Bitcoin Forum

So basically there's some people that theorize about what will happen in 2140+ (to put the extreme scenario in which no further Bitcoin will be generated, but this should be hit in practice way sooner due very low inflation rate) when no longer there will be more BTC issued, so then the question remains to be answered:
How will miners find an incentive to mine blocks if there is no block reward left?

Some people claim that the transaction fees will be enough. Others say that the price will be so high that there will be not enough transaction volume for this to compensate miner activity (then again wouldn't the high fees be enough incentive?).

Well what if the fees are not enough, or there are not enough volume tx going on, or anything else? Would non-monetary use cases like Ordinal inscriptions solve the problem? I imagine in the future blocks will contain massive amounts of second layer transactions and it will be a settlement layer 1 network only. The fees will be very high, so maybe between regular transactions + other use cases it will be enough to keep miners working?

I have been trying to decide if ordinals and basically anything but monetary transactions are a massive amount of time and resources, or perhaps they would actually help with the miner incentive structure?

Obviously, the top priority should always be monetary transactions, then if that wasn't enough, the incentive could be filled with non-monetary transactions. Perhaps at the protocol level, make it so that monetary transactions go first, then with less priority process non-monetary transactions. We should consider if ordinals can be saved or not long term, if these use cases could deliver something useful. If they will only bring problems, then perhaps a full frontal attack on them and planning to stop them may be best. But im still open to discuss this before I pick a side.

Look at testnet3, and try to use your CPU to mine something there. Then, you will find your answer.

They are enough on testnet3. Why wouldn't the same situation repeat on mainnet? Those networks are very similar, just 20 minute rule is one of the main difference between them, other things are pretty much the same (which is also the reason, why test coins got some value, even though they shouldn't, and why they often behave better than some altcoins).

You cannot take all of your coins to the grave. Sooner or later, people would need to spend their coins, because if their plan is to never move them, then sending all of that to OP_RETURN will achieve the same outcome.

Then, people will abuse 21 million coins limit even further than today, and that would "solve" the problem. Because note that even though we have altcoins, they are not all pegged into existing coins. They usually create additional supply, which abuses the system. If you have 21 million BTC, but if BTC has only 50% domination, then the situation is the same, as it would be with 100% BTC domination, and 42 million coins.

No, they would create the problem instead, for example by raising fees sooner, than they should, and taking down regular payments. It is not only about the incentive they bring to the miners. It is also about the incentive they removed, by discouraging people to transact, when they saw some mempool stats, and switched into other payment methods.

Assuming bitcoin is still used in 2140 and there is no mining rewards your question is whether miners can sustain only with the fees from transactions right?

It should be or miners will stop then the difficulty reduced but still the blockchain will just go on, and ordinals are non-monetary is a 'subjective' word because there is no difference between ordinals or an actual transaction, it's someone willing to spend high fees due to hype up the network but you can look at the mempool (https://mempool.space/) they are not capable of doing that forever cause they are simply burning their money for some shit which just gained attention in short term and in my opinion it won't even be there for another next 5 years.

Then mining difficulty will drop, until remaining miner can mine at profit with lower difficulty.


No, unless block usually isn't full.


With current way of Bitcoin protocol, miner can choose any transaction they want to include in a block. In practice, miner usually include as much as transaction as they can starting from transaction with highest fee rate. Forcing miner to prioritize certain transaction would be difficult for various reason, such as each node have slightly different mempool.

Well, first let's see the current blocks (post-2024 halving), and see if we already can draw some conclusions.

If we take out the fee spike near the halving and the OKX spike, then most of the time the fees are hovering around 10% of the total block reward in both waves when Runes cluttered the network (including subsidy and fees), and 5% in low-fee phases (when no such event occurred).

https://talkimg.com/images/2024/06/13/ccyE1.png
Source: bitbo (https://charts.bitbo.io/fees-percent-of-reward/) (blockchain.info charts currently don't work for me)

That means that if block subsidy was cut off completely, then the hashrate would probably drop by 90 to 95%, from currently around 600 Eh/s to perhaps 50-70 Eh/s. You may say that's dramatic, but that is approximately the hashrate Bitcoin had in 2018/19 (see this chart (https://www.coinwarz.com/mining/bitcoin/hashrate-chart)), and Bitcoin was considered safe at this time too.

This would be basically the situation in 2140 if the Bitcoin price was the same than today (65-70k) and the transaction activity was the same too.

@vjudeu's comments about testnet would indicate something very similar: even on testnet there's a significant hashrate even if there's no real block reward.

However I think there is perhaps a problem: mining difficulty could become more volatile, because in bear markets usually transaction rate drops and thus the fees drop too. What I think will happen is however what already today happens when there's an upcoming halving: miners diversify their operations, for example offering services to AI and cloud companies, and server colocation.So miners would find a way to survive - and take into account this is a very gradual process. Hashrate would probably however then get lower in bear markets. But the hope is of course that Bitcoin's volatility will be much lower at that time, as it will be an established financial asset.

I believe in the future merged-mined sidechain rewards could become a significant income for miners. That's also what some people like Paul Sztorc (who proposed Drivechain, an interesting sidechain mechanism) wrote.

As for tokens like Ordinals/Runes I think other blockchains will be more attractive. Take into account that Ordinals/Runes were/are mostly a fad phenomenon, at least on Bitcoin - they came from the idea to "have an NFT on the OG blockchain". It is very unlikely that this fad will continue for more than 100 years ...

This whole question is wrong because you are only focusing on a tiny insignificant thing (fees) while there are a million different things that are much more important. For starters in 100 years from now, Bitcoin may not even exist. The cryptography used in Bitcoin is going to will have been obsolete for years by 2140, so by then Bitcoin either will have had significant changes (hard forks) that would fundamentally change it or a more probably thing is alternatives will have been invented that replace Bitcoin.

As for fees, people don't tend to use a payment system that is expensive. If that becomes regular in Bitcoin, it would kill it. That means it stops being used by regular people for payment and it will lose its "hype" and subsequently there won't be any reason for scam attacks such as Ordinals to be using the "bitcoin name".

Disagree. The fee structure is a fundamental pillar on Bitcoin. Without that incentive there is no one to process blocks which contain transactions. As far as the cryptography being compromised, that is a problem that everyone will face eventually. And yes, that may be the first time a hard fork has 100% agreement, I mean that is a clear Schelling point in benefit of all Bitcoin holders.

As far as high fees, people will just use 2nd layers. Unless some impossible to imagine alternative is invented that is so good that people want to migrate their money there, Bitcoin could perfectly exist in 100 years. This is a software, it's a digital asset, it's not some physical thing that has to be replaced. It's interesting to think about the far future and see what will happen. I believe BTC is here to stay unless a major discovery is made that like I said, invents something that we cannot even imagine now.

You're talking about 2140 but I would be worried much sooner, unless you believe the market cap will double after each halving.

Two more halvings and the security of Bitcoin could be in great danger due to economical factors and before this happens we will see higher centralization due to less companies mining it.
The tx fees will not be enough and block reward will be too small, if you can't see it - you're blind.

There are only two solutions to this problem:

.1 Switching to tail-emission and forgetting about "there will only be 21m" mantra.
.2 Switching to PoS (which will make it even more centralized)

Tail emission quick summary:
https://www.youtube.com/watch?v=sRwSqM0YBto (https://www.youtube.com/watch?v=sRwSqM0YBto)

Another problem is the Bitcoin project competition, mining companies don't care about the project and securing it, they only care about profits.
So why would they buy another batch of power hungry ASIC's if they can buy a lot of general purpose CPU's and mine with them more profitably ?
Add the fact that you have more resale value of this CPU's, better warranty, etc.

Numbers will speak and if Bitcoin price craze stops, this companies will reevaluate their mining business and alternatives.

If they see they can have double the profit mining with CPU's on a stable enough project... they will switch in a blink of an eye.
They could as well trade to BTC if they want, but the matter is - security of Bitcoin will be in danger.

Some hints about current Bitcoin mining profitability (or rather lack of it):
https://www.youtube.com/watch?v=QCFwmJXx7dc (https://www.youtube.com/watch?v=QCFwmJXx7dc)

It will only get worse as the electricity costs rises and more mining taxation come.
If nothing changes, there will never be 21m - So the mantra will die anyway along with the project.


If you're gonna use second layer solutions, you can as well use PayPal.

If there will be a decentralized way to handle much more transactions per second than today, it can as well be incorporated into first layer.
Centralized "solutions" are not solutions and LN has failed in this regard.

It's a shame "hard fork" term was even invented, because before that we just called it "Update" and all software needs updates.
Due to this protocol code stagnation, right now I don't see bright future for it.

Maybe someday it will change but it could be too late.

PS. I know I sound very pessimistic but it's safer to be incorrectly pessimistic than incorrectly optimistic. 

I think it will happen anyway, probably as an altcoin. And then, in the future, those altcoin users could even burn some overprinted coins from tail supply, to raise the price.

Again, another good candidate for some altcoin. I wonder, if some people coming from ETH want to deploy some stuff on BTC, because they don't like PoS. Before Bitcoin, we already had Proof of Stake. It is called fiat, and existed for years. And Bitcoin raised partially from that, after what happened in 2008. So, I wonder, if people will go back, when Proof of Stake will collapse again, and the cycle would repeat again?

Because if you have an ASIC, then your blocks are stronger. In BTC, there are many places, where one, two, or something up to six confirmations is sufficient. But if you have something, that can be mined on CPU, then it requires much more confirmations. For testnet3, people require for example at least 100 confirmations. And for many altcoins, it is also the case: You have less power? Then you need more confirmations.

Second layers are there not only to replace the main solution, but also to bring more features. For example: if you have LN or sidechains, then you can remove data, if they are deeply confirmed on the first layer. If you have only one layer forever, then you can never delete your data. So, do you want to live in 2140, and still download all historical data, starting from 2009? Or maybe you want to say, for example "data below 2100 are deeply confirmed, are grandfathered in, and should never be reorged again". We already had some upgrades, which have exceptions in the source code, because it was easier to add them (for example duplicated transaction hashes), than to re-mine the whole chain, to reorg hundreds of blocks, and to invalidate previously valid coinbase transactions, which were already spent.

So, to sum up: my advice is simple: you want tail supply? You want Proof of Stake? Go on, do it. Test it, observe it, make your own conclusions. And remember: nodes can leave and join the network at will, so you can always return, if you conclude, that your ideas were wrong. After all, even coding a bad idea is something, which can be used to improve your skills, and get some new experience.

This is not true, more confirmations on other chains come from different problems and you can have 0-conf transactions when it comes to some altcoins, I have paid numerous times with 0-conf.

Mining hardware has nothing to do with how many confirmations are needed and ASICs certainly does not make it stronger (rather weaker due to more centralization).

One project that I use has 10 confirmations needed due to privacy reasons and more hashrate/ASIC's wouldn't solve it.

Bitcoin requires less confirmations - 6 but time between blocks is 5x higher (10 min) vs 2min and 10 confirmations of the competition project.
So we have 60min vs 20min on non-ASIC chain when it comes to confirmations.

If the fees are not enough, it will be because there are not enough transactions to compete for the block space. In that case, the system will stop operating:

This will have the following consequence: Ordinals will switch to another protocol, with transaction structure that is not biased by the Bitcoin protocol.

We should never censor or discourage any valid transaction that does not fit our description of Bitcoin, no matter what. The only objective factor that defines which transaction gets into the next block is the law of demand and supply.

Where is, in your opinion, the limit between "secure" and "not secure"?

As I wrote in my last post, the hashrate would probably go back to 2018/19 levels (90-95% less than now) if the complete block reward was cut and the BTC price was like today. It will be a bit higher due to Moore's Law and even could be higher than today for this reason but as hashrate is not the relevant factor for security but the "attack cost", the security will definitely be much lower than now (in 8 years Moore's Law based hashrate increase should be * 4 approx. without increased attack cost).

But: Is that not secure enough?

Just a small comparison:

Bitcoin today has a block subsidy of roughly 200000 USD (65000$ * 3.125 BTC). Transaction fee income is about 10000-20000 USD per block. A block is generated approximately all 10 minutes.

Litecoin has a block subsidy of roughly 500 USD (6.25 LTC * 80$), 4 blocks per 10 minutes, so the miner income is about $2000 per 10 minutes.

So LTC has only 1% of the current Bitcoin subsidy, and in addition there are no relevant transaction fee rewards. The LTC reward is also much lower than the BTC transaction fee income. But it is considered safe enough for many exchanges to accept deposits of less than 24 confirmations (24 confs are equivalent to 6 Bitcoin confirmations, i.e. "one hour of confirmations"). And it is also safe enough that financial companies are basing regulated products on it (e.g. the CoinShares/ETC Group ETPs).

Litecoin can also serve as a model to see what could happen if the long term price increase is not pronounced enough to balance the subsidy decrease due to halvings. Miners income is thus now less than some years ago.

So in general I'm a bit more optimistic. I think Bitcoin is now much more secure that it needs to be, and after several halvings more, it will be likely that it is still secure enough - probably even in 2140.

We can only speculate for the near future not 100 years from now. And we already have the history to work with.
For example 10 years ago miners were being paid 25BTC per block with nearly no fees but that was a $5,000 reward.
Today they are being paid 3.125BTC and even without any additional fees that is like a $200,000 reward.

That's been more than enough incentive.
In the near future (like next 10 years) people all agree that price rise will continue. So there is no valid arguments left there for "incentive for miners" either.

I disagree.
First of all for a secondary layer to function well and give people incentive to use it, the primary layer that this 2nd layer depends on has to work smoothly. People aren't going to be able to open channels and settle them easily if the fees are outrageously high.
Secondly if you checked the LN stats over the past year, there has been drops in usage (number of open channels, capacity, etc.) each time that the Ordinals Attack was at its peak and had caused a huge fee rise. That proves my first statement.

This is true and a bit scary, because one of their goal with this is to destroy medium players (small are already dead).
Higher hashrate does not always mean higher security, as security comes mainly from decentralization.
If only few players control majority of the hashrate, I would say it's not secure at all and the project is at the mercy of the regulators.

Remember China ban on mining when in 2021.05.13 65.5% of hashrate disappeared in little over a month ?
Another threat is the inevitable black listing of transactions forced by government on miners, as if the tainting wasn't enough.

The security of the Bitcoin is in hands of the few companies, if profitability drops or they find something more profitable to mine then Bitcoin network will be doomed.

After each halving the centralization is growing as more mining companies get out of it as reported by several sources.

So I'm still very pessimistic when it comes to Bitcoin security in the upcoming years and if nothing changes then I'm 100% sure that this project won't last more than a decade.

I wouldn't compare Bitcoin to much lesser projects when it comes to security as there is much more at stake and someone will be willing to attack it with higher budget like state actors from North Korea.


When the balance between decentralization and hashrate is broken, we have seen a huge increase in hashrate but we can't say this about decentralization.
In fact, decentralization has been dropping for quite some time because smaller players can't compete in this model.

I just don't believe a trillion+ market cap will magically double after each halving, and that's what is needed for this to work.
It's unsustainable.


And yet, profits back then were much higher due to much lower difficulty.
https://bitinfocharts.com/comparison/bitcoin-mining_profitability.html#alltime (https://bitinfocharts.com/comparison/bitcoin-mining_profitability.html#alltime)

Hint: Profitability - It's not going up.
And there's nothing wrong with it until it hits the wall of no profit.

1. It is not all about mining: https://en.bitcoin.it/wiki/Bitcoin_is_not_ruled_by_miners
2. Big server farms:

And note one thing: most users will not mine in the future. They will be "client nodes".

3. If you have 51% in any Proof of Work system, then you no longer compete with other miners. You compete mainly with yourself. Going above 51% simply means, that you produce more Proof of Work than needed, and you make your own blocks harder to reverse, than anyone else's blocks. Which also means, that not only it is costly to compete with you. It is costly to also reorg your own chain, which you mined.

Some example: you have regtest-like network, but with difficulty adjustments. Is it profitable to run some ASIC here, and have blocks with 64 leading zero bits? Obviously not. Reorging that would be costly not only for your competitors, but also for you. Which means, that if you are a regulator, and you find out in the future, that "hey, this decision was wrong", then you would need to reorg your own, very strong blocks. Because if you don't, then the truth will be written in the chain, and shared with every non-mining node.

The article talks a lot about breaking consensus by miners and that it can't happen but I'm talking about regulators forcing miners to use black-listing (not including certain tx) in a block, which doesn't break the consensus.

Here's some more info on this subject:
https://thebitcoinmanual.com/articles/bitcoins-battle-with-ofac-compliance/ (https://thebitcoinmanual.com/articles/bitcoins-battle-with-ofac-compliance/)


Here are interesting stats about current regulators impact on Ethereum:
https://www.mevwatch.info/ (https://www.mevwatch.info/)

Suppose regulators force the miners of their country to blacklist certain transactions. What happens with miners outside their borders? They cannot force their behavior. The anti-censorship miners will normally mine any transaction, and the pro-censorship miners have the choice to either accept the chain with the most work (which would make their blacklisting pointless and financially damaging), or blacklist the most worked chain if it contains blacklisted transactions.

Since the former is not an option, good luck with convincing the miners to attack the very essence of Bitcoin.

Marathon is already blacklisting transactions and it didn't split the chain.

The tx will be mined by different company but the problem here is that it's already happening and other miners could be incentivized if not forced, to join this OFAC compliance.

I don't remember how this incentive would work, there's somewhere article on it - if I find it, I link it.

Even if just 1% of the hashrate remains anti-censorship, it has a high chance to produce 1 block everyday. That's enough to clear the mempool from all blacklisted transactions every day.

The only way to censor Bitcoin is to attack it, which incidentally will affect your own income either directly through waste of electricity or indirectly through capital devaluation.

*anti-censorship, makes sense...

I disagree with this logic.

First, if "they" (I guess you mean the big mining companies like Riot, ViaBTC etc.?) find something more profitable, then difficulty will go down, and it will become easier for smaller players to enter mining again. Decentralization in this case will probably grow, although more slowly than in the past due to higher entry barriers.

Second, this will be a gradual process. While halvings are unique events, the consequences are different for each miner, depending on its exact business model (e.g. if it relies more on cheap electricity or on other scale effect, on speculation [HODLing] or not, etc.). So the most likely scenario is that if some big miner retires from the business or goes bankrupt this would be first good for the other mining companies as the difficulty pressure will be lower (hashrate will probably grow more slowly), and will help them to stay in the business.

There is only one factor I could be a little bit more afraid of - if a country becomes too dominant, then an event (be it a new tax, a ban or a sudden rentability problem) driving out miners from there could drop hashrate again like during the China ban. However, the US has currently probably around 35-40% of the worldwide hashrate, so in the current situation it would be less of a problem than in 2021, and 60% of the current hashrate would be still "secure enough". But such events could be problematic in the far future when the attack cost has lowered. So I share a bit your concerns about decentralization problems.

But I anyway expect a change in miner behaviour in the next 5-10 years. As electricity cost is about 70% of the cost of each Bitcoin mined (for the miners) and renewable energy share is growing around the world, I expect to switch them slowly from a 24/7 mining model to a model where they use a wholesale energy tariff (these are becoming more common now) to benefit from very low prices in times of high solar/wind production and thus at least turn off their gears voluntarily when energy is expensive, mining e.g. only 70-80% of the time but to a even lower price than now (< 5 ct/kWh is easy with this strategy, even < 3 ct is feasible. i.e. better than the famous Riot agreement of 3.5 ct). This model is much more likely to be adaptable all around the world (the "duck curve (https://en.wikipedia.org/wiki/Duck_curve)" is starting to look the same in several countries), so mining could decentralize again due to this evolution. This is controversial and has been disputed by some forum members, but they haven't convinced me that my thoughts are wrong.

There is also an additional point. Bitcoin mining companies are basically server farms. In periods of low profitability, they can diversify their income sources offering servers for other cloud services. This is already been done by most big mining companies, and it is also one of my reasons why I think that if profitability drops there will not be a "big hashrate crunch", but a very gradual process, because bigger miners will slowly dedicate more percentage of their farms to other cloud services once their hardware becomes unprofitable, instead of going offline completely. And that will re-balance with difficulty growth again and again.

IIRC, what you're suggesting is already a reality in Texas.

Miners there don't work 24/7/365, but only when the state has a surplus of energy.

When there's high demand for energy (during heat waves or harsh winters), they switch off ASICs.

Why should we trust them? Are they the sole arbiters of truth or what? ::)

I've been saying for a long time that BTC + renewable energy + AI (to manage the energy distribution grid faster and more effectively than human beings) will be the ultimate combo until 2050 at least.

How it started:

https://www.weforum.org/agenda/2017/12/bitcoin-consume-more-power-than-world-2020/

How it's going:

https://www.instagram.com/worldeconomicforum/reel/C35FfRytPtJ/

 ;D

Net Zero/Greta fanatics still cannot believe that Bitcoin/PoW is more ecological than they thought! :D

Imagine if BTC was centrally planned (like ETH) and switched to PoS:

https://cleanupbitcoin.com/

Bitcoin will help speed up the green energy transition. Few people understand this (even among the BTC community, let alone no-coiners). ;)

ps: I'm not saying Satoshi invented it for this purpose, nor that WEF had this planned from the get-go.

The hashrate is constantly rising and it has risen massively. That means mining has to be increasingly more profitable to create that incentive for people to join the mining scene by making that commitment and investment every year.

Yes, it's beginning, but I think it's still a 95-99%-on model. AFAIK those miners only turn off the gears in periods of extreme demand surplus, when they're "ordered" to turn off operations by the electricity distributor, as a condition to benefit from energy credits (Riot case).

Anyway, I think soon we'll start to see miners doing the same but voluntarily chosing their on-/off-cycle due to them using the wholesale price (i.e. participating in auctions in the energy exchanges themselves, or with access to a tariff which follows the wholesale price - something already available in several European countries even for retail energy consumers). By the way this method would even stay relatively attractive if the "DAME" mining tax is implemented in the US, as the prices they would pay when they effectively consume would be low, and thus also the tax (Anyway I don't like the tax concept, above all own renewable generation should be always excepted.).

Of course, that's also why I will repeat this argument until somebody proves I'm wrong - but then I've no problem to admit an own error. But I think evidence is clearly pointing into this direction :)

I think you can't deduce from the risen hashrate that mining has become more profitable. Only that it stays profitable, but currently only for those who can afford efficient ASICs and cheap electricity. The hashrate increase is very likely largely a product of hardware improvements (Moore's law).

However I agree with you that MeGold666's fears are probably unfounded - if profitability was that bad we would see a slower increase.

That's a good point but that's also only one of the things affecting the hashrate rise.
Keep in mind hardware improvement doesn't happen every day. For example Bitmain released S19 miner in 2020 but then the next major improvement which is S21 released 4 years later in 2024.

BTW I don't disagree that block reward falling can become a concern regarding the incentive for hashrate to continue rising. But I don't see us there yet. I believe we are far from that time, at least a couple of more halvings are needed for it to become a real concern.
Also we definitely don't need spam attacks to increase the fee to provide that "unhealthy" incentive.

Even if this is of course correct, the "hardware upgrading process" for the miners should be in general a gradual process. Not all farms will upgrade their gears at the same time, not even in the same year. And that's in my opinion the explanation that the hashrate increase looks "gradual".

Let's see how much can be explained by Moore's law with a short calculation:

- Antminer A19 (2020): 95-110 TH/s
- Antminer A21 (2024): 200 TH/s

So we have roughly a duplication in 4 years. The hashrate however quadrupled in this time (from 120 to 550 Eh/s), so Moore's Law explains about half of the hashrate growth probably.

Of course however the other big factor is the Bitcoin price increase, but the impact on miner profitability is not 1:1 as you always have to take into account the previous difficulty increases as well (that's the reason why some talk about "production costs" of Bitcoins, even if I strongly disagree calling it that way ...). And it's a bit difficult to calculate the impact due to the cyclic price behaviour. In early 2021 the price was still much lower than today, in mid-2021 and late-2021 the price spiked to similar values than now, but with a big "valley" of 35-50k in mid-2021, and in 2022 it fell drastically again.

Fully agree here :)

This assumes the remaining 99% of hashrate won't reorg the block created by the anti-censorship pool.

That's exactly what I mean by "attack".

I don't understand why in the name of god it's 2024 and we still cling to things like hashrate is going up, everything is fine the networks is more secure, and we ignore when it comes to security basic things about mining, things that guys like Burtw mentioned decade ago!!!
Quick example of why hashrate on its own is just a useless metric, so, assuming attackers need 10 billion (number pulled out of my *) now at 600exa that means back in 2013 at 400th/s you could have attacked the network with $10k? No!

So why not go the easy path of doing things, best case scenario:
- all miners are ar 3cents kwh, we completely ignore ROI, last day read was $27,476,575.85, let's round it up to 30 million, so that's 1 full TWH a day, your best efficient gear at the moment draws 85kwh so we can feed with this 11 million ASICs worth 50 billion! Nice one!
Bad scenario:
- we no longer have rewards, but Bitcoiners are spending like nuts so the fee is not 1 mil but 5 mils, all miners have 4 cents per kwh, but they also have a ROI of 2 years on the gear so we're down to 125Gwh, before ROI so with that we need additional 8$ per day per gear , so we're down to 400 000 asics being the maximum of economically viable, and that's 2 billion! Still out of my pocket reach, we're safe!
Worse case scenario:
- fees are the same as now 1mil, kwh is at 5, adding expenses keeping 2 years ROI, and we can't feed more than  70 000 asics worth 350 million.

350 million to attack a destabilize a network worth 1.4 trillion, cheap, expensive?
The good part in this is, why would Buttercup do this? Or Kim Join Us! Or the reptilians and the bildenbergs!  ;)

Oh my bad, forgot in the worst-case scenario the attacker might not want to buy the most expensive gear but a cheaper one in cost of TH/s/$, so in head to head battle you won't need $5000 of gear to overpower a s21 you would do it with s19 that are selling for almost scarp on the market for $1800.
Which would bring down our scenario from 350 million to 126 million  :D

But why would anyone who feels threatened by Bitcoin spend twice the wage of the CEO of JP Morgan on killing it.... ;D

Peter Todd has actually been entertaining that idea of adding a tail-emission or extending Bitcoin's inflationary period indefinitely. But it's going to be very hard to convince the community, and many people believe that the network has ossified to the 21,000,000 total coin supply. But perhaps if the situation will be life or death for Bitcoin, the community might start to entertain a discussion of a tail-emission.


POS doesn't work.

https://youtu.be/NYPGub6XzjE?si=YnAFxEvDmKdsudAx

If the diff drops, so will security - small miners will join but the security will never be as high because this small miners will never be as profitable as large scale miners and the network will be at constant threat of state sponsored 51% attack from North Korea or some other sources because it will not be enough to secure a trillion+ dollar network.

Which means the mining will be even more centralized than today, more power in the hands of few.
This is not good for many reasons, one of which is Government control - possible forcing of OFAC rules on mining company or straight out ban like in China.
In this situation network will again have low security.

Bitcoin mining is done on ASIC's, not on general purpose CPU.
No one in the world needs this power aside from Bitcoin and other, less profitable coins running on SHA-256.


It would split the community for sure, the discussion should be done now as this tech should be running on Bitcoin years ago.
It will be too late for any action when problems arise.


Here's an interesting info on the subject:
https://www.youtube.com/watch?v=sw6aMxaNmXA&t=416s (https://www.youtube.com/watch?v=sw6aMxaNmXA&t=416s)

I suspect that hashrate was increasing only because mining companies wanted to eliminate competition.

Currently we can observe a ~30% drop in network hashrate:
https://bitinfocharts.com/comparison/bitcoin-hashrate.html#alltime (https://bitinfocharts.com/comparison/bitcoin-hashrate.html#alltime)

https://i.postimg.cc/JzCnCgCN/Untitled.jpg

I was warning in my other posts that it's a matter of time until this big companies reevaluate their mining business and switch to more profitable projects, and it's already happening:
Source: https://cryptonews.com/news/bitcoin-miner-profitability-halving.htm (https://cryptonews.com/news/bitcoin-miner-profitability-halving.htm)

Remember: There's no love in this industry, there's only $$$ and they have great financial advisors.


Even fewer people understand that green energy is not green and it also produces waste.
Solar panels have a life span, batteries have a life span, electronics that run it have a life span, and nothing of it is degradable or recyclable.
Other "green" solutions are even worse...

The greenest solution is nuclear power plant, snowflakes will disagree.

Here is some information to all the glorifying disinformation propaganda circling around "green" energy when it comes to solar:
https://hbr.org/2021/06/the-dark-side-of-solar-power (https://hbr.org/2021/06/the-dark-side-of-solar-power)

The whole green industry is a scam to earn $$$ for companies who produce this waste.

We can argue all week about it but I let the time show it's madness.
I'm out from the forum for a longer while as I have some work to do, see you when the roof starts collapsing (if nothing changes to the Bitcoin protocol).

It's interesting that you just mention North Korea, weren't that the guys making billions with Bitcoin hacks and ransomware? ;) Why should they destroy their source of income?

Anyway: again, where is the limit between "secure" and "insecure"? Was 2019 hashrate secure enough? I believe a state actor could possibly currently carry out a short 51% attack if he wastes a lot of taxpayers money, but that wouldn't be disruptive enough to destroy Bitcoin, more like what happened to some altcoins. Even dictatorships need to justify their actions at least a bit to avoid becoming unstable, and it would be really hard to justify such a waste of money.

I think we agree that a centralized mining sector would not be good, but I just explained why I don't believe that your logic here is correct, just a paragraph before. Yes, the difficulty lowers in this process, and so does security. But it is still safe enough.

That is not the point. The point is that the housing/cooling infrastructure can be used for CPU servers as well. And that is already been done for some years. Many US miners are already diversified.


No, that's only 30% if you take the highest single value and the lowest single value. The drop is about 15% and that's expected because it occurred in each halving until now (see 2020, 2016 ...). It will grow later this year probably.

Neither the waste nor the energy needed to produce solar panels, wind turbines, hydro facilities etc. compensate the energy they produce.

Those that really understand about the subject know that there is some waste but that this is not really a problem. There is no free lunch (still). But recycling tech is advancing too, and household waste is still a far bigger problem than silicon which can be used for landfills.

It's funny you mention nuclear energy as being better because ... well, didn't they produce waste?

Ah ok. :P

Is your mission currently failing? ;)

By the way I'm not against a tail emission, the Monero folks are doing well with it and I think it's a reasonable model, but in the case of Bitcoin I don't think it will ever materialize due to the "ossified" 21 million limit, and I also don't think it's needed.

No, it didn't!  ;D
The halving in 2016 was in July, we had just a 5% drop in August.
The May halving difficulty history looks like this:
so a bit harder to find a pattern there, oh, the adjustment wasn't on the -15.95 it was on the -6  ;D

I seriously don't understand on what you base your recovery enthusiasms, other than maybe being completely bullish on the price, miners don't care about charts, we care about income per $ spent on electricity, and this I how it looks now:
The best of the gear launched now and available makes you 11$, burns 84kwh and costs 6k,2 years ROI on zero costs.

I was pretty bullish on the growth myself, but the end of Q2 is coming and a lot of contracts will probably expire, only the big guy with dirt-cheap energy will keep on mining.

Also, in terms of network security it's the same in both cases:
300 exahash when the bets gear was the 19 with 100th/3kwh and $5000
600 exahash when the best gear is the s21 with 200th/3kw and priced at $5000

Hashrate without the cost per th/s and the amount protected is a bit meaningless!

While you're correct that in 2016 the pattern was weaker (a >10% drop was present but only for a short time, about 2-3 weeks, so you could say it did not leave the normal fluctuation range), in 2020 it was in the same order of magnitude than now:

https://talkimg.com/images/2024/04/11/juMo2.png
Source: Coinwarz (https://www.coinwarz.com/mining/bitcoin/hashrate-chart)

There was of course before the March "hashrate crash", so the halving drop looks like "one drop more", but this one was clearly caused by the price crash due to Covid, leading probably to a short panic in the mining sector too, because if the price had stayed at $5000 or less with the incoming halving the drop would probably have been even harsher. What I can't explain at this moment is the sharp November drop (this was ATH time in '20 ...).

btw: Is your quote related to the 2020 hashrate evolution? If you mean the 15% drop was earlier (or later?), there always can be some delays, in either direction.

My recovery expectation is not based that much on price expectations but mainly on gradual continuous hardware improvements plus rising longer-term average price, see my answers to @pooya87 above.

With the rest of your post I more or less agree. It is possible that recovery takes longer if for example Riot runs into difficulties. But eventually a new equilibrium will be found. Let's see how it plays out. At least I don't think the hashrate drop caused by this halving will be as large as the one in 2021 caused by the China ban.

A new trend in miners (at least PubCO companies that im tracking) they are moving to HPC services, mostly for companies that use AI. Core Scientific just got a 12 year contract with CoreWeave, so that's a solid source of income. IREN is also planning to announce some clients and they are building infraestructure with HPC on mind. WULF as well. There are others like CleanSpark and Riot that choose to remain pure Bitcoin miners. Im ok with diversification of MW if they can leverage this properly in order to continue to remain profitable even during long bear markets, so when the bull Bitcoin market is back they will have more power ready to mine. We'll see what model turns out the best.

-5, not >10!



I only count difficulty adjustments, daily hashrate is just all over the place, completely unreliable!


End of China's rainy season!
Short story, cheap hydro energy during the rainy season, end of October-November, gear gets shipped back to inner Mongolia to feed on coal, it was an annual occurrence!



Equilibrium could be found even at 5 exa and price per BTC at $600, but I have a feeling in terms of security nobody would be happy about it!

@stompix: Ok, thanks for clarifications. The >10 was referred to the approximate 1-week average of daily hashrate, not diff adjustments.

Difficulty adjustments is indeed a better metric than daily hashrate. But difficulty until now fell only 5.1% since the maximum in April according to Coinwarz (https://www.coinwarz.com/mining/bitcoin/difficulty-chart) (88.1 T to 83.7 T). I guess it will fall a bit lower in the next adjustment but I'm still not convinced it will get lower than -15 or -20% from the pre-halving maximum.

Again, let's see :)

What I however don't understand is why you think that this time it's different to previous halvings and recovery will not occur or be slower than in 2016 and 2020.

If we look at the 2024 halving from a reward perspective, it was less devastating for miner income than 2016 and 2020, because the reward is already so low that transaction fees are increasing their share, so it's no more a 48-50% income reduction but more close to 40-45%.

But let's explore the possibilities a bit. One argument could be the stagnant price evolution since April, but a strong counter-argument is just 2016, when the post-halving price evolution was stagnant for a long time, basically until November/December, but hashrate/difficulty recovery was very fast. In 2020 we also had much more bearish conditions than now in 2024, if we take into account that the price in 2019 had reached almost $15.000 and at halving date was around $10.000, and also there was price stagnation for some months (and the real bullrun only began in November).

Another argument could be that the current hashrate is "inflated" due to publicly listed mining companies like Riot who diluted their shares but are probably mining on a loss actually, i.e. operating close to a "ponzi" but preserving a high hashrate. This would be more convincing for me, as I already wrote in the last post. But does that apply to all bigger miners? I have some doubts here.

Edit: If I understand your posts well then your argument is that "attack cost" variable is going to lower with each halving, as hashrate could be going up but the attack cost not. I sorta agree here but don't think it will come into dangerous territory as long as we have significant transaction fees.

Yes, miner diversification is definitely an interesting evolution. It could lead to more income stability for mining companies but the hashrate could stay relatively volatile, because it's easier to simply change hardware than to set up or dismantle entire mining farms.

Income! That's the problem!
I mined at 45 cents/th back in 2021, I could take a 66% drop because I would still be profitable even at 15 cents/th.
If I would still have mined at 10 cents/th (before halving) and got even a 33% I would be losing money!

Income at 45 cents and expenses at 8 cents/ths means a cut of 50% I still make 18cents/s profit!
Income at 10 cents and expenses at 8 cents means that a 50% I'm in red by 3 cents!


A lot of different things, profitability was still higher by 40%.
The s19 was shipping in May after a shitton of preorders because it was outperforming the s17 from 0.045j/Gh to 0.034j/Gh.
The summer had cheap energy from China, which is not here anymore see the power curtailing in the US now .
ROI at 0 costs for the s19 was still just over a year, not two and a half years for the s21 as now.


Well Riot has never made money, Stronghold is for sale, Core is taking a $75m loan not to enter bankruptcy again, Mara is sitting on 600 million of debt and "investors" money.....


Exactly, each halving when the price doesn't double means less security overall, and even if does double it still means 1/2 security to market cap.
As for the fees, they are 4% of the reward, if you expect people to pay 30 million for 400k tx day, well, how can I say it... don't!  ;D

Thank you for this post, I now understand your arguments better.

Basically so your point is that in the last halvings the profitability was much higher than now and miners had margin to "manage" the halvings with minor adjustments to the business model, While now the margins for miners are much thinner, so with every income problem (like halvings or price crashes) there's danger to lose larger parts of the hashrate without immediate recovery. So it's basically what I meant with the "inflated" hashrate.

I still don't see this as critical for Bitcoin's wellbeing / security, at least not in the short/mid term.

If the assumption is true, then the mining market was never really mature: Until 2021 we may had seen a market with lower supply than the optimum (and thus high profitability margins). It's possible according what I read in the forum and other communities that this was mainly caused by hardware shortage - there were less gears available than people willing to mine. And according to this assumption, we are now in a phase with a higher supply (of hashrate) than the optimum, with several miners mining at a loss.

The consequences would be that we could see a sharper hashrate drop, but in general this would mean that the market would be maturing and that would be in general favourable for Bitcoin because if we see really a major hashrate drop this year, this will probably not happen again then so easy and the curves would become smooth again. Miner diversification into housing / AI / HPC is a sign of that business models are evolving to more sustainable ones.

The problem that halvings are still a reduction of security vs. market cap of course remains, I never disputed that. This is of course also the root of the big block/small block debate or of ideas like "Bitcoin should go PoS" or "tail emissions". My argument is that if nothing changes, we could lose 40-45% attack cost vs. market cap in this halving, the next time it will be closer to 30-40%, then 20-30% and so on because fees will make up a larger portion of the income even if the fees in USD stay constant.

But we could improve that equation with models like merged mined sidechains (higher throughput for the same fee level -> slightly higher cost for miners but much larger attractivity for users -> higher usage -> higher price -> higher overall security).

I am in the same opinion as d5000. The total hashing power may have reached a major saturation point that some miners, especially the unprofitable miners, may need to turn off their ASICs until there's a more "favorable situation" - but that probably will not last very long either.

BUT there's another argument. Because we know that miners could be speculators too, it could also mean that they will add more hashing power if they think that Bitcoin's price will surge, no?

Let's not forget that higher fees (in BTC terms) are a result of higher transaction count, if Bitcoin will be used only for big transactions as some say then of course transaction count will be smaller than what we have today and thus the fees will be smaller.

You can't expect to have more big transactions than small transactions we have today.

How do you expect Bitcoin miners to survive on fees and too small block reward then ? the only way for this to work is the FIAT evaluation doubling after each halving.
That's a big gamble on the network security.

This mining companies are already switching to training AI models as there is more cash to earn, they will not bet hundreds millions of dollars on future Bitcoin growth when they can earn it today from different investments.

Bitcoin mining is quickly becoming thing of the past, current landscape is AI.

Does anyone really believe Bitcoin will be alive 10 years from now if nothing changes in the protocol ?

Bitcoin mining companies have farms of ASIC miners which are designed to only do one job, to mine Bitcoin. How can they offer other cloud services? If they had GPU mining farms (which isn't profitable and as far as I know, everyone is on ASICs), then they would be able to rent their GPUs to AI labs and anyone who wants to use it for AI.

It's basically a fight. The one who surrenders the last, keeps playing. It's really very interesting.

No one is going to attack Bitcoin. Simply it doesn't worth and is completely meaningless. In fact, it would be one of the dumbest thing that humans could ever possibly do. What if they attack the Bitcoin network? The code is here, there are countless number of altcoins, if Bitcoin falls, another coin rises and I don't believe someone is going to spend billions of dollars every time a coin is created. Also, too many serious companies are involved in adopting and integration of Bitcoin and too much money is put into it, we talk about trillion.

This mining companies are already switching to AI, it's worth for them to buy new hardware which they would have to buy anyway for Bitcoin as new generations of ASIC come along and kill old generation profit to the ground.

Source: https://time.com/6993603/ai-bitcoin-mining-artificial-intelligence-energy-use/ (https://time.com/6993603/ai-bitcoin-mining-artificial-intelligence-energy-use/)


If the security (diff) drops significantly, anyone will be able to attack it with a botnet for 0$ cost.


The money will dump faster than it pumped and only very rich will be able to escape this Titanic in time due to temporarily extra large fees, small investors will not have enough for fees when this panic starts and will be able to withdraw only after all the big investors money is gone.

Rich people feel safe knowing they will have priority on rescue boats when shit hits the fan, if you have less than $1m in Bitcoin, you're not on the VIP list.
If you have it, you'll be lucky to get 1/10th of it as it will be a fast fall, people in crypto space believe in 10x up but not in 10x down but it can happen and 10x is basically death for Bitcoin.
Some will still hold it at this low levels and get eaten by market manipulators who will pump and dump it, some will hold it regardless and die on this iceberg.

https://i.postimg.cc/7678Pm4n/St-wer-Titanic.jpg

The problems are already arising but people are like frogs in a slowly heated water and they're not realizing it's already boiling...

If Bitcoin falls to a 51% attack POW is dead! So in case of a revival, every PoS coin is at an advantage! And we have an incentive for ir already!
Also, I think you've mistaken the code for open-source projects and the ability of sites like torrent websites to come back with investors losing a few hundred billion on it. It's one thing to see an exchange go down it's a different thing to see the entire ecosystem being fucked up!

Second, you can't have a thing worth a trillion, a new revolutionary idea that threatens the world order or any of that thing, and at the same time have no enemies  ;)


Of course not, but again, I started from the OP first post, and that has 2140 in it so all my arguments are for that!
If we debate 2030 or 2040 , that's a different thing!


Same way VISA does, not charging 100$ for each of 1000 transactions but charging 1 cent for a billion! ;)
You know what I mean but let's not go deeper cause we're going to end with the same discussion about nodes not affording a 2tb drive.

Bitcoin would be so much more if community and developers wouldn't be so stubborn on keeping protocol at certain version.

There would be no need for altcoins to exist, there would be no need for convulsed Lightning Network.
The price would be way over 100k today because it would actually be used instead of horded by people wishing for it to go up by not using it while praying more big investors would pump their bags, praying for more regulations instead of fighting the machine and become money that everyone needs.

Instead we have "Digital Property" with collapsing roof and idiots who cheer when Bitcoin miners who secure this "property" get their profits cut down every few years by 50%.

Imagine having a company and knowing your revenue will go down by 50% every 4 years and that you have to invest a ton of money in new hardware even sooner and the only thing that can keep your company afloat is new investors buying shares of X company* in hope other investors will do the same and they can sell faster than them.

Does it seems reasonable ?

Ridiculous, I can see why this same miners are so happy that something new came for rescue in form of AI.

*Analogy to Bitcoin, if someone didn't understand or will try to fight using words taken out of context.

There is a very apparent reason why altcoins exist: We don't need to come into one agreement, if some of us disagree with each other. Remember: coercion holds no sway in this space. I cannot force you follow my proposal for 100 MB block size. I can only implement it, and invite you to join.

Bitcoin is ingenious technology because it gives you the freedom to define it yourself. If you think Bitcoin has failed, it's because you've overlooked at this feature. But, let me ask you this question: What should the Bitcoin developers do, if not keep it "at certain version", as you claim? What should it change, in your opinion, so that it would make it thrive?

First off, the halving should stop and tail-emission added.
This approach would give more assurance for security, why it's in danger with current scheme I have already explained.

Block size should be dynamic and not static so the fees would not skyrocket and block poor people from using or even withdrawing their money.

If this two things were fixed, Bitcoin would make sense as an alternative to fully obfuscated chains.

PS. I know anyone can fork Bitcoin and make changes but starting a new PoW chain with no premine, no dev-tax etc. would be very hard to secure if not impossible today, maybe with merged mining it could survive.

No! This logic is flawed.

To get to the point that people "only use Bitcoin for big transactions", fees must be much higher than now.
On-chain transactions are always the safest form of a Bitcoin transaction, compared to Lightning, sidechains and L2s. So people will always prefer on-chain when they can afford it.
In the (unlikely) case that really sidechains and LN become so popular that the blocks become emptier as a consequence, there's still the option to decrease the mainchain blocksize to boost the fee market, like Luke-jr and Paul Sztorc have proposed. (I personally think this is insane from today's point of view, but it may be an emergency strategy if things really go terribly wrong).

Yes, but it's a diversification strategy in times where the mining market is saturated and margins are thin, see also my last answer to stompix. Once difficulty falls enough mining becomes profitable again. And I'm sure that will not mean a 50%+ hashrate/attack cost drop. Let's talk about that again in a year. :)

By the way what you are aiming for (non-premined, tail emission and bigger blocks) already exists. This list probably contains some options (https://bitcointalk.org/index.php?topic=5456378.0). From the larger ones, Dogecoin has tail emission (and community seems to be open for big block ideas (https://www.reddit.com/r/dogecoindev/comments/u3wc71/increased_block_size/)), XMR too, and there are a whole armada of non-premined coins where some should have also a big blocker ethos. I think you should simply support Doge and XMR thus instead of being angry about Bitcoin not following your enlightened ideas :)

Ok, that's so far away that a lot of variables will be in play and we do not even know if Bitcoin will still be needed in 2140 :)

I already wrote that elsewhere: they simply change the hardware. That's no big deal. For example, if a generation of ASICs is phased out because it's no longer profitable, in the sections of their farms they had these old miners running, they simply install "commodity" servers with strong GPUs.

Yeah let's decrease the block size so it can only fit 1 transaction and make block times longer from 10 min to 1 week...

Instead of scaling up, their "plan" is to scale down  :D

My logic is not flawed, your logic involves a lot of praying and I rather have a scalable chain than chain that lives on prayers.


People always prefer the cheaper option.
No one except purists (minority) care what it does in the background.

You don't need to start a brand new chain. Hardforking implies that you'll have the exact same chain, until date X, when your rules will start being enforced. Bitcoin Cash is an example.

A scalable blockchain is one where payments occur in second layers, not in the main layer. If you think that by simply rising the block size, you can solve the scalability problem, then you'll have to explain to me why Bitcoin Cash and numerous of other altcoins were such terrible failures.

I know you can do that but the problem of low hashrate and possible 51% attack still remains in this scenario.

It's about removing bottlenecks from the protocol and pushing scalability problem to the hardware layer.
When Bitcoin network was recently spammed even LN had trouble to work properly.

Bitcoin shouldn't be slow because code says so, it should be as fast as hardware and network infrastructure allows.
Bitcoin is outdated today and you want it to flourish 100 years into the future  :D

Then please refute my argumentation convincingly instead of using words like "praying" (in this case I promise to not continue using words like "enlightened" then ... ;D )

At least the assumption that the main chain will not lose attractivity in relation to L2s if fees lower is difficult to refute, I think.

Why is Lightning then not much more popular now, and L-BTC and Rootstock stay in niches? My interpretation is easy: the current on-chain cost is still low enough for people not being forced to use these layer2's. People don't do micropayments on Bitcoin (mainchain) since long ago anyway.

If you think that big blocks and a high tail emission (for example, if we continue with 3.125 BTC per block forever) are the solution, then how do you ensure that the following negative consequences don't occur:

1) full node centralization? (see BSV, Solana ...) (and again, this is NOT because of storage but because of relay/validation costs ...)
2) negligible transaction fees income, making miners totally dependant on the tail emissions?
3) exodus of investors attracted by Bitcoin's deflationary nature?

Problem 3 would lead probably to a hefty dump in price. And problem 1 is crucial in my opinion. If we sacrifice censorship resistance then we don't really need Bitcoin anymore. With a highly centralized network, not only censorship becomes easier but also 99.9999% of the users are forced to use SPV clients, and SPV clients can in some cases be attacked.