Bitcoin is valuable even if adoption never does pick up. It's a money system with certain properties that can't be found anywhere else, and it's a money system that's that's half decent.

The existence of Bitcoin forces all other money systems around the world to be at least as good as Bitcoin, or citizens will flee into Bitcoin.

Bitcoin adds value simply by existing, even absent serious adoption.

How many products and services are bought every day with conventional monetary systems?

How many products and services are bought every day with Bitcoin?

I suspect BTC will have as much impact on conventional currencies as a fly landing on an aircraft carrier.

Bitcoin also spends about as much money as a fly landing on an aircraft carrier. Sorry Argentina, but you just don't use that much electricity in the context of the global economy.

Companies like Tesla are buying Bitcoin, not because they intend to transact with Bitcoin but because it gives them protection against inflation within the US economy, and it gives them a stash of money that cannot be forcibly shut down, allowing them to do business even in jurisdictions that don't want them there.

If you are looking at just the "products and services" you are missing the forest for the trees.

There’s so many things wrong with this comment.

Until people start accepting bitcoin for bread, rent and salaries, Tesla most certainly does not have a “stash of money that cannot be forcibly shut down”. It has something that might be able to be exchanged into a local currency to buy those things.

What’s more, that stash of bitcoin most certainly could be forcibly shut down at the stroke of a pen. Tesla is a public US company listed on a US stock exchange. If the USA decided tomorrow to criminalise BTC, Tesla would be left with a huge 1.5bn hole in its accounts and a huge number of very pissed off investors.

Moreover, if Elbonia doesn’t want Tesla there, how in the hell are Tesla cars going to arrive at ports? How are they going to be registered? How are Tesla going to deploy charging infrastructure? BTC has exactly squat to offer in terms of dealing with jurisdictions that don’t want them there.

BTC isn’t some magic wand that lets you operate in a fairy dimension where none of the rules of reality apply.

Just because Tesla doesn't have the ability to sell cars in Elbonia doesn't mean they don't have some form of commerce requirement with Elbonia. Maybe they are paying devs to work on the UI, or paying a computing company to run simulations.

Bitcoin is already accepted by many for rent and salaries. Bread is not likely to ever happen because transactions are too inefficient, but that doesn't mean you can't viable have an economy based on larger exchanges.

I often wonder if the release of Zelle around 2017 (near instant, fee-free transactions between personal and business bank accounts between certain US banks) was motivated by a perceived crypto threat.

I mean, that’s been pretty standard in every country I’ve dealt with outside the USA for a long long time.

If anything, I’d say that’s more a reflection of the poor state of the USA banking sector than its currency.

vs the rise of Venmo and Square Cash?

The power consumption is linked to network size and has nothing to do with the block itself. If anything the marginal power cost is linked to the price of a Bitcoin, as the higher it climbs the more incentive to mine, the greater the competition and therefore the greater the mining difficulty/energy required. The only way to really lower energy consumption is to lower the price I can get for a Bitcoin but with the current economic conditions I don't see that happening anytime soon.

How is power consumption related to network size? I agree that the power consumption of the network is directly related to the value of the bitcoins mined (and I supposed the transaction fee/tip size).

I think by 'network size' the GP is referring to the network hashrate.

For proof of work networks to function securely, the network as a whole must have a much higher computation rate than an attacker.

Obviously capabilities increase the whole time, so the bitcoin network needs to be able to adapt - to do this, it adjusts its 'difficulty' approximately every two weeks. The power usage corresponds directly to the difficulty level set by the network.

If you add more miners, the difficulty will increase, and power consumption will go up, if miners leave, the difficulty will decrease and power consumption will go down.

Network size increases competition to mine blocks faster, leading to demands for higher hash rate, leading to higher power consumption.

That's not how Bitcoin mining works.

A certain amount of bitcoin -- the amount determined by a schedule that was defined before the network became operational -- is given as a mining reward every 10 minutes. The incentives of the individual miners is such that the expenses of the miners (collectively) equals the mining reward -- and the major mining expense is electricity.

If the mining reward is cut in half, the electricity consumption of the network is cut in half, too.

In contrast, if the rate of transactions changes or the number of miners change, electricity consumption stays the same. (More precisely, the expenses of running the network equals the mining reward plus any transaction fees, and since the blocks are of fixed size, for more transactions to compete for space in the blocks increases transaction fees, but I am guessing that transaction fees are currently a small fraction of the mining reward.)

I think you are _very_ mistaken.

> If the mining reward is cut in half, the electricity consumption of the network is cut in half, too.

That's not true it all -- there is no difficulty adjustment when the reward is halved. There may be pressure on some miners to stop mining, but that adjustment is not immediate and can also be compensated for by change in the price of bitcoin.

> In contrast, if the rate of transactions changes or the number of miners change, electricity consumption stays the same.

Given the network is operating at peak transaction rate already, there's not much change here. Neither block size nor the transaction count has a meaningful affect on the computations required to "solve" a block. The merkle tree for the transactions is computed once, but most of computation is finding a nonce, that combined with the rest of the block header, produces a hash with a certain number of leading zeros.

Changing the block size or transaction count doesn't meaningfully change electricity consumption.

I agree with your final 2 paragraphs.

>that adjustment . . . can also be compensated for by change in the price of bitcoin.

It can. When I wrote that if the mining reward is cut in half, the electricity consumption of the network is cut in half, too, I assumed that the price remains constant.

A miner must pay for the electricity he or she uses. Where do you think the money comes from to pay for the massive amount of electricity used by the network?

Do you imagine that rich people (i.e., people who can afford to lose money) are buying the electricity for pro-Bitcoin ideological reasons?

I don't: I believe that it comes out of the revenue made (collectively) by the miners. In other words, the payments for the electricity are parts of (individual) plans to make money through mining.

And I believe that if that (collective) revenue were cut in half -- especially if miners and prospective miners knew of the halving in advance (particularly, before they decided what mining hardware if any to buy) -- then money spent on electricity is approximately cut in half, too. (Because otherwise the plans to make money would not work.)

You are correct that the adjustment in the hash rate is not immediate after a halving of the mining reward. Mostly that is because miners who recently bought mining hardware have to continue mining after the halving to continue to pay for their hardware.

If the halving was announced in advance, then the halving will start exerting downward pressure on the mining difficulty months in advance of the actual date of the halving. The major cause of that downward pressure is miners opting not to upgrade their hardware and prospective miners opting not to enter the mining business in the first place (the effect of which is to make it take longer for the remaining miners with their non-upgraded hardware to solve the proof-of-work puzzles, which in turns causes the software to reduce mining difficulty to return the average time between successive blocks back to 10 minutes).

Basically, it takes many months for newly-manufactured mining hardware to pay for itself, and that delay is the main reason the response to a reward-rate halving is slow. But again the response starts months before the actual halving; and the cumulative effect of the halving on the rate -- more precisely the effect the rate has on how much electricity is consumed by the network over the years -- is approximately the same as it would be if the effect of the halving on the rate were instantaneous.

It is the fact that one of the major expenses (namely, hardware) of the miners is "lumpy" (requires an upfront expenditure that is then recouped over many months) that obscures the simple relationship whereby the collective expenses of the mining community approximately equals the collective revenues of that community -- where most of that revenue is from mining rewards, which is equal to the price of bitcoin (or the value of bitcoin if you prefer) times the rate at which the network dispenses bitcoins from miners as rewards. What makes me confident that expenses = revenues is that miners are rational and consequently are capable of taking into account scheduled halvings of the rewards to mining and the "lumpiness" of the cost of mining hardware (and many other factors).

I agree with pretty much everything you said about the costs and rewards of mining and where the pressure points are. What I don't agree with is the simple statement:

> if the mining reward is cut in half, the electricity consumption of the network is cut in half

This would only happen if half of the miners shut off their equipment as soon as the reward adjustment occurred, which does not happen [1]. Even then, that doesn't change the electricity consumed by a single miner producing a single block, since that is governed by the difficulty parameter, which also did not halve and took multiple weeks to bottom out before increasing again [2].

There is no simple function between block reward and network or block electricity consumption.

1. There was a drop of 120PH/s to 90PH/s within a week after the halvening. https://www.blockchain.com/charts/hash-rate

2. You can see a similar curve of difficulty _following_ the hash rate of network, albeit over multiple weeks. https://www.coinwarz.com/mining/bitcoin/difficulty-chart

It is how the mining works from a bird's eye view.

> A certain amount of bitcoin -- the amount determined by a schedule that was defined before the network became operational -- is given as a mining reward every 10 minutes. The incentives of the individual miners is such that the expenses of the miners (collectively) equals the mining reward -- and the major mining expense is electricity.

The 10 minute rate is true, but the expense is not collective. I as a miner am in active competition with my fellow miners and am incentivised to find ways to decrease costs (electricity), but only mine. In fact it would be in my best interest to increase the electricity cost of a competing miner in order to force them off the network, and if I win the block it is only my personal power backing that block.

> If the mining reward is cut in half, the electricity consumption of the network is cut in half, too.

Not necessarily. If the reward is cut in half but the price of BTC has risen considerably since starting mining operations then I will remain on the network as the profit is sufficient to continue. This is why the "halving" event that Bitcoin goes through has never had a major impact on either the price of BTC or the scale of mining operations globally.

> In contrast, if the rate of transactions changes or the number of miners change, electricity consumption stays the same. (More precisely, the expenses of running the network equals the mining reward plus any transaction fees, and since the blocks are of fixed size, for more transactions to compete for space in the blocks increases transaction fees, but I am guessing that transaction fees are currently a small fraction of the mining reward.)

This is very wrong. The expense of running the network is equivalent to all the energy burned by the entire network in order to find the next block. This isn't felt by the individual miner in any way other than a potential increase in competition to find the block, but taken as a whole the network is for example now considered to be using the total energy consumption of Argentina.

Almost, but because of the incentives, the greater the adoption, the more valuable each individual block reward actually is. The more valuable each individual block reward is, the more energy you can afford to spend attempting to mine it.

You can't fit 10,000 transactions in a block on BTC.