> Bitcoin uses about 32 terawatts of energy every year

terawatt is a unit of power, not energy. You probably meant 32 terrawatt-hours every year.

Instead of converting to energy per year, I would just say that bitcoin consumes a few (between 1 and 10) gigawatts of power. That way you can compare to power plant output to get a sense of what that means.

Is that really a fare comparison? I mean Visa and Mastercard use the global banking infrastructure to complete its transactions. I cant open a Visa or Mastercard account but I can generate a wallet in one second and use it instantly.

I have two questions:

1) Is this wallet going to have any spendable money in it one second after creation, or do you have to wait for the network to process your transaction like everyone else?

2) How often is it useful to create a wallet in one second?

My point wasn't really about comparing Visa/Mastercard connected accounts with cryptoaccounts.

I just wanted to highlight that for a traditional transaction to be completed between to accounts requires a lot more energy and infrastructure than just the energy consumed directly by those services.

Nevertheless I still think its outrageous that so much energy is wasted mining BTC.

Please quantify the energy use of the conventional approach — many people want to make the same point as you, but there don’t seem to be any (easily referenced) numbers estimating the energy cost per transaction of the alternatives.

The number I've seen is about 1 gigawatt, +- 50%. That's about one big power plant.

It could be worse. Bitcoin mining is centralized in mining centers running custom ASICs, mostly located in cold areas with cheap hydroelectric power. That wasn't the original plan. If Bitcoin mining was widely distributed, we could have a billion laptops mining in the background. Laptops running 24/7, wearing out their ICs through electromigration, and consuming power from all sources at retail rates.

They're not running in places with cheap hydroelectric. It's cheap coal. It's all coal.

https://qz.com/1054805/what-its-like-working-at-a-sprawling-...

In Sichuan it's at least partly hydroelectric. In northern China it's probably mostly coal. People mine in other places, too, even here in Shenzhen.

You're wasting energy for posting comments too. If we think a way like this, we can't no anything.

What on earth is this logic. Bitcoin is _by design_ inefficient. Its scaling inefficiency is the principle method by which it ensures fairness. That's what PoW chain commits are. That's how they work.

Bitcoin was ALWAYS going to be inefficient. It was assumed that hardware would scale to keep up with demand, but that hasn't proven true in the advent of ad hoc mining pools.

> It was assumed that hardware would scale to keep up with demand,

No, it was designed to stay at "one block every ten minutes" regardless of the hardware added or overall hashrate of the system. Which is why you hear a lot of people arguing over block size and block compression: you always have one block every ten minutes, how many transactions can we squeeze into that block?

The only thing that adding more hashrate does is waste more power, since the network will adjust.

The power consumption is what I was alluding to. I guess I was too vague.

If commenting burned ten cents of electricity, no matter how efficient computers get, that would be a huge problem too.

Each post uses as little electricity as is feasible. Efficiency improvements lower the amount of electricity used. Reducing the use 90% would mean 90% fewer posts and ruin the posting experience.

Bitcoin doesn't try to minimize electrical use. Efficiency improvements increase the amount used. Reducing the use 90% would lower the transaction rate 0% and ruin nothing, but we can't do that because it's set up as a competition.

This is needlessly pedantic. OP is advocating for mindfulness, and lists a number of statistics for reasonable context.

The dose makes the poison.

Except Bitcoin is inefficient by design (it’s what proof of work is) and the dosage is the market cap and transaction load. So what you’re really saying is bitcoin at the scale it was meant to operate at is poison.

No, the dosage is the hash rate (if more hash rate joins, the difficulty is automatically adjusted to keep it at 1 blocks / 10 min). The transaction output is constant, and market cap is irrelevant to the energy inefficiency discussion.