PoS is more energy-efficient than PoW, but is it better?
We’ve all heard that Bitcoin mining consumes as much energy as an entire country. Why is it so inefficient compared to Proof-of-Stake networks and to HBBFT blockchains, such as DMDv4?
Some years ago, people were shocked to find out that Bitcoin mining used up as much electricity as all of Ireland or Czechia. Now, according to Cambridge researchers, it uses more than all of Argentina — 121 terawatt-hours, to be precise. Partly this is due to the inflow of miners attracted by the growing BTC price: back in October 2020, when one BTC was still worth $11,000, Bitcoin’s annualized consumption stood at 67 terawatt-hours. Still, 67 TWh or 121 Twh, there’s no denying that Bitcoin was designed to devour resources — here’s why.
Proof-of-work: resource-hungry by design
As you may know, in PoW networks miners compete to solve mathematical puzzles. They do earn transaction fees, too, but that’s not what a miner is after: the real prize is the 6.25 BTC (over $300,000 at the current price) that you get if you add a new block to the blockchain.
Here we won’t go into the details of nonces, hashes, and block headers. Suffice it to say that the puzzles are designed in such a way that the solution takes a huge amount of computational effort to find, but very little to verify. Whenever a mining device (ASIC) thinks it found a solution (which happens thousands of times a second), it broadcasts it to the network, and other miners verify it. Once everyone agrees that the solution and the new block are valid, the lucky node gets the prize.
The problem is that everyone is working on the same puzzle at the same time. All that effort spent by those who don’t get the reward is essentially wasted.
PoS is 99% more efficient, but is it secure?
In Proof-of-Stake, the system picks a small set of validators to verify transactions and form new blocks for a set period of time (called epoch). There is no competition, no race to solve the block first. The others just wait on the sidelines, waiting for the algorithm to pick them next.
Once you take the cryptographic puzzles out of the equation, the need for electricity falls dramatically. Processing transactions isn’t a resource-intensive process in itself; that’s why PoS minting (validation) can be done on a regular laptop or even smartphone. In fact, it’s been calculated that Ethereum will reduce its energy consumption by 99% once it fully transitions to PoS.
At the same time, you are left with a problem: how to make sure that all validators behave? Once again, we won’t delve into the intricacies of PoS security here, but the basic answer is that you make them stake their own coins. This way, they literally have a stake in the game, and anything that harms the value and security of the network will harm the value of their own assets.
Some say that PoS networks have a much lower cost of attack, but this is incorrect: acquiring enough tokens to interfere with the network is extremely difficult and expensive. Rather, the issue is that the validators with the largest stakes will accumulate all the minting and voting power.
The verdict
So, does all this mean that PoS is better than PoW? Well, if you judge the effectiveness of a cryptographic defence by the amount of effort that it takes to break through it, then PoW can seem like the best of them all. Indeed, it would take an impracticably huge amount of energy to mount a successful attack on Bitcoin.
However, the Bitcoin network is only invulnerable because it’s so big. Smaller PoW blockchains often fall victim to 51% attacks — as it happened to Ethereum Classic, Firo, Bitcoin Gold, for example.
To say that a huge Proof-of-Work network is ‘better’ or more secure than an average Proof-of-Stake network doesn’t mean much. You have to look for a consensus mechanism that offers the optimal compromise for networks of all sizes, including those that are only starting to scale. From this point of view, PoS is a much better solution: you still have good (if not perfect) security which is why Vitalik Buterin initiated Ethereum’s move from PoW to PoS. By the way, he once called PoW’s burning of resources ‘a tragedy’.
How DMDv4 improves on PoS: HBBFT + dPoS
PoS is still a competitive consensus model — and any competition implies a waste of resources to some extent. That’s why DMDv4 goes even further and implements the world’s first cooperative consensus engine that uses smart contracts (see here for the difference between competitive and cooperative consensus).
This allows DMDv4 to increase energy-efficiency and ensure decentralization and security at the same time. HBBFT (Honey Badger Byzantine Fault Tolerant consensus) does away with set block times: it’s an asynchronous consensus model, where the creation of a new block can start just one second after the previous one. There are no empty or orphan blocks, and transactions become final as soon as 2/3+1 one of the validators give their approval. All this means that no energy is wasted.
Moreover, our implementation of HBBFT together with dPoS (delegated Proof-of-Stake) also guarantees a high level of security. Even if 1/3 of the nodes become malicious, the consensus will still work. Meanwhile, our staking and node election mechanics ensure that it’s in the interest of every validator to play by the rules, and the fact that a validator’s maximum stake is limited to 50,000 DMD means that there will be no excess concentration of influence, i.e. no centralization. You can read more about our consensus mechanism in this article.
We believe that Proof-of-Work was the necessary first step in the evolution of blockchains, but that it has to be superseded by Proof-of-Stake. We owe this to our planet and the environment, and there are surely many industries that can use all that energy much more effectively and for a more productive purpose.
But while Proof-of-Stake requires up to a hundred times less energy, it isn’t a perfect alternative, because it’s subject to centralization risks. That’s why we went a different way with DMDv4, choosing a combination of HBBFT and staking/delegation. If you’d like to learn more, check out our Light Paper.
