Hacker News Reader

Comment by steve19 | Hacker News Reader

It says a single transaction generates almost 300kg of CO2. Surely not. 95g of e-waste also seems incredible. And 600 kwh of energy per transaction? Really.

Edit: based on a cost of electricity at $0.1/kwh, and assuming bitcoin operators break even and are rational, one transaction costs less than 3kw of energy:

https://bitcoinfees.info

If it cost 600kwh, we would assume a transaction would cost at least $60.

Electricity is probably a lot cheaper in China. If it was half as much we would be looking at under 6kwh of electricity at the most.

View on HN

They're missing that the reward has fallen below the cost of non-renewable energy, such that the only profitable way to mine Bitcoin now is with renewables. The only way to profitably mine Bitcoin with any country's electric grid would be if someone else is paying the electric bill for you.

> Electricity is probably a lot cheaper in China.

Specifically, miners in China are famous for colocating with overprovisioned excess hydroelectric power that is not transportable elsewhere via the grid.

Easy to miss the point that the power would be lost if it wasn't used.

An average block has something like 2500 transactions in it. It means that there is about 1-2 transactions per second. Bitcoin network is running at 100 million TH/s. Modern mining hardware does 10 GH/Joule = 10^10 H/J, so 100 million TH = 10^8 * 10^12 H = 10^20 H is 10^20/10^10 = 10^10 Joules. 1 kWh is 3.6 * 10^6 joules, so 10^10 Joules is something like 3 * 10^3 kWh, so 3 MWh. One ton of coal produces something like 2 MWh. So yeah, assuming some of the electricity is not coal, the numbers do add up.

Do you have some back of the napkin counter calculations? Or critiques of their methodology?

I'm not sure if the article is right or wrong, but the commenter above you was probably getting at this:

You can see how much money a block miner receives for a block of transactions, and given the money they received, and an estimate for the cost of power, you can estimate how much power they might have been willing to expend on it

Going off the guy's link, I guess it's 24 cents to get into the next block, and a block has about 500,000 transactions, so whoever mines that block gets $125,000. Assuming 4 cents per kWh, the presence of only one miner, and no other operation costs, he'd still be breaking even if he expended (1 kwH/$0.04)*$125,000 =~ 2,500 kWh per block

(and by the same napkin math, that's 6 kWh per transaction)

What's weird is TFA seems to be using a similar methodology (but looking at total bitcoin mining revenue instead of an individual block), and yet they came out with a number 100x as large. I'm not sure why there's such a big discrepancy.

625 kwH is microwave oven running continuously for about 3 weeks. This is a lot of energy.

If 625 kWh per transaction is accurate, then 300kg CO2 is plausible depending on the grid. Many electric grids emit 500g CO2 per kWh or more. Others are much cleaner depending on the production mix.

See: www.electricitymap.org

I don't think most Bitcoin mining is powered by a grid at all. I think the reward would be less than the cost using grid power.

Reply on news.ycombinator.com