Bitcoin Forum

Been wondering. What happens when all the coins have been released? Does everyone just stop mining? Would that leave just people who want to contribute to the cause hashing to verify transactions? That seems like it would make it quite easy for someone to gather a modest amount of hardware and hit that 51% mark.

OR will there still be rewards for hasing? Such as the transaction fees being distributed?

Yes, bitcoin will get killed by attacks or alternatively by high fees when this happens unless the protocol is changed. Actually, death due to decreasing rewards will likely happen much sooner than when rewards drop to zero. Even with the current maximal reward we are already getting botnets and lone individuals who aspire or have some share of hashing power in the 10-30% range. This is known as the tragedy of the commons problem. Proof-of-stake is the solution.

https://en.bitcoin.it/wiki/Proof_of_Stake (https://en.bitcoin.it/wiki/Proof_of_Stake)

Note that there is a strong desire among many of the bitcoin community to wish this problem away. Demand a logical argument providing an answer to your question. Don't let them get away with hand-waving.

 Ask what will happen to txn fees as well. Will the network be secure? Will txn fees be competitive with paypal? In my view, the current protocol can potentially give a yes answer to only one out of two of these questions (that is pick one). It is impossible to obtain yes answers to both under the current protocol.

The hope is that transactions are enough of an incentive to keep the network happy. Currently, it wouldn't be, and that is proven by the fact that 1/3 of the blocks are purposely mined empty, to increase profitability.

Yes, miners receive transaction fees in addition to the new coins being minted. As cunicula points out, there are doubts that transaction fees alone will be enough in the current system.

This same question has been getting asked quite a bit lately:

Transaction fees will need to replace the currency issuance in the block reward.


 - http://en.bitcoin.it/wiki/Myths#After_21_million_coins_are_mined.2C_no_one_will_generate_new_blocks


 - https://bitcointalk.org/index.php?topic=71176.0
 - https://bitcointalk.org/index.php?topic=75138.0

I seem to recall seeing something here about one of the alternate blockchains dying from inattention, someone posting something to the effect that several episodes of corruption and a need for recovery preceded the ultimate demise.

The "spinning plates" ( a reference to the stage show where dozens and dozens of spinning plates are balanced on sticks and kept spinning by the performer running from one to the next to keep them spinning ) model of bitcoin operation has always given me doubt. The state of the blockchain freezing completely is not so bothersome, but what happens if mining declines is. With all the hashing horsepower that has built up out there it seems rather likely that an attempt to seize control of the blockchain could be successfully executed. Unlike the alternative blockchains there is enough value locked in bitcoin to make it an attractive target.

Miners will continue to expect more and more value out of blocks. By completely removing block rewards and relying solely on transaction fees, the miners will eventually expect nearly infinite value out of blocks. When that is not possible mining will happen less often, leading to lower confirmations. It could happen soon after a block-reward-halving. In fact, I suspect that for at least this first block reward halving, we will see the price double or more than double to accommodate the eventual change.

Whichever are the governments of the world at some point in the future will inevitably step in and demand X% transaction fees. It will be a very low number at first and gradually rise. Miners that refuse to include transactions with these fees in their blocks will be punished, even if they are doing so out of protest of the government, at this point controlled by the subsidized miners who now funnel lobbying money into the government.

I mean, we're talking decades probably before this happens again, but like with all things, it will happen. I'm guessing maybe mid-century some stupid mining operation run by an incompetent fool (I insult them here for the sake of posterity) will purchase equipment prior to a block-reward-halving and expect to have it paid off as price rises, but the block reward will halve without a significant move in the price of coin and they will lobby the government to force every Bitcoin enterprise operation in whatever the biggest Bitcoin adopting country is (Japan or China?) at that point to increase or mandate fees.

Again, the idiotic populace (again, insulting for sake of posterity) will conform to these fees and willingly give them up in the name of equality or security or some other contortion of values that the power that be deem as necessary for the "good of man".

In reality, the only threat that happens when the coins dry up is that the difficulty drops to a low enough point that the cost-benefit-ratio for doing a 51% attack increases. That threat alone pretty much ensures that anyone who has a significant stake in Bitcoin (processes lots of transactions) purchases their own equipment instead of piggy-backing on the miners themselves.

So the answer to your question: "Does everyone stop mining?" is yes, people stop mining, and the transaction processors are the only ones still finding blocks. We have yet to discover who that will be. My guess is that the transaction processors will essentially be the governments' biggest lobbyists at the time, though.

Yes, I have a very pessimistic and dystopian outlook. In reality I'm hoping it's not that bad, with multiple alternative currencies to compete with Bitcoin that do not have some of the same flaws. I'm worried that rather than make a new currency, the developers will be under increased pressure (vis-a-vis the miners, perhaps?) to fork Bitcoin to always provide some form of block reward. I think I'd rather see an alternate crypto-currency than a fork of Bitcoin.

How much is the typical transaction fee awarded when a block is found right now?

~0.04 BTC

Currently, cost per txn is about US$4-5. Of that, about US$0.000571428571 is txn fees [4 BTC per day in total fees/8000 txns per day; source blockchain.info]. The rest is block reward. That is txn fees make up about 0.01% of the funds supporting network security.

Currently, it would cost roughly $7-8 million to 51% the network. If the network was supported solely by these tiny fees, the cost of 51%-ing the network could be expected to drop to about US$700-800. A couple of 5970s would do the job nicely.

That's the amortized cost. I hope nobody mistakes that for the marginal cost (which is close to 0).

The coins won't dry up for over 100 years.  And the last set of blocks before the "drying up" will be for  0.00000001 BTC per block.  And the before that period, the blocks will be worth  0.00000002 BTC each.

It's a very gradual process, and there will absolutely be a crossover point where transactions become the larger part of the block reward.

Someone with a brain.

Whuuuua?

Problem?

1/10 because I replied.

The idea is that in 2033 or whenever they do dry up, the reward would have halved so many times that we are used to the tiny payouts at that point and the value of a bitcoin (should) be dramatically higher than it is now, making mining profitable even at the end.

You'd be better off just letting the topic die. If you can't think of a convincing answer to your opponent, the best strategy is to ignore him.

This:

http://msnbcmedia.msn.com/j/MSNBC/Components/Photo/_new/pb-110414-easter-splash2-ps.photoblog900.jpg

FYI, transaction fees per day: (logarithmic scale)

https://blockchain.info/charts/transaction-fees?showDataPoints=false&show_header=true&daysAverageString=1&timespan=&scale=1&address=

I do wonder what would happen when the coins dry up?

I hope I can get some mining things (a BitForce single or a cluster as Gfx cards use too much energy for what they give back to you-The HD5970s and HD6990s are an examples) so i can get a piece of the pie while the block rewards are still worth getting (50BtC,25BTC,etc). .I'll have to take out some (large) loans in future to help buy them then pay them off at some point in the future as I need to consider some things first (how many computer like things,can I fit into the airing cupboard (as I live in a place where very small budget like apartments are) without too much heat,noise,energy use as I hear of coppers busting ppl for using too much energy and to stop the neighbours  complaining about noise as I live next door to them)

The thing that will happen is that people with dedicated rigs will start pulling out as block rewards drop.  But on the flip side difficultly will also drop.  Once the block reward drops enough (and the difficultly with it) you would see more people solo mining (outside of pools).  The first drop in block reward (and I believe bitcoin's first real test of longevity) will be in this December.  The thing to watch is how the total hashrate and difficultly and how quickly.  I think there is a built in max change in difficultly of 400% up or down.  As long as the total hashrate doesn't drop too much faster than difficultly, you wouldn't notice a thing.  Also, as long as there isn't a sustained exudes of people from the community, you could expect the price for coins to rise (slowly) as people turn to buying coins instead of mining.

As long as the btc value keeps rising steadily to some point, fees will cover just fine for the miners i think. If the use of bitcoin doesnt grow, yea, then we are screwed.

The eventual lack of new coins in misguided imo. There should be at least some new generation for replacement of lost coins and to encourage a low level of inflation.

Fees don't seem like a viable option. Coins are realistically in competition with real world payments, and at best would garner a couple % for fees (capped by the CC processing fees). However, since they are paid by sender and do not offer benefits that CC affords, I would venture that the fees should be more in line with fees for ACH (or equivalent systems), and thus, not that much.

An implosion of miners is NOT desirable since there has been much effort put in to lower costs of hashing. One could amass large amounts of hash power on the cheap and mount a 51% attack.

Paypal transaction volume is ~ 2.2 billion tx annually (VISA is ~ 78 billion and all electronic tx is probably on the order of a trillion tx annually).

Still lets consider a 30 year goal of 40% the size of Paypal =~ 1 billion tx annually.  That is ~ 20K transactions per block.

ACH fees are ~ $0.20 per tx.  So 20K tx * $0.20 ea = $4,000 per block block reward.  Still I doubt fees would need to be that high.  At ~$0.05 (whatever that ends up being in BTC) per tx avg block reward would be ~$1000 which would support a network 5x as strong as the current one.

This.

This is ridiculous. Paypal is worth about $50 billion. This is the discounted present value of collecting monopoly txn fees from money sent through paypal. You can currently buy a similar monopoly over bitcoin for $10 million. Let's do some back of the envelope calculations to show that D&T's suggestions are idiotic.

D&T proposes that the network will be 5x as strong as the current one, so that the cost of obtaining monopoly will increase to $50 million. However, the value of the monopoly will be about 40% of the value of paypal's with 20k txns per block. That is monopolization of the network will be worth $20 billion. Hmm. $20 billion dollar asset priced at $50 million. Sounds like a very good deal. The network would quickly get owned with such puny fees. The owner would jack up fees to paypal levels and become very rich indeed.

So what kind of fees would you need to protect bitcoin from getting owned. Well you need to get that $50 million up to $20 billion. D&T's figures are off by about a factor of four hundred. How much do the fees need to be 400*$0.05=$20 per txn. Wow! $20 per txn, great low cost payment system that would be.

Proof-of-stake is the only long-run viable solution. Stop joking around with this nonsense about txn fees coming to the rescue. It doesn't pass the laugh test.

You bring up a good point, but also consider:
- That as Bitcoin becomes larger and more stable in value, people will be willing to take smaller ROI's on their mining equipment investments, thus increasing the "monopoly buyout" of Bitcoin.  Right now, the ROI is around 7 months for a BFL miner (for example).  Judging that a good stock might be somewhere in the 15 P/E ratio (in other words, ROI would be roughly in 15 years if no earnings were reinvested into the company), you could expect size of the mining network to eventually reach around 26 x as much as it currently is.
- That the fact of immediate devaluation of Bitcoins due to a monopoly would likely mean no one would ever attempt it.  What good would it be to own a monopoly over a transaction system that no one uses anymore?  After all, right now, assuming one could buy a monopoly over Bitcoin for $10 million, they could gain a potential $250/block, or $36,000/day, or $13M/year.  But if the price crashes, that monopoly does no good.  It would be a terribad investment strategy.
- That it would also take a lot of time and coordination to pull off a majority share of mining.  Time in which others may hear about the plan, and fight against it by buying and utilizing more mining hardware themselves.

Mmm, I might agree to disagree with you on that one.  It's certainly possible, but I tend to believe that people would revolt against a hostile takeover, and would likely just continue to utilize Bitcoin on the same fork/rules than follow a new fork controlled by a centralized agency.

You can't 'buy' it. If it is worth $20B and somone 'buys' it for $50M then someone else will come along and 'buy' it for $100M and so on. But that won't happen since you don't get to keep it for $50M and anyone with $50M will know that.

Maybe the same thing will happened as when the gold mines in San Fransisco dried up?  The price went up.

For all the complaining you do about your perceived economic inferiors relying on intuition and belief instead of "real data", you sure do come across as a zealot when it comes to your own imaginary musings...  

We are a long way from pure transaction fees.  We haven't even seen the first reduced block-reward yet.  The suggestion that proof-of-stake is the *only* solution to a *hypothetical* problem is what doesn't pass the laugh test.

Oh another cunicula rant. 

So imagining a hypothetical world where Bitcoin is 40% of the size of Paypal.

1) The "market cap" of Bitcoin wouldn't be 40% of Paypal because Paypal market cap isn't just based on tx volume but also margins (fees) and barriers to entry.  Things that Bitcoin lacks.

2) No company could just pay $100M and "buy Bitcoin" and then raise fees to generate the $20B in value.  It is idoitic.  THERE IS NO BARRIER TO ENTRY.  Company A "buys" Bitcoin for $100M.  Ok.  Company B "buys" more hashing power than Company A for $200M so Company A loses everything.  Except company C can send $500M and bankrupt both company A & B.  Company D can spend $1B to "control" the network except Company A, B, C and private small miners could combine to destroy Company D investment.  With no barrier to entry there is no way to force prices high and no way to prevent catastrophic losses when someone else does exactly what you did.

3) Obviously anyone who has hundreds of millions of dollars to throw around can go through the thought excercise in #2 and realize it is massively risky to even attempt #2 which means it likely will never happen.

4) Today margins are high on capital because risk is high.   If hypothetically 20 years from now Bitcoin is 40% the size of Paypal with 2 decades or operating history, billions of tx annually, and millions of users the risk is much lower and as a result the return on capital will be much lower.   Today ROI% on deployed capital are >100% annually.  If that falls to even 10% annually (i.e. cost of network is 10x the annual revenue) then we aren't talking about $50M in deployed capital but $500M in deployed capital chasing the same block rewards.

I agree however it just tiring to hear the never ending half thought out rants and the venemous bile coming from him.

There are significant issues with proof of stake, namely it is almost certain that either a monopoly or oligarchy will form.  Potential hashing power is infinite and is fees rise that create an incentive for new entrants to enter the network.

With proof of stake once an entity (cough cough bank) or cartel of entities (Chase + BofA + Wells Fargo) gain 51% of the coins they do OWN the network.  Their control is absolute and the "barrier to entry" becomes infinite.  Under such a scenario Bitcoin's fees (assuming network is widely adopted) would rise to be similar to Visa. Paypal, facebook tokens, etc.  It is naive to think otherwise.

I don't want to say PofS is impossible or unviable but there are serious serious issues which need to be addressed.  It is almost a certianty that Bitcoin will never use PofS so if Sir RantsAlot had any common sense he would be working on a ALT-CHAIN which implements PofS.  PofS would be a breaking fork of Bitcoin.   Even if it had popular support, support will never ever ever be absolute and that means implementing it creates a permenanent fork where both chains co-exist. It simply isn't happening.

So the logical plan instead of spitting venom all over every single thread would be to implement PofS in alt-coin and if he is right then eventually it will supercede Bitcoin and become the dominant crypto-currency.

This is essentially what D&T said. Yes, an absurdly low price (due to market failure under a competitive network) would likely attract multiple competing would-be monopolists. Competition among these entrants would bid up the entry cost until it was sufficiently large to deter further entrants. Competing companies would end up merging, since this merger would add tremendous value for their shareholders. You would end up with one monopolist and astronomical fees. Barriers to entry would now be large because the hashing horde of the single monopolist would be large. The monopolist would impose monopoly fees, depressing coin value. Proof-of-stake is the only solution.

Maybe you are hoping for antitrust policy to save the day? Or perhaps a directly government controlled network?

I'm surprised that a system supporting such a depressing range of long-run outcomes could ever be widely supported. It is probably due to myopia and weakness in abstract reasoning ability among forum members. Certainly, no has ever provided logical support for the possibility of sustainable competitive proof-of-work under txn fees. If you mistakenly believe that this is possible, please explain why you think this and I will try to clear up your confusion.

No, idiot. Try to use your brain when formulating a response or defer to me if you don't have one.

With proof-of-stake there is no incentive to raise fees to high levels under monopoly. This is a critical difference from proof-of-work.

Under proof-of-stake, the monopolist would own 51% of the coins. Increasing fees would depress the value of these coins grievously injuring the monopolists bottom line. Rather than choosing fees to maximize fee revenues (as in proof-of-work), the monopolist would jointly maximize market capitalization and fee revenue. This would lead to very low fees. The fact that high fees depress the value of the network would restrain the monopolist from behaving like paypal.

Finally, proof-of-stake makes monopoly less likely because monopoly becomes more expensive to achieve. It also removes any possibility of increasing returns to scale in hashing technology which is a feature encouraging monopoly under proof-of-work. Nevertheless, I agree that a monopoly may still emerge even under proof-of-stake. It is just not as certain to occur.

I don't see how or why you think someone owning 51% of all coins would be better than someone mining 51%.  Either one would be a fatal blow to Bitcoin.

If this is your point of view, then note that owning 51% of the coins is substantially more difficult than mining 51%. Stop there.

If bitcoin became widely adopted, no one would care if it was owned by a single entity or owned by a community. It is only the current enthusiast community that cares about stuff like this. I'm not sure why. Regular people care about getting payments from point A to point B with minimal hassle and minimal fees. Bitcoin can handle minimal fees under monopoly provided that it is organized as proof-of-stake. If it is organized as proof-of-work, then monopoly will bring ruinous fees.

Come up with another solution and then watch me tear it to shreds.

Scenario when a monopoly forms in a proof-of-stake network:

• Large mining pools merge or a monster miner emerges.
• The possibility of potential terrorist organizations take control of Bitcoin
• Enemies of the monopolist are targetted for reversals.
• Double-spend attacks are attempted.
• Stakes have no geographical location and can use rented servers, they cannot be shut down by any authority.

Solution A:Since the government regulators would be managing an unknown group that can be communicating covertly. This would be very difficult and expensive. It would require a central governmental clearinghouse of approved addresses for any business accepting Bitcoin, identification, and transaction approval of all commerce to prevent terrorists from controlling the network. Bitcoin is more likely to be outlawed. Either way Bitcoin fails.
Solution B: Fork the blockchain. Bitcoin fails.

Scenario when a monopoly forms in a proof-of-work network:

• Large mining pools merge or a monster miner emerges.
• The possibility of potential terrorist organizations take control of Bitcoin
• Enemies of the monopolist are targetted for reversals.
• Double-spend attacks are attempted.

Solution: Find allies to bring reserve mining power online before this attack has 6 confirmations and reject those attacks. Government miner stockpiles are brought online and thwart the attack. These stockpiles cost very little to maintain because they don't require electricity until they go online. These can even be patriot volunteer organizations. The attackers are identified by their IP addresses and shut down. Bitcoin survives.

These are simply two different philosophies about how to operate a cryptocurrency. Trusting in a monopoly is hardly a disrupting enterprise. While monopolies can be efficient, historically they require government regulation to prevent corruption. In a PoS model, the government regulators would be managing an unknown group that can be communicating covertly. This would be very difficult and expensive.

Lol! What?

Can someone translate for me? I don't speak idiot.

IT MEANS YOU GOT NOTHING! LOL WTFSTFU
or maybe you are a terrorist.
or maybe you want Bitcoin to fail.
or maybe English is not your first language.

Haven't you figured it out yet? The bolded item is the correct answer.

I just don't understand why he won't admit that the idea of PoS is a PoS. If he really cared about Bitcoin, he would throw out the bad parts, keep the good ones, and offer a better idea. I actually think he is sort of on the right track, but instead of using an acquired stake (which can be stolen), use a proof that you earned those Bitcoin by processing transactions, and that fact weighs even more than hashrate or the stake itself. I just don't know how you would do that.

Because it's not. Getting a majority of the bitcoins is much harder than getting a majority of the hashrate, meaning that increasing the reliance on the former improves security.

Also, it looks like the PoS detractors haven't taken the time to study how the proposals for PoS actually work. In particular the proposals aim for combining PoS and PoW.