Hacker News Reader

Comment by durkie | Hacker News Reader

so this has been an issue that keeps coming up for a while, but i don't really understand the issue.

* As others have asked, how much is bitcoin supposed to use? What's the right amount and why? How is that number better/worse than the amount of resources spent on existing financial infrastructure?

* We are rapidly progressing towards an era of free (extremely cheap) clean power. Just today (https://electrek.co/2017/11/08/chilean-solar-down-26-as-impo...) incredibly low bids for solar projects were submitted in Chile at 2.15c/kWh. That's insanely cheap. We're not there yet in having ubiquitous cheap clean power obviously, but neither is bitcoin "there" in being able to replace much of traditional financial infrastructure. But the point still stands: will bitcoin's energy usage still matter in the not-too-distant future?

View on HN

The problem is that with a competitive mining market, Bitcoin energy use will rise until the cost of mining a block is equal to the reward for that block (i.e. market equilibrium). So if the cost of energy goes down but the price of Bitcoin stays constant, the amount of energy expenditure will be (approximately) equal.

Likewise, if the price of Bitcoin goes up, the amount of energy used will go up accordingly.

The reward amount will decrease over time and mitigate this, but not enough by a long shot IMO.

This is a key observation. Now apply it to the "eco-friendly" and PoS schemes. The same economics apply. For Bram's coin and PoS systems, the costs spent will rise to equal the block reward. So, you are basically doing something more complicated to end up with the same amount of economic waste. In the proof of space version, maybe you will be building millions of hard drives, running them to failure and then throwing them away.

This answer written by the Ethereum team addresses why cost spent won't necessarily rise to equal block reward in their PoS formulation. https://github.com/ethereum/wiki/wiki/Proof-of-Stake-FAQ#doe...

Mining rewards will be gone in a couple of years, then all that's left are fees to pay for mining - meaning, for securing the blockchain.

Higher energy costs mean higher transaction fees, which mean a disincentive to transact, and vice versa. The end state of bitcoin is a hash rate which corresponds to how much people are willing to pay for transactions. It may be a lot lower than it is today.

I haven't seen the hashrate drop when the price drops. We may be more limited by supply of ASICs than the price of electricity.

> how much is bitcoin supposed to use? What's the right amount and why? How is that number better/worse than the amount of resources spent on existing financial infrastructure?

Something approaching zero joules per transaction seems like the number to try to achieve.

There is also a chicken and egg problem: if Bitcoin was actually to be used for normal day to day financial activities then it could be more excused (as you say, financial transactoins don't come for free today). But as long as it's basically a ponzi scheme of processor farms and the transactions used are basically either illegal or speculative - then it doesn't provide the same vale that "existing financial infrastructure" does. So while I do think there may be a future for cryptocurrencies , I certainly don't hope Bitcoin (or any PoW currency) is it. And to be honest I'm not optimistic about the idea of "anonymous" money.

> We are rapidly progressing towards an era of free (extremely cheap) clean power.

That's great. And we need to speed that up. And while we are making progress, we are suddenly eating a lot of that progress by computing hashes for a currency used mainly for speculation. I think it's a cool concept but it must be one of the stupidest things humans have done in recent years.

Once we do have cheap clean energy I'm all for using it for whatever. But right now we keep adding demand for energy (which then includes dirty energy prices also go up, which means it's economical to keep running coal plants a while longer, and so on).

Ideally >99% of the power used by bitcoin would be in sharing and storing the blockchain and pending transactions. The security mechanism would use as few resources as possible.

Be careful when you compare to existing infrastructure, because bitcoin has to do all that too, plus mining costs on top.

The energy consumption of the incumbent financial infrastructure is extremely difficult to measure, and so it has never been seriously considered a "problem." There hasn't been an alternative. What does it cost to put a roll of quarters into the hand of a laundromat customer in Faibanks? ¯\_(ツ)_/¯

within roughly 5 years or so (very rough guess on my part), all of the available Bitcoin will have been mined. At that point, the mining part of the equation will be moot, leaving only the blockchain infrastructure to consider. Note that this may not hold true for various "forks" of the bitcoin blockchain, or for other cryptocurrencies.

Not true at all. The security of the bitcoin blockchain is fundamentally based on mining; after the inflation schedule runs low enough that it's not the major component of mining rewards (which won't be for decades), miners will continue mining, but for the transaction fees instead.

No need to guess, the reward schedule is written into Bitcoin:

https://bitcoin.stackexchange.com/a/162

Reply on news.ycombinator.com