I'm starting a new thread, because this has grown out of control and has some incorrect (vastly underestimated) figures.

I've been analyzing #CryptoArt #NFTs on @superrare and @niftygateway

I'm afraid the figures are even worse than I thought. https://twitter.com/memotv/status/1308683955705655296

Avg ETH transaction consumes 35 kWh, with 19.5 KgCO2 emissions.

However, avg transaction relating to an #NFT (across ~80K transactions) consumes 76 kWh, 47 KgCO2 emissions
=1 week's elec for avg EU resident.

This is for a single mouseclick that takes fraction of a second.

wtf

But it gets worse.

Avg #NFT (across ~18K on @superrare) has 4.5 transactions, 340 kWh, 211 KgCO2.
=EU resident elec for a month
=flying for 2 hours
=boiling 3x Olympic sized swimming pools.

For a single #CryptoArt #NFT, just to keep track of bids and sales.

wtf

But it gets worse.

The avg #cryptoartist (~649 on @superrare), has minted 28 #NFTs. 9.5 MWh, 6 tonnes CO2

Just to keep track of bids and sales.

WTF

BUT IT GETS WORSE

@superrare does 1 offs
What abt editions? Like on @NiftyGateway

In 2 months #cryptoart superstar @beeple's #NFTs consumed 300 MWh w 190 tonnes CO2

In 5 months @muratpak's NFTs consumed 440 MWh w 275 KgCO2

100+ yrs of an EU residents elec
200+ transatlantic flights each

WTF

And it's not bc they're famous & make big money

Financial cost =/= eco-cost

eg NFT (by #slimesunday) selling $40/ed (avg $226 in secondary mkt), consumes 120 MWh, 73 tonnes CO2
in 2 months
https://niftygateway.com/itemdetail/secondary/0xb9abb6e7fdc06fe25ac96d3da1c3ef4651ec660a/500010375

they all have 4-5 tx/NFT edition (same as @superrare avg)

WTF

What makes all of this even sadder is that #NFT Market is built around a scam.
I don't mean that in a philosophical sense; but a very practical one.
The blockchain does not
- provide authenticity or scarcity - not even artificial scarcity.
- guarantee secondary sales royalties

- and for the #cryptopunks : the blockchain decentralizes a *database*.
Decentralizing a database does not decentralize the ownership of assets, or mitigate the centralization of power.

The American Dream promises winners, and delivers.

But success is not evenly distributed.

Of the 633 artists on @superrare:

The top 0.1% of artists earned 8% of total income
The top 1% earned 21% of total income
The top 10% earned 57% of total income
The top 20% earned 75% of total income
The bottom 40% earned 2% of total income

I wouldn't mind the fact that the #NFT market was built around myths, if it wasn't so stupidly ecologically disastrous.

Every #NFT related mouseclick consumes on avg enough energy to evaporate an Olympic sized swimming pool.


wtf

Yes ETH2 is in the works, but it isn't here yet. Mainnet won't dock w Beacon Chain for a while. Right now & in the near future, PoW is still dominant.

At a time like this, for digital artworks to have emissions of 100s KgCO2 (let alone 100s tonnes for editions) is bonkers.

wtf

The aim of this thread is not to single out #cryptoartists mentioned above. I am sure they - and everyone else - have *no idea* that this is happening. I had to dig deep to get these numbers.

But hopefully now everyone does know.

This technology is simply not ready.

wtf

Working on this, I have seen 1000s of #CryptoArt #NFTs, and I am blown away by the amount of *amazing* work out there. And that do not fit into any other existing venues. This work absolutely needs a home. But works should not need to consume MWh

That is just plain bonkers

wtf

I invite the wonderful communities that are emerging to reconsider. Pls don't use unethical platforms.

I invite the platforms to reconsider. Pls build ethical platforms.

http://cryptoart.wtf 

More considered write-up not constrained by twts: https://memoakten.medium.com/the-unreasonable-ecological-cost-of-cryptoart-2221d3eb2053

PS Congratulations to my friends in the US (and all of us around the world) for getting rid of the menace in the WH. Hopefully the new admin will help move in a better direction.

Stay safe y'all.

I'm sure many will ask impact of #DL. I too was curious.

here is a comparison with #NFTs

and (via @joz_freeman) this looks good too: https://mlco2.github.io/impact/ 

I am incredibly sad for having written this. But sitting on these numbers (which I have been for a while), was just too difficult. Esp seeing the growth.
I hope my figures are wrong, but I doubt they'll be orders of magnitude wrong.
I hope ppl smarter then I can find solutions.

A lot of ppl have been saying that the figures are wrong because renewables. It is true that there are a lot of assumptions, all the way down. Truth is nobody knows exactly.

But even renewables is complicated, unless farms use their own off-grid plants...
https://www.wired.com/story/bitcoins-climate-impact-global-cures-local/

... or - as many articles claim - 90% of China's miners use underutilized hydroplants, & globally 'most' miners use renewables.

All of these link back to one single report, by a crypto-investment agency coinshares who essentially say their product is green and we should buy it.

I don't know if this has been independently verified. It might be true that most mining is green, but I wld love to see this report by an org who's business is not selling crypto. Otherwise it's a bit like the tobacco industry saying smoking doesn't cause cancer.

digiconomist (where I get some of my data from) is criticized (by crypto-investment bankers) for being a 'clueless blogger'. This could also be true. Though the estimates there are *conservative* when compared to the estimates from cambridge uni (for BTC) https://www.cbeci.org/ 

This discussion wrt the ecocost of crypto feels like the discussions (decades ago) around tobacco. Does it cause cancer or not? So many assumptions are made. Stakeholders say it's clean. Non-stakeholders (which may have other conflicts of interest) say it's not.

Most authoritative critique of the @DigiEconomist kWh estimation by J. Koomey (an expert, not just a crypto-banker). He calls the method 'simplistic' & 'hand-wavey'
https://www.koomey.com/post/179556571967

and
https://www.coincenter.org/estimating-bitcoin-electricity-use-a-beginners-guide/
(I know this is funded by crypto, but criticisms still hold)

Koomey shows another method by Bevand, which he says is more accurate. Which (for 3 days in 2017) are half of @DigiEconomist's more simplistic model.

Today, CambridgeUni use Bevand's method, & are getting 40% higher than digiconomist's estimates
https://www.cbeci.org/cbeci/methodology

that is regarding kWh estimation.
(and for BTC. This doesn't imply that ETH kWh estimation is correct. But it does mean that just bc method is simplistic, it's not necessarily overestimating, cld be underestimating too)

regarding renewables (CO2)
https://www.jbs.cam.ac.uk/wp-content/uploads/2020/09/2020-ccaf-3rd-global-cryptoasset-benchmarking-study.pdf

= IMPORTANT CORRECTION =

I had a chat w a founder of NG, & he explained that my figures were way off.

It's true I grossly over-counted the # of tx (on NG).

(It's unf that he didn't mention this 500 tx includes batch-tx w upto 100s of internal tx) https://twitter.com/DCCockFoster/status/1352349510069006337

But still, my estimates for NG are off. It turns out NG doesn't record any activity on the blockchain. Just first mint. All bids & sales are logged on their own db, which is great.

Of the 500 tx, I found 348. They sum to 275M Gas, or ~140 MWh (just under half my orig estimate)

I don't know abt the other 150 tx. If they're similar to the 350 I found (which they may or may not be), that sums to 400M Gas, 200 MWh, (2/3rd my initial estimate).

This cld be wrong, but luckily I've been offered help to calculate it properly :)

https://twitter.com/DCCockFoster/status/1352349511298019328

(here's the 348tx I used for Beeple's NFTs btw. Less than 1/10th the transactions that I'd initially counted. But adds up to half my initial consumption estimate. Still missing 150tx) https://docs.google.com/spreadsheets/d/1PK5q1BRaMqlhYEwgEDxUJa_bZhoxSF5NuPRDvy5VXeQ/edit?usp=sharing

hey D, as agreed, I've publicly recalc'ed
(spreadsheet )

Beeple: 140 MWh (45% orig estimate)
Pak: 260 MWh (60% ")
TLSOTNS: 74 MWh (62% ")

Will you also post the new numbers? It wouldn't be nice to mislead your followers, implying that I was 10x off.

https://twitter.com/DCCockFoster/status/1352349510069006337

(tbh I don't wanna argue over whether it's 100 transatlantic flights a month, or 50. Kinda misses the point).

Again I want to underline, as I've said many times in this thread & article. This was never an attack on the *artists*. I know nobody knew. & I realise how painful this is for many (clearly not all), & it kills me to bring this out. But platforms must be transparent & accountable

I'm drowning in DMs & replies, but I can already see lots of initiatives, optimistic rollups, platforms already using PoS, more off-chain activity etc. I'm hoping this will all end for the better (otherwise I obviously would not have dove in)

Artists *should* be able to release 100s of digital artworks. There is absolutely no reason that releasing 100s of digital artworks should consume 100s of MWh.

For decades we (I included) have been putting work on vimeo, youtube, tumblr, Instagram etc for free, while the platforms profit from our work. A platform which generates income for the artists is ofc essential. But can we not do better than wasting so much energy in the process?

Many ppl still believe mining is 'mostly powered by renewables / surplus hydro-power in Sichuan". This is simply not true. Misleading propaganda from crypto-lobby. It's tobacco all over again.

imgs:
Sep2020, Apr2020
https://cbeci.org/mining_map 

pp24-27
https://www.jbs.cam.ac.uk/faculty-research/centres/alternative-finance/publications/3rd-global-cryptoasset-benchmarking-study