3 ms·
I think the price of Bitcoin does have a relationship to energy consumption. It’s obvious it’s on large time scales. And unless one has a very cheap energy sour
by randomhodler84 5y ago
I think the price of Bitcoin does have a relationship to energy consumption. It’s obvious it’s on large time scales. And unless one has a very cheap energy source it will always be cheaper to buy from the market. So it’s a good in the ball park metric, the core issue to transfer value to old coins is that older coins grow exponentially more difficult to undo — so the fact that a few kilowatts could have mined many thousand coins in the early days does not decrease those utxo security and this value. There is an additive value of every block, and while head block at difficulty D virtually upgrades all the blocks before it to D, as you would need at least D+1 to undo just head-1, we build bigger defense walls every 10 minutes on average.
- root_axis 5y ago> so the fact that a few kilowatts could have mined many thousand coins in the early days does not decrease those utxo security and this value utxo security doesn't matter - a unit of bitcoin that expended a fraction of a kWh to be mined is worth the same as a unit of bitcoin that cost hundreds of thousands of kWh to mine, thus the energy spent to mine a bitcoin has no meaning with respect to its price.
- randomhodler84 5y agoThere are multiple feedback loops. When energy is expensive, mining slows. This impacts yields from miners, and supply of new coins to the market. Also, there is a point where it is cheaper to buy the coins at certain energy prices; thus cost to produce the coins is a component in their market price. It is ignoring reality to say they are not linked concepts. One can demonstrate this in the myriad of shitcoins over the years — their value collapsed when mining stopped. When people did not want to pay to protect the network. Exactly which leads which is probably something that will switch over time in various cycles of its adoption.