go to miners each day will be their relative share of the total fees being paid by users transacting on the network. Miners choose which transactions to include in a mined block so in the future a large miner may have some pricing power over transactions because they could refuse to process any transaction with a fee that falls below some threshold.
The question I haven't seen answered that I'm intrigued by: Assuming I put my 2015 pretty-normal-specs (i7 processor, 8gbs ram) laptop to work bitcoin mining 24 hours a day, how long before I have a single bitcoin?
I find this style of commenting passive aggressive... "Nobody has pointed out this great fact?! I guess I will be forced to do it." Just lead with "This is BitFury..." we don't need to know how annoying it is for you to have to post this.
(Note: Since most HN readers are technical, this comment is being downvoted by bitcoin enthusiasts among them. You won't see many replies from them to this comment, since their reaciton is political and not a technical statement. Still, I'll keep it up. This comment is currently at -2.)
To put this into context, by comparing it with a centralized version, an equivalent number of actual transactions could be done at a bank with a crud app running on a $5 complete desktop PC which the Raspberry Pi foundation just released[1]. It is a 1 ghz computer with 512 MB of RAM. It draws up to 0.7 Watts.
Wait, wait, wait. did I say an equivalent number? Since Bitcoin is limited to 7 TPS (7 transactions per second), I should modify this to 100x more transactions (700 transactions per second), if you wrote it in C, or if not a hundred times, then at least ten times as many in python. If it were a crud app running at a bank, a $5 computer could do 100 times the transactions that 650 Phash/s of wasted effort collectively give. And 40 megawatts.
Bitcoin is a supreme waste of resources through the proof of work hack. It is like emulating a GPU in a CPU...in javascript. I mean, sure, you can do that. You can run Open GL by emulating a GPU in single-threaded Javascript.
But it's a stupid, wrong solution. The basic idea sucks. It's bad.
(by the by my target price for btc is $600,000 based on a comparison with how bad gold is physically, and the market cap of gold. :-P.)
I love fiat currencies, they're one of the great achievements of modern civilization. (This sentence isn't ironic, it's actually how I feel.) Despite my target price I don't hold any bitcoins at all. I hope some of the numbers I've given can put into perspective why.
> But it's a stupid, wrong solution. The basic idea sucks. It's bad.
instead of explaining that the cost is the cost of guaranteeing the authenticity of the transaction. HackerNews likes technical comments, and will happily burry expletives.
You could find more compelling example, such as estimate the ecological cost of having most of the current world-wide banking transactions on BitCoin, and prove your point.
You're wrong. Even without any mining, running a full node requires 2GB of RAM. Even if you do less than a full node, you'll find it hard to fully process 100x transactions on less than 2GB.
Also, you don't seem to understand the point of PoW.
He's talking about running something that does bitcoin, but without the inefficiencies of PoW. I pointed out that the RPi 0 doesn't have enough power to handle that scale, even if we took out the PoW stuff.
Hash rate has increased by 41.9% over the same period, so the difficulty has kept the time-to-generate relatively fixed - just the way it's supposed to work.
The real news is that someone or someone(s) have added ~200PH/s worth of processing power to the network in the past 30 days. This is probably from some high-power ASIC miner being released, or from some consolidated mining concern going live.
The billion dollar question is how we can have completely decentralised (and not merely distributed) consensus without proof of work. It's a difficult question, and various proposals like Proof of Stake (not secure) and Consensus Ledger (not secure, not decentralised) have all failed technically and on the market.
For comparison, how much power does Visa, Mastercard, the Federal Reserve (and its printing presses), banks, and all of the buildings and employees that work in the traditional financial sector use? Now think about that in every single country on this planet. It's a lot more than 40MW. The Bitcoin network is a steal by comparison.
The amount of power wasted by BitCoin indirectly depends on the monetary turnover of the system. Because if somebody can spend $1mln to perform 51% attack and move around $1bln, that will likely be done.
Hence, the power wasted is proportional to the turnover of the system. Otherwise, it does not work.
Also, when a bank charges me 1% for a wire transfer, they pay salaries from that money and some people buy food. The power burnt by BitCoin is actually literally immediately irrecoverably physically wasted.
> How would bitcoin's power consumption stack up if it were tasked with servicing tens of millions of transactions per second?
There is no way to do this calculation. At the moment Bitcoin mining uses, let's say 150MW and can do around 4 transactions per second. If you gave the network a 1000 times more hashing power, you'd be using 150GW, and could handle... around 4 transactions per second.
Well a miner like the one going live is also responsible for managing the transaction volume. So the capacity of Bitcoin to manage many more transactions just went up.
The federal reserve has no printing presses. VISA can process 10,000x as many transactions per second as Bitcoin and banks do a lot more than transferring funds.
Bitcoin feels a bit like the gold standard. Massive mining operations dumping huge volumes of resources and energy to acquire some thing that is only mildly useful. It seems quite wasteful.
Excerpt:
According to our estimates then, the whole Bitcoin network is consuming maybe 10% of a large power plant’s worth of electricity. Although this is not an insignificant amount of power, it's not yet a large amount of electricity compared to all the other things that people are using electricity for on the planet.
Any payment system requires energy and electricity. With traditional currency, lots of energy is consumed guarding and moving gold bullions around, running ATM machines, coin sorting machines, cash registers, and payment processing services, and transporting money in armored cars.
Some people say Bitcoin wastes energy because the energy expended computing SHA-256 hashes doesn’t serve any apparent purpose. But you could make this same argument for traditional currency as well — there’s a lot of energy being wasted and it doesn't serve any purpose besides maintaining the currency system. So, if we value Bitcoin as a useful currency system, then the energy required to support it is not really being wasted.
That said, we can ask if there’s a way to do better ...
I believe that most of money consumed by the financial system actually goes back into the economy. Some bankers buy sushis, then sushi chefs buy gasoline, then oil drillers... and so on.
BitCoin tends to waste energy irrecoverably.
Apples-to-apples comparison is a challenging task, but the modern financial system hardly has that feature of burning more for the sake of burning more.
> I believe that most of money consumed by the financial system actually goes back into the economy. Some bankers buy sushis, then sushi chefs buy gasoline, then oil drillers... and so on. BitCoin tends to waste energy irrecoverably.
Bitcoin is best technical solution we have for an existing societal problem. As soon as better technical solution will be created (e.g. proof-of-stake which can be relied upon), Bitcoin will be modified to use it. And Bitcoin will continue to operate with lower costs - burning less energy.
Money is returned to the economy via energy company profits and wages just as effectively as via bankers' salaries. It's the labour and energy itself that is wasted.
> I believe that most of money consumed by the financial system actually goes back into the economy. Some bankers buy sushis, then sushi chefs buy gasoline, then oil drillers... and so on.
Yes and no. If a banker is being paid $100/hour then presumably (in an efficient-markets sense) their labour is worth that much, and they could be doing something else with their time that would be creating $100 of value instead.
>Any payment system requires energy and electricity. With traditional currency, lots of energy is consumed guarding and moving gold bullions around
Whenever the topic of bitcoin's inherent wastefulness is discussed, someone always brings up this point, but it's a fallacious comparison because most of the power consumed by the traditional financial system is spent in its capacity as a ubiquitous pillar of modern society, wherein bitcoin would be completely subsumed were it to become anything more than a technical novelty.
Even when you subtract the common denominator that represents the vast majority of the world's financial intuitions, bitcoin is still the only currency that burns resources as a function of its circulation.
>most of the power consumed by the traditional financial system is spent in its capacity as a ubiquitous pillar of modern society
Either I missed your point or it would be nice if you brought this back down to Earth. Yes, traditional ledgers help mediate economic exchanges. Bitcoin, as a ledger system, also helps mediate economic exchanges.
> ... wherein bitcoin would be completely subsumed were it to become anything more than a technical novelty.
You put this forward like we should all nod and say of course. Care to tell us about this scenario?
That paper estimates the bitcoin network's total power consumption at around 117MW. This latest increase is estimated at 40MW. So it is a significant increase.
I forget where I read it, but when countries and banks buy and sell large amounts of gold bullion they don't usually move it around because it's expensive and inconvenient.
The ownership changes, but it's usually left where it is.
Stellar solves the same problem in a different way - nodes pick whom to trust in the network. So it doesn't burn vast quantities of energy just to stand still, and it's not vulnerable to someone with vast computing resources.
I'm not an expert but it makes a lot of sense to me. Most real-world decentralized institutions work kind of like this.
And all of the myriad redundant banks, central banks, credit card processors, buildings in every city in the world aren't wasteful? Bitcoin is a monetary system for the whole world and if it succeeds none of the things I previously listed will need to exist.
Proof of work is the most secure way we have by far of running a decentralized currency. I am hopeful that a better, more energy-efficient method will be developed, but until then, I think advancing the state of the art is worth the energy cost.
It's not obscene to me that a truly global currency, that is decentralized and not controlled by any government, would cost a tenth of the energy output of a modest-sized power plant.
That's my main issue with bitcoins (aside from how they're marketed to anyone; it is most definitely NOT anonymous!).
Instead of 'proof of work' it should require actual, /useful/ work. I think it should be a mix of work /types/ to promote general purpose computing, instead of ASICs. Imagine if a comity decided, and the owners of existing coins voted on, what work was worthy of being included. Folding proteins for medical research, SETI, attacking DRM/bootloader encryption keys... things that benefit the world.
Ok, yeah that would be nice. But no one has figured out how to securely do productive work for proof of work. In fact, there are pretty good reasons to believe that it may even be impossible to create a secure yet productive proof of work algorithm.
You can't use SETI@home for proof of work. The work target has to somehow involve a fully random process.
Lots of very smart people have tried this and failed. The best effort so far has been to find prime numbers, and even that turned out to not be a robust proof of work.
It might seem mildly useful until you realize why bitcoin, gold, or fiat currencies all have value - their properties lining up with the properties of ideal money. Gold is not valuable because of jewlery or industrial uses. It is valuable because it was the original decentralized shared ledger. Easy to verify, easy to divide, hard to create, hard to destroy (and some more if you look it up).
Once the 16nm chips become a commodity, it will make sense that the hashing power would get more distributed.
I hope figures out how to get a solar powered, interchangeable bitcoin miner in a box at a positive ROI. It may seem impossible now, but solar prices are falling faster than expected.
Why would you expect hashing power to become more distributed? Mining is a race to the bottom/thin margins, and the differentials in power rates far exceed normal miner profit margins. It would still only be profitable to mine at scale in a few select locations in the world.
"Valery Vavilov, CEO of BitFury, said: “We are very excited to launch mass production of our super 16nm ASIC Chip. The final results of our hard work have fully met our expectations. We understand that it will be nearly impossible for any older technology to compete with the performance of our new 16nm technology. As a responsible player in the Bitcoin community, we will be working with integration partners and resellers to make our unique technology widely available ensuring that the network remains decentralized and we move into the exahash era together. BitFury warmly welcomes all companies interested in joining our integration and reseller program.”"
Yes it does. If the article is correct, the 16nm BitFury chips will be sold to interested parties, so new mining capacity won't be dominated by a single company, thereby ensuring newly mined BitCoins can be distributed throughout the network.
If your concerns were about something other than distribution, please clarify.
They will be sold to interested parties.. and hosted in the same data center. Back to square one.
It takes a price per kW-hr differential of about $0.03 before it makes sense to put miners on a plane. That's far less than the power differentials one would expect, so those miners will stay put. Just paper ownership will change hands.
Isn't that logically impossible until renewable is cheaper than fossil fuels?
I'd have thought that because fossil fuels are cheaper, it will literally never be possible to have have a positive ROI bitcoin miner on a renewable energy source, assuming that people are willing to mine using $0.99 of energy to make $0.01 profit.
EDIT: Downvotes and no comments for a simple question? WTF is going on with HN these days, can we not ask questions?
Does the new capacity make a bitcoin more valuable or less valuable?
Intuitively seems like more valuable, but the average cost in energy to mine the marginal block has gone down (otherwise the new miner wouldn't be mining) and that's often though of as the floor on BTC value.
Seems like having a stronger network is a net plus, and since they're probably not near the 50%+1 threshold it probably is in fact a stronger network. Although maybe they will get close if marginal miners are forced to turn off if the price of XBT drops and BitFury is enough more efficient.
Although BitFury is probably not a bad actor, technically a 50%+1 attack is not obviously illegal (IANAL), although a government might step in ironically enough. It seems to me these types of more centralized setups do introduce some tail risk to the system.
> we now know that there is a near precise
model that describes the cost of running and maintaining the network. The way the
cost estimate is determined is through how Bitcoin acts as a decentralized waste heat creator
that activates and deactivates heat generation based on market participation and pricing
signals. What do the randomizations necessary for cryptography and the waste heat produced
by computing devices have in common? One word: “exergy,” a term of art describing the
maximum useful work possible during a process that brings a system into equilibrium with a
heat reservoir. Exergy is always destroyed in the seigniorage hashing process - for example - if
a token's value increases to $1,000, this means that at most $1,000 worth of waste heat will be
generated somewhere in its creation.
From my reading of the text, it's not so much the additional hash power that's valuable, it's the additional money spent building and operating the ASICs. In theory, the market cap of a proof-of-work system should approach its total cumulative cost to secure. The more watts you see being dumped into the environment calculating hashes, the more you should value Bitcoin.
If Bitcoin is worth less than its cost to mine, no rational miner will mine (if they want BTC they'll just use their electricity budget to buy it on the market), so the competition (and therefore the cost) to mine each block goes down. If Bitcoin is worth more than its cost to mine, mining becomes profitable to anyone willing to put up the capex, so the competition (and therefore the cost) to mine each block goes up. There's an equilibrium where the value of Bitcoin is equal to its cost to mine.
The actual price of bitcoin as seen by the average consumer is insulated from the cost to mine because of effects like speculation, perceived future movement, and the value provided by ease of spending / anonymity, so the economic theory isn't really accurate, but that's why it's a theory!
>No one is afraid of the ecological cost of crypto currencies ?
I am. But this is a drop in the bucket compared to financial sectors that deal with "real" money. Really, isn't nearly all modern money "virtual"? Nevermind the fact that a lot of money only exists electronically, modern money doesn't actually physically represent anything. Therefore: virtual.
The problem is not the virtual character of the money, but the energy involved to create it.
The idea behind it is that you need energy to gather money (metalic money for example). but this is pointless for bitcoin, it's just wasted without any utility.
looking at the chart there are plenty of places there was more increase in difficulty in 30 days. In the whole 2014 difficulty seemed to have multiplied 40 times! I understand the magnitude is a whole different story now though..
More powerful miners beget higher work needed to mine. A small cadre of people with ultrpowerful mining gear have pushed a more difficult task on the rest of us.
Not 100% sure, but from your tone it sounds like you think they are performing some kind of service. This is not true. Miners do not increase the number of transactions the network can handle. An infinitesimal amount of the electricity going into mining is actually goes to process transactions. A raspberry pi in a shoebox running mySql is capable of processing more transactions than the entire bitcoin network, liquid nitrogen and all.
All that mining does is give the person with the most hashrate more voting power in which transactions will be accepted. Innovation in mining hardware and data centers does not in any way increase bitcoin's security.
> All that mining does is give the person with the most hashrate more voting power in which transactions will be accepted. Innovation in mining hardware and data centers does not in any way increase bitcoin's security.
This is completely wrong. The higher the difficulty, the harder it becomes to attack the network, which is a critical feature. If anyone who rented a data center could attack Bitcoin at this point, it would be even less stable (much less!) than it has been already.
No. Read what I said. "Innovation in mining hardware and data centers does not in any way increase bitcoin's security."
If everyone can use technology to push a higher hashrate, then what has changed? The people who can get their hands on new mining chips first have an advantage for a little while, but in the end it's a zero-sum game. The pie is 100% of hashing power/block validation power. How can you divide a pie into more than 100%?
If a dedicated data center is more effective than a commodity data center then that increases bitcoin's security, because it means a greater proportion of the hashrate resides with people who have an expensive long-term commitment to bitcoin and it's more expensive for an outsider to match that.
That's circular logic thogh. If you built a data center capable of staging a 51% attack, you haven't made the network harder overall. You've already gained enough power to do an attack.
It might encourage others to compete, and the arms race continues. But if too much of the network is controlled by a minority, then it's not automatically more secure.
Only certain types of attacks are based on hashing at a rate which is a significant percentage of the network hash rate. Many other types of attacks have nothing to do with hashing, for example, the transaction malleability attacks.
>Innovation in mining hardware and data centers does not in any way increase bitcoin's security.
Not sure exactly what you mean here - mining, and mining faster, generally increases the computational resources another third party would need to 51% attack the network.
Only by as much as the miners themselves spend. And it's a running cost, you have to spend it every day, while an attacker would spend it only during their attack.
The only way to make a 51% attack impractically expensive is to make the network impractically expensive to run.
Anyone with enough compute can attack the network in this way. There's various scenarios depending on how much compute the attacker has, but in general as they approach 51% of the network's compute, they can start to reliably do double-spending attacks.
But, and this is arguably one of the cleverest parts of Bitcoin's design: Who is going to get all that compute setup to mine Bitcoin, and then break the very system that makes it worth having?
Not saying there couldn't ever be scenarios where it happens, but its a pretty good first deterrent to bad behavior - and its clearly intentional, mentioned in the original Bitcoin paper:
"If a greedy attacker is able to
assemble more CPU power than all the honest nodes, he would have to choose between using it
to defraud people by stealing back his payments, or using it to generate new coins. He ought to
find it more profitable to play by the rules, such rules that favour him with more new coins than
everyone else combined, than to undermine the system and the validity of his own wealth."
In some ways, the existence of custom mining hardware is a bad thing, by reducing the spread of participants who can mine cost-effectively - essentially un-democratising the running of the system.
But on the other hand, its meant that an attacker has to invest in a lot of special-purpose hardware to attack the system (rather than just being able to e.g. rent enough EC2 nodes, or turn their entire government's cluster against Bitcoin, or whatever); which probably increases the cost of (then) destroying the system, and makes Satoshi's original Incentive argument stronger, imo.
If they wanted to torpedo the price, get routed around, and lose all that capital investment in 16nm mining fabrication. Bitcoin was designed so that attacking the network will be less profitable than joining it.
But I can imagine scenarios where its game theory might not hold - because "profitable" is measured in bitcoins. For example, Bitcoin might not be suitable for use as a reserve currency if doing so would create "political profit" for destablising bitcoin that might exceed the monetary cost.
It does specifically because it increases the barrier to entry. This means the likelyhood of an attack is lower and the cost of an attack is higher. If your Rasberry PI ledger had a market cap of multiple billions of dollars, the likelyhood of an attack that would either steal people's money or shut it down would be pretty much %100.
There are many shades of grey. It creates some centralization, thought incentives of the miners are still aligned with the incentives of the people using the currency.
You seem to be implying total centralization, which is not at all true.
I also have paper draft about better Proof-of-Stake protocol, and would like to share it with people from academias to get a feedback. Please write me ( kushti at protonmail dot ch ). I also have half-written paper draft about PoW+PoS hybrid chain.
Like precious metals? Deflationary currencies have been used successfully for over 1000 years.
The only people it is bad for are governments. Throughout history there is a patter of promising, spending, and becoming insolvent.
The idea that an individual wouldn't choose a deflationary currency is ludicrous. Inflationary currencies are what you want everyone else to use.
And by the way, the cat is out of the bag. Cryptocurrencies are here, bitcoin or not. When people can choose to use any currency they want, will they choose one that inflates? I doubt it.
So let's see... We have an inflationary currency which is behind an increasing divide between the rich and the poor[0], and keeps people hooked on unsustainable growth to just survive, leading to intensifying environmental destruction and wanton consumption of limited resources... It's not like an inflationary currency is necessarily a good idea, either.
[0] before you say that inflation hurts the rich, among other reasons the if inflation didn't hurt the poor, we would never feel the political pressure to increase the minimum wage.
Then blocks continue to be created, but no more coins are mined (in practice, we will be mining tiny little fractions of a coin for a very long time until it drops off to true zero).
If no more coins are mined (or the mining rate drops to almost negligible levels), is there still any incentive for people to keep the blockchain alive by computing power? Aside from keeping the whole system alive, I mean.
What a colossal waste of electricity. Great, they've created a general ledger and currency formed by individual untrusted participants, but who in the aggregate are trusted.
What about this: Why not just diversify your risk by doing transactions or investing in currencies/assets across a diversified set of untrusted counterparties? Same net effect, and a lot less electricity wasted.
Huh? How does this "diversifying risk" idea let you do any of the major bitcoin applications? (For example, how would your idea that has the "same net effect" as bitcoin allow anonymous markets?)
Bitcoin is not anonymous. The entire ledger history is exposed. Through network analysis you can figure out who the original anonymous holder is. And as soon as that holder tries to convert to a fiat currency their identity will be exposed. It's precisely because Bitcoin is way more traceable than cash, the US government has not tried to shut it down despite a lot illegal activity being paid for via bitcoin.
As soon as Satoshi tries to convert any of his coins into dollars or any other legacy currency everyone will know who he is.
Actually, we're both wrong. Just because you have not been caught does not mean there hasn't been loss of anonymity. And if you mix enough coins you can decrease the chances of being pinpointed exactly as the source, but you cannot definitively remove yourself from the bucket of suspects.
>And if you mix enough coins you can decrease the chances of being pinpointed exactly as the source, but you cannot definitively remove yourself from the bucket of suspects.
Not in my understanding. I have some coins X that are tainted. I send them to you, Y. You, Y, happen to have other coins completely unrelated to the address I sent my X coins to, and you send them to me at address Z. There's no blockchain link between X and Z.
I suppose physically tracking you down, assuming you keep logs, could hurt me, but if I trust you not to keep logs then I'm safe after you delete the logs.
40e6 (watt) / 0.06 (joule/gigahash) / 1e6 (petahash/gigahash) = ~650 petahash/second
BitFury's immersion cooling tech:
http://datacenterfrontier.com/immersion-cooling-bitcoin/
https://www.youtube.com/watch?v=uV7MDhqNyXE&t=0m42s (shows the fluid boiling - starts at 0m42s)
https://www.businesswire.com/news/home/20151211005837/en/Bit...
BitFury's 16nm chips:
http://www.businesswire.com/news/home/20151216005453/en/BitF...