This needs to be repeated in every HN thread about Bitcoin: no, transactions don't consume energy. The proof-of-work is completely independent of the number of transactions. A block could have 1 or 1000 transactions, but the energy consumption would be the same.
People have this wrong idea that more transactions imply more energy consumption. That's just not true.
The calculations are based on the maximum throughput of the Bitcoin protocol, which is a fixed value due to the 10-minute block time and the limited block size.
The average number of transactions per block is always around 2K: https://www.blockchain.com/charts/n-transactions-per-block
> The proof-of-work is completely independent of the number of transactions.
Right, which makes this whole thing even more ridiculous. If number of transactions went to zero we'd still have to burn 120+ TWh per year just to keep the system going.
The 165kWh per transaction calculation is the best case scenario assuming every block is filled to the maximum with transactions.
The Bitcoin blockchain could subsist on a minuscule fraction of todays energy usage if the majority of miners decided to throw away their investments and stop mining.
Whatever else you may think of those features, it's a bit weird to call it pure waste when clearly millions of people find this incredibly valuable.
There's a block limit, though, right? It doesn't scale up infinitely -- you might have 1000 transactions in a block, but not a million. Therefore, it is accurate to attribute the energy per block to securing the transactions within.
With the current concentration of mining capacity in China, that's not exactly a sure thing any more.
It can be safely increased over time as bandwidth, storage and processing costs go down.
For now there's still a lot of viable efficiency enhancements to be implemented/adopted so increasing the limit isn't strictly necessary.