- Two successive successful 51% attacks on Ethereum Classic.
- A new, more realistic estimate of Bitcoin's energy usage; it is only as much as Belgium
Follow me below the fold for details and commentary.
Energy ConsumptionSince Blockchain: What's Not To Like?, I've been working with Tim Swanson's 2018 estimate that the top 5 cryptocurrencies use as much energy as The Netherlands. Swanson's estimate covered Bitcoin, Bitcoin Cash, Ethereum, Litecoin and Monero. Swanson estimated that just powering the Bitcoin mining rigs used ~50TWh/year, but:
this estimate is probably a lower-bound because it doesn’t include the electricity consumed within the data centers to cool the systems, nor does it include the relatively older ASIC equipment that is still turned on because of local subsidies a farm might receive.I regret not noticing a considerably more detailed 2019 estimate by Stoll, Klaaßen & Gallersdörfer, The Carbon Footprint of Bitcoin, which concluded that:
as of November 2018, the annual electricity consumption of Bitcoin had a magnitude of 45.8 TWh.Swanson's Bitcoin analysis cited a 2018 estimate from Alex de Vries that Bitcoin alone used as much energy as Austria. Now, in Bitcoin’s energy consumption is underestimated: A market dynamics approach Alex de Vries shows that:
most of the currently used methods to estimate Bitcoin’s energy demand are still prone to providing optimistic estimates. This happens because they apply static assumptions in defining both market circumstances (e.g. the price of available electricity) as well as the subsequent behavior of market participants. In reality, market circumstances are dynamic, and this should be expected to affect the preferences of those participating in the Bitcoin mining industry. The various choices market participants make ultimately determines the amount of resources consumed by the Bitcoin network. It will be shown that, when starting to properly consider the previous dynamics, even a conservative estimate of the Bitcoin network’s energy consumption per September 30 (2019) would be around 87.1 TWh annually (comparable to a country like Belgium)
|De Vries Fig. 1|
To evaluate the accuracy of the estimates for Bitcoin’s energy demand and their respective assumptions, we propose dividing the developments in the Bitcoin mining market in three different stages based on the estimated network hashrate. These stages are growth, stability and decline (highlighted in Fig. 1). The growth stage is identified by observing the increasing amount of computational power (hashrate) in the network. Likewise, stability and decline can be identified by a relatively flat or decreasing amount of hashrate.
Declining Hashratede Vries shows that:
This shows that in a declining market real-world mining facility efficiency rapidly approaches the theoretic optimum, as suboptimal choices by market participants are increasingly punished. It also shows that during this phase a simple back-of-the-envelope approach (though corrected for market share) can provide a good sense of direction when estimating the electricity consumption of miners.Stoll et al's estimate coincided with the sole period of declining hashrate, so their estimate of 5.23GW was close to one using Swanson's methodology, 4.36GW. Because the proportion of older, less efficient devices falls rapidly during periods of declining hashrate, basing estimates on the most recent, most efficient devices works well.
Unfortunately, we have to establish that most of the time the situation is a lot more complicated, as a declining market only occurs once (and only briefly) over a period of 33 months in total.
Growing HashrateFigure 1 shows that, except for four months in late 2018, the hashrate has been growing since September 2017. Thus it is most important to analyze energy usage during periods of increasing hashrate:
Here we can immediately notice that the link with miner earnings is not as obvious as for a declining market. In fact, the Bitcoin mining industry is booming for nine months after the crash of Bitcoin miner earnings early 2018 (also confirmed by record device production during the entire first half of 2018 ). The simplest explanation for this is that, while cutting losses in a declining market merely involves switching off devices, taking advantage of growth opportunities typically requires the acquisition of new devices. This takes time, and even more so if devices are only available in limited quantities due to production constraints.Thus, because the proportion of older devices falls only slowly during periods of growing hashrate, basing estimates on state-of-the-art devices underestimates Bitcoin's energy demand during these periods. And because the hashrate is growing most of the time, this underestimate is significant.
de Vries analyzes the market during 2019 in Section 2.2.2. The first half of the year saw the introduction of new, more efficient devices from Bitmain (Antminer S17 with ~40% of the power draw of the S9) and Caanan (A10 with ~70% of the power draw of their A9). But supplies of the chips were very constrained, so the prices for the new devices were very high:
With an average selling price of 7038 RMB per unit (compared to 1008Thus the older devices continued to dominate newly introduced hashrate. Caanan:
RMB and 1526 RMB for the A8 and A9 series respectively) the [A10] was sold only 490 times during the first half of 2019.
sold 252,862 units of its older A8 and A9 series ... in the first half of 2019, as compared to 292,826 units in the second half of 2018. These sales (representing 22 percent of all computational power sold in the first half of 2019) reflect a steady demand for older device types in 2019These large lags in responding to the introduction of new devices are very significant:
From the total increase in the Bitcoin network hashrate of around 30 EH/s during the third quarter of 2019, about 25 EH/s can be attributed to the sale of newly produced devices. ... During the first half of 2019 we observe an increase of around 20 EH/s in the total network hashrate, of which almost 18 EH/s can be attributed to the total computational power sold during this period. That means that, out of the average estimated network hashrate of 92.5 EH/s per September 30 (+/− 4 days), at least 49.5 EH/s of computational power must be generated by devices produced prior to 2019.In other words, despite the introduction of new devices, over half the hashrate is due to devices more than 9 months old.
Conclusionde Vries concludes:
By analyzing sales and market shares data instead, we conclude a conservative weighted average power efficiency of all mining units in the network amounts to 0.0917 J/GH per September 30, 2019 (assuming only the most power-efficient devices produced prior to 2019 are still active). This translates to an annual power consumption 85.8 TWh after performance 5 and PUE corrections ... This number increases to 87.1 TWh annually if we consider a small delay (of one week) before delivered mining machines become active, as this primarily affects the most efficient devices being considered during the third quarter of 2019. ...The one-week delay is due to the time it takes a miner from receiving new devices to having them in production.
To put this number into perspective, it represents close to half of the current global data centre electricity use (200 TWh ), while equaling the electricity use a country like Belgium (87.9 TWh) .
The bottom line is that a major country's electricity supply is devoted to supporting pure speculation, massive fraud, rampant theft, money laundering and other crimes.
51% AttacksAs I wrote in Proof-of-Stake In Practice:
At the most abstract level, the work of Eric Budish, Raphael Auer, Joshua Gans and Neil Gandal is obvious. A blockchain is secure only if the value to be gained by an attack is less than the cost of mounting it.Back in January last year, this observation was made obvious. First, Dan Goodin reported that:
Attackers have stolen almost $500,000 worth of the Ethereum Classic digital currency by carrying out a compute-intensive hack that rewrote its blockchain, officials with Coinbase, one of the leading crypto currency exchanges, said on Monday.Then Catalin Cimpanu reported that attacking ETC cost $4903/hr and thus:
According to coinmarketcap.com, ETC is the 18th biggest cryptocurrency, with a "market cap" of $523,350,707. But that's not enough to keep it secure. Only a few of the biggest altcoins have enough mining power relative to their "price" to deter 51% attacks.
Coinbase also updated its original report with details on another 12 double-spend attacks, bringing the total of stolen funds to 219,500 ETC (~$1.1 million)."In Why the Ethereum Classic hack is a bad omen for the blockchain, Russell Brandom explained the rash of 51% attacks and quoted Nicholas Weaver:
As Weaver puts it, it’s “a nice illustration of how proof-of-waste schemes cannot be both efficient and secure.” The more it costs to mine a block, the more expensive it is to outspend the honest miners for long to reverse a transaction. Electricity prices vary from miner to miner, but Weaver estimates that the Bitcoin network currently runs through about $300,000 in electricity each hour, while the smaller Ethereum network runs at roughly $100,000 per hour. For Weaver, any coin much smaller than that is at risk of a 51 percent attack. Ethereum Classic clocks in at roughly $5,000 per hour.But, as always, the answer to "is our children learning?" is no. 18 months later, ETC is still not wasting enough electricity to be secure. David Gerard reports that:
There’s still nobody who cares about Ethereum Classic, leaving the blockchain open to a 51% attack on 31 July — in which someone spent $192,000 on hashpower to mount a double-spending attack that netted them 807,260 ETC, notionally worth $5.6 million. And another attack on 6 August. [Bitquery; CoinDesk]Alexsey Studnev reports on the first attack:
Attacker double-spent 807,260 ETC ($5.6 million) during this attack and spent 17.5 BTC ($192K) to acquire the hash power for the attack. The attacker also got 13K ETC as a block mining reward, which we are not including in our double-spent calculation.So an investment of $192K returned around $5.7M, almost 30x. That is a return even the "vampire squid" would not turn up their nose at.
Sebastian Sinclair reports on the second attack:
Ethereum Classic has suffered its second 51% attack in a week after more than 4,000 blocks were reorganized Thursday morning.
Mining pool Ethermine’s parent entity Bitfly and crypto exchange Binance reported the reorganization, announcing all Ethereum Classic payouts, withdrawals and deposits had been suspended due to the attack.
|Bitcoin mining pools 8/7/20 |
Both Bitcoin and Ethereum mining are very centralized, with the top four miners in Bitcoin and the top three miners in Ethereum controlling more than 50% of the hash rate.It is impossible to know whether or not these pools are conspiring together. For more discussion of this problem, see the update to Cryptographers on Blockchains: Part 2.