Hacker News Reader

Comment by graeme | Hacker News Reader

Right just thinking through the claims of the boosters to their logical end.

If it did happen in nine years the next halving is 2028 so the cost would be ~$85 billion per year. And that would be less than it is now due to inflation.

Ok that’s actually not as bad as I thought, though still pretty expensive. You need $85 billion from new entrants in burnt capital just to maintain the value of existing bitcoins at that price, and people need some reason for putting in capital other than price appreciation assuming appreciation stops.

View on HN

You seem to have a good grasp of creating high level models.

I always see people referring to the energy consumption of bitcoin as though it’s strictly additive.

What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies.

Second, what people tend to ignore is that innovation doesn’t come in a vacuum - there are second, and third degree improvements triggered from it. In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin.

Third, and this isn’t directed at you, just a general statement based on arguments I see on HN - cryptocurrency itself does not have high energy consumption. Proof-of-work mining does. There are methods to support cryptocurrency that do not result in high energy consumption (Proof of Stake, DAGs). It’s wrong to generalize energy consumption issues to all cryptocurrencies.

> What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies.

Very hard to compare. One thing I’m unsure of for bitcoin: do costs go up if there are more transactions or does mining difficulty adjust so that costs don’t rise no matter how many transactions bitcoin does? I realize this is a major gap in my understanding.

The finance and currency system does much that bitcoin doesn’t currently do. So comparing as a whole isn’t adequate unless bitcoin network costs don’t scale up with numbers of transactions because mining gets easier.

Actually how do miners make money once there are no more coins to mine?

> In a world where bitcoin is $1million, it means the world has agreed on the benefits, and we will see innovations to address the tradeoffs that come from this. For example, we could see significant investment into renewable energy as a means to support the energy consumption of bitcoin.

That actually doesn’t follow. All that follows is that the world has decided to move $170 billion in capital into bitcoin per year.

The world is presently set to move about $20 billion of capital into bitcoin per year, and there’s no real consensus above usefulness.

The world could be putting $170 billion in because there is value, or it could be a massive waste of resources. The world has certainly misallocated resources before.

And yeah I was restricting my own comments to bitcoin. Ethereum et al I know much less about.

> Actually how do miners make money once there are no more coins to mine?

With BTC (and I would assume most other cryptocurrencies), by transaction fees.

Wait is my whole calculation wrong then? I thought miners only got paid for receiving bitcoins.

They also get additional payments in fractions of a bitcoin for transactions and this is in addition to the costs I wrote above?

> They also get additional payments in fractions of a bitcoin for transactions and this is in addition to the costs I wrote above?

Yeah, the basic design of bitcoin is for the block rewards to dominate early on with no/low transaction fees, but transaction fees to gradually take over as the main reward for miners maintaining the network.

Ah interesting. Any estimates of how transaction costs scale? Currently iirc it is 100 million transactions per year at $20-$30 each, so adds about 10-15% to network upkeep costs at that transaction volume.

miners are getting payed for every transaction. if you don’t include transactions fees your transaction will spend a lot of time in limbo and may actually never be mined.

the way it works, let’s say you have A bitcoins and you want to pay someone B bitcoins. you create a transaction that say: I want to move A to B and A-B-fee the miners see it and see the fee. it gets included (ie mined) in a block.

actual example 50BTC -> 20BTC + 29BTC

when the miner mines it, it will send the tx fee to its own wallet

also, most miners sort transactions by the most profitable to less profitable and mine only the most profitable ones

good question about baseline infrastructure energy cost

> >I always see people referring to the energy consumption of bitcoin as though it’s strictly additive. What I’d be interested to know is how it compares relative to energy costs for transacting with and securing fiat currencies.

the base load of the combined major central banks, transfer networks and custodial institutions is undoubtedly considerable

not least I'm thinking how much still runs on Itanium and OPENVMS. Just ten years ago I would have wagered the majority is on Itanium running vms or nonstop or the Honeywell legacies virtualized by Unisys. now a significant UK clearing bank ( at clear.bank) has run on Azure with white label customers who would have brought plenty of attention in the event of any problems, I'm no longer sure at all that I have any real idea what infrastructure is on.

re the energy costs of the network and that growth, I'm baffled - surely bitcoin should have moved to a custody model of settlement in view of the energy and other costs and risks of on chain transactions by now?

imagine I have a bitcoin account and my bank lends me to trade against the bitcoin as collateral.

if the account is a true bitcoin account I borrow btc and owe a greater amount of btc to settle.

it's not unreasonable for my bank to expect me to close my trading profits into my bitcoin account.

that's just two regular bitcoin demanding and regular transactions creating more demand

Presumably the 85B would have to be paid in network fees assuming price doesn’t move?

Reply on news.ycombinator.com