Texas paid a Bitcoin miner to use less electricity during hottest month(texasmonthly.com) |
Texas paid a Bitcoin miner to use less electricity during hottest month(texasmonthly.com) |
Texas paid Bitcoin miner more than $31M to cut energy usage - https://news.ycombinator.com/item?id=37428605 - Sept 2023 (130 comments)
Texas paid Bitcoin miner Riot $31.7M to shut down in August - https://news.ycombinator.com/item?id=37418866 - Sept 2023 (55 comments)
An aluminum smelter uses a humongous amount of power to, well, smelt aluminum. Due to the square-cube law the aluminum takes a really long time to cool back down. When you are trying to heat something from 20C to 700C it doesn't really matter if you take a break in the middle of the heating cycle. Sure, it might cool down from 500C back to 450C, but who cares?
As long as they get paid enough to make up for their workers getting an unscheduled paid day off, the smelter really isn't going to complain too much.
Some lobbyists must have lined some pockets.
The issue with "pure" surge pricing is that it doesn't solve the actual problem. Most businesses - even "non-essential" ones will have fixed-price electricity contracts. They are not affected by surge pricing on the electricity market and have zero incentive to reduce their usage even when the market price spikes 100x.
In practice the "market forces" here are a bunch of Wall Street investors losing a shitton of money because their bets went sideways, but that is not going to fix the grid itself.
Again, grid stability at the lowest cost is the name of the game.
As usual lots of hot takes from people who just read the headline
Not really. NIST did a really good introductory blockchain technology whitepaper in 2018:
* https://csrc.nist.gov/pubs/ir/8202/final
Figure 6 is a decision flowchart laying out whether blockchain or some other technology would be best for various needs / use cases, and blockchain is very niche:
* https://twitter.com/justincormack/status/1048875246437195778
Needing both distributed and trustless simultaneous isn't common, especially in the modern, connected world:
> Reputation is a technology. We solve a lot of problems in the economy by turning one-shot games, in which people literally take the money and run, and turn them into repeated games, in which people have an incentive to behave well in this period so that you will trust them next period, and so on, into the future.
* Paul Krugman, https://www.youtube.com/watch?v=Y_IYGeZLLhI&t=10m30s
It’s so slow and cumbersome that the vast majority of use is via banks, making it effectively quite centralized. It’s so volatile it’s a horrible currency so its main role is gambling and speculation. It’s pretty useless as the basis for decentralized communication or storage systems for a variety of reasons.
The only decentralize app even tangentially related to Bitcoin I’ve ever seen be marginally useful is Nostr, and Bitcoin has nothing to do with how that works. It’s tacked on as an afterthought. Nostr is just a straight up relay-facilitated peer to peer network with cryptographic addressing.
If anything the presence of Bitcoin makes Nostr worse by flooding it with spam, endless chatter about sending Satoshis, and stupid laser eye memes.
I'd like to see the data sources and working on that.
As for decisions, a WWII European economy would have ditched both as excessive and not contributing to the current situation.
> According to a recent post by the Center for Global Development, Christmas lights consume a whopping 6.63 terawatt-hours in the US alone, with the global figure being substantially higher across the rest of the world. The tiny state of El Salvador, for instance, consumes 5.35 terawatts for Christmas lights, and the African nation of Tanzania stands at 4.81 terawatts. Add these figures up all over the globe and you soon discover that bitcoin isn’t the energy consumer that many environmentalists lead people to believe – at least not compared to Christmas lights.
This is the post from the Center for Global Development that Mawson is using [2]. The relevant paragraph says:
> A 2008 study from the US Energy Department’s Energy Information Administration (EIA) found that decorative seasonal lights accounted for 6.6 billion kilowatt hours of electricity consumption every year in the United States. That’s just 0.2% of the country’s total electricity usage, but it could run 14 million refrigerators. It’s also more than the national electricity consumption of many developing countries, such as El Salvador, Ethiopia, Tanzania, Nepal, or Cambodia.
That is followed by a chart that shows the national consumption of El Salvador, Ethiopia, Tanzania, Nepal, and Cambodia, and the decorative holiday light consumption of the US.
Mawson misunderstood that chart. The 5.35 tWh for El Salvador and 4.81 tWh for Tanzania that Mawson cites as for Christmas lights are those country's total consumption.
In other words, what Mawson has shown is that the electricity used for Christmas lights in the US plus the electricity used for everything in the rest of the world (including Bitcoin!) is more than the electricity used for Bitcoin.
Let's see if we can figure out the right numbers. First, just for the US. Most figures I have found are for cryptocurrency in general not just Bitcoin, but I believe that Bitcoin is by far the biggest proof of work cryptocurrency so we probably won't be far off to attribute almost all of it to Bitcoin. Adjust all the following by whatever you think the ratio of Bitcoin to total cryptocurrencies is.
Annual US electricity use is about 4 x 10^12 kWh per year [3].
US Bitcoin electricity use is 0.9 - 1.7% of US electricity use [4]. Going with the bottom of the range that is 3.6 x 10^10 kWh per year.
The CGD number of holiday lighting, 6.63 x 10^9 kWh per year, were from 2008. Assuming that the proportion of electricity that goes to holiday lighting is about the same, that would be about 5% more now.
That puts US Bitcoin electricity use at least 5x higher than US Christmas light electricity use, and possibly up to 10x.
For the whole world, total electricity consumption is around 2.3 x 10^13 kWh per year [5].
The US is estimated to host about 1/3 of the crypto operations [4], so the global Bitcoin energy use should be about 3x that of the US or about 1.1 x 10^11 kWh per year, which is about 0.5% of world electricity use.
I haven't found anything on Christmas lights around the world. In the US, Christmas lights are around 0.2% of the total. If we assume that is the case for the rest of world that would give crypto uses 2.5 - 5x the Christmas light usage.
Also note that the Christmas light data for the US was from 2008. I'd expect Christmas lights nowadays to be much more likely to be LEDs, which would use less power.
[1] https://mawsoninc.com/bitcoin-mining-uses-less-energy-than-c...
[2] https://www.cgdev.org/blog/us-holiday-lights-use-more-electr...
[3] https://www.statista.com/statistics/201794/us-electricity-co...
[4] https://www.whitehouse.gov/ostp/news-updates/2022/09/08/fact...
[5] https://www.iea.org/reports/electricity-information-overview...