If you google search "how much power is used by online banking" there are multiple analyses answering this question for you along with consumption estimates for Bitcoin. One pro-bitcoin one from my first 4 results attempts to estimate all the power consumption for every bank server, branch office, and ATM 24/7 around the entire globe and produces an estimate of 140TWh/yr vs bitcoin's (at the time) 32.56 TWh/yr estimate. You can compare their average transaction values for yourself to decide how favorable this comparison actually is for bitcoin.
An article I read today provides a comparison point for realtime transaction processing, Fedwire, along with its transaction volumes and server count. It's interesting to compare that with Bitcoin's volumes. The article also makes claims about the energy consumption (suggesting FAANG's energy consumption is already dwarfed by Bitcoin) but I can't manually verify those myself: https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-c...
The above article estimates a best case total of 55.1 TWh/year for the major Proof of Work chains, and also provides estimates for various types of mining equipment, etc.
Bitcoin's energy usage isn't determined by the transaction count but by its valuation. Raising the limit of transactions on the Bitcoin network wouldn't affect the energy consumption (but it would have other trade-offs).
Aren't there significant technological obstacles to raising the transaction limit, not to mention the fact that large miners have actively obstructed attempts to raise it in the past? My understanding is that we only finally got a block size increase with segwit in 2017 and that only happened after people worked around miner sabotage.
There are no technological obstacles per se, but it will increase the rate which the size of the blockchain grows.
It's true miners have obstructed attempts to make these kind of changes but they don't control the price of Bitcoin. If market demand for a higher transaction rate becomes great enough and they don't adapt, then another coin with bigger blocks will simply outcompete Bitcoin.
An article I read today provides a comparison point for realtime transaction processing, Fedwire, along with its transaction volumes and server count. It's interesting to compare that with Bitcoin's volumes. The article also makes claims about the energy consumption (suggesting FAANG's energy consumption is already dwarfed by Bitcoin) but I can't manually verify those myself: https://www.ofnumbers.com/2021/02/14/bitcoin-and-other-pow-c...
The above article estimates a best case total of 55.1 TWh/year for the major Proof of Work chains, and also provides estimates for various types of mining equipment, etc.