Is bitcoin’s energy efficiency all we think it is?

Faith is a powerful drug. It is easy to believe that computer technology and electronics will make the world more efficient and reduce its dependence on fossil fuels, not least by reducing how much of that energy must be carried out by physical things that work in a wasteful way. But this is the same set of core technologies that gave us bitcoin.

In a demonstration of how perverse incentives can be in economics, bitcoin and the other blockchain protocols that evolved around it fell into an arms race that makes just burning currency for heat look vaguely sensible.

The first miners used precisely the Intel processors on desktop PCs. But that quickly changed as the value of bitcoin and interest in mining increased. The protocol behind the cryptocurrency was designed to increase the amount of work needed to calculate the hash codes that guarantee the authenticity of the blockchain in order to maintain a more or less constant rate of bitcoin creation. People armed with GPU cards capable of crunching the hash faster quickly supplanted the x86s. After a while, GPUs gave way again to teams that could deploy arrays of programmable logic chips, each running custom hash-generating algorithms. In the end, even those on custom chips lost out: owners traded the millions of dollars it took to develop and manufacture them for the lower power efficiency and hash rates of GPUs and programmable logic.

This arms race has made it difficult to estimate the pollution generated by bitcoin. Miners have their own accelerators, each connected to power grids with varying degrees of dependence on coal and gas. Some researchers have argued that bitcoin miners are even moving their operations around depending on the availability of cheap electricity, with some apparently shifting between hydropower-rich areas during the Chinese rainy season to inland provinces closer to Russia where coal dominates at other times.

To be fair, some of the more lurid claims about energy consumption may be inflated. To address this issue, a team from the University of Southampton used machine learning to try to reduce uncertainty in a project that found a previous mining claim generated around 70MtCO₂ei 2017 was overestimated. The bad news was that the trajectory of bitcoin mining puts it on course to rise from just under 15MtCO₂ei 2019. Although a smaller number, this is still more than the total emissions from Bolivia and similar small states.

Despite the amounts of air conditioning needed to keep racks of accelerators cool enough to keep running, energy is probably not the biggest cost for miners. The faith has ended up costing them much more.

According to work published late last year by Charles Bertucci and colleagues working at a group of Parisian universities, the energy still ends up being a relatively small part of the cost even though the energy required can now be as much as 100TWh each year, which works out to be a third of the total consumption of large non-crypto data centers worldwide in 2020.

The conclusion Bertucci’s group reached about why energy is responsible for less than a quarter of bitcoin miners’ costs is that they believe they need to keep buying new hardware to stay in the mining game. Even if the miners have collected a significant amount of bitcoin, as with the gold rush of the 19th century.th and 20th For centuries, hardware manufacturers have profited from crypto. The implication is that miners are dependent on further increases in the value of the tokens they hold, as well as those they will win in the future to justify their continued purchases. Given bitcoin’s collapse in price in tandem with falls in the US and other stock markets, this belief may have taken a hit. But at the moment, many of those in the crypto world believe that it is only a temporary setback in a continued climb up the mountain of monetary value.

Measuring the net effect of bitcoin, provided we have the right power consumption figures in the first place, is relatively easy. That’s because, for the most part, it doesn’t replace much that is physical. You could argue that it replaces physical coins and notes, which require resources to mint even though the former are easily and routinely recycled. In reality, it is mostly replacing other electronic transactions that managed to go quite efficiently and have done for decades.

Figuring out the effects of other technologies is a more difficult task as explored in the last issue of E&T in this piece about the problems facing large-scale AI models meant to run the world better and how rebound effects can occur, where the convenience of digital activity causes the overall level of energy or resource use to increase. The rebound effect for automated driving is quite clear to see, but difficult to quantify. Others can be much more subtle. An example is video conferencing.

The principle is, as we have seen with the Covid pandemic, that online conferences can quite well replace travel which may well generate a lot more carbon. It can also generate rebound effects as it becomes easier to have geographically separate teams that then travel to see each other for more in-depth meetings, increasing carbon from air travel. The research so far tends to indicate that video conferencing results in less travel overall, although researchers such as Kelly Widdicks, a lecturer at Lancaster University, believe much more is needed.

If we bring in the possible evolution of video conferencing into virtual reality and the metaverse, the resources needed for the electronics may make it even less of a clear trade. There are limits to how much a headset can do before it gets too hot to wear, but the industry is looking to offload much of the processing to local servers and the cloud. Given the level of detail that full VR needs compared to what we currently accept from video conferencing, it will likely result in increases in both electricity and emissions: from all the silicon and other high-precision devices needed throughout the network.

This week, ABI Research published a report that underscored how much change a full metaverse implementation will need at all levels of the network. And, as the ABI analysts point out, crypto could end up embedded in these applications through non-fungible tokens (NFTs) and similar digital money and ID devices. Even as Ethereum and other blockchains move to less energy-intensive protocols, supporting NFTs directly instead of other, potentially more efficient transaction mechanisms could turn the metaverse into a major power hog.

As with AI, there is a clear need to measure how much these technologies need in terms of resources. But even with greater transparency, there’s a lot that can be missed since the effects are so subtle given that we can’t be sure how much metaverse sessions will balance against flights or other high-resource activities.

Widdicks and colleagues argue that the solution may lie in the expanded use of carbon pricing, as long as it is used over the lifetime of the systems used to implement these services. An ideal benefit of carbon pricing is that it allows the market to drive towards a minimum rather than relying on measurements that may not be balanced that way, and which in turn leads to perverse outcomes. “We will still need transparent evidence of ICT emissions and their reductions as we align the digital sector, and global sectors in general, with the Paris Agreement,” Widdicks added.

Is it too late to start? A hiatus in both the AI ​​and metaverse now seems possible as FAANG companies batten down the hatches in anticipation of a downturn or recession, providing some breathing room. However, there is a clear danger that the next wave could get underway before governments can agree on how to tackle the problem, not least because an extended carbon pricing regime would be politically difficult territory and measurement itself could look much easier. implement even if it is far less than optimal. The first step is to change beliefs. Digital is not necessarily the efficient option. It depends on how you go about it.