There’s been some coverage of recent comments by Ethereum creator Vitalik Buterin saying that the second-largest blockchain can’t get any faster than it is now.
Which seems to contradict what he and all the other top developers on Ethereum 2.0 have been saying for years: That Ethereum 2.0 will be an order of magnitude more scalable than the current blockchain, which has been groaning under the weight of its own success for several years.
See also: PYMNTS Blockchain Series: What Is Ethereum? The Blockchain That Moved Crypto Beyond Currency
So, which one is it? Is the highly touted Ethereum 2.0 stuck at the same speed?
Technically, yes. In reality? No, not at all.
In the payments industry, speed tends to refer to transactions per second, or TPS. In blockchain, however, “scalability” is the term used to account for TPS. Speed refers to something else entirely: block time.
Read more: Bitcoin’s 10-Minute Block Time Batches and Fluctuating Transaction Fees Give RTP a Leg Up
When Buterin talked about “the limits on making block time faster” in a recent Reddit discussion thread — which he reposted on Twitter — he was referring to how fast new blocks of transactions are added to the Ethereum blockchain. And that’s a matter that’s more about security than scalability.
Ethereum’s block time is currently about 13 seconds. And while that’s a whole lot better than Bitcoin’s 10 minutes, Solana, one of the current “Ethereum killer” blockchains seeking to steal away developers and projects, has a 400 millisecond block time — so a new block is added every 0.4 seconds. It also can handle 50,000 TPS.
Read more: PYMNTS Blockchain Series: What Is Solana?
But the ongoing project to convert Ethereum into Ethereum 2.0, built on the greener and far more scalable proof-of-stake consensus mechanism (as opposed to Bitcoin-style proof-of-work mining) will enable it to reach 100,000 TPS.
Read more: PYMNTS Crypto Basics Series: What’s a Consensus Mechanism and Why Is It Destroying the Planet?
Now, block speed is very important to payments, but what it effects is transaction time — latency — and final confirmation of the transaction, which is all about security. For all of their vaunted immunity to changing information once it has been recorded, blockchains transactions are vulnerable for a short time after being added to a block. You can see the details in the link below.
See more: The 51% Attack: Crypto’s Double-Spending Achilles Heel
The Blockchain Trilemma
“Trilemma” is a word coined by Buterin to describe a problem with the three core features of a public blockchains: They are decentralized, they are secure, and they are scalable.
The dilemma part of the word trilemma refers to the widely accepted theory that you must always sacrifice one of those three to strengthen the other two — thus making blockchain design a balancing act.
The way to increase decentralization is by adding more nodes — which have full records of all the transactions on a blockchain. But that means more people who are vying to win the right to validate transactions and add them to the blockchain in exchange for a reward of newly minted coins and transaction fees.
The way to make a blockchain more secure is to spend more time and effort validating transactions. They must all agree — reach a consensus — that a newly written block is accurate; nodes that disagree are forked into separate blockchains. And that takes time.
Adding scalability means adding more transactions to the blockchain, either by speeding up the block time or increasing the size of each block.
Here’s the trilemma: The more nodes you have, the more decentralized and secure your blockchain is, but it takes longer for all of them to reach a consensus, hindering scalability.
Similarly, increasing scalability and decentralization means cutting into the security, as you have to do more validation more quickly, and security will suffer.
To be more secure and scalable requires reducing the number of nodes and making the blockchain more centralized, as fewer people can reach consensus faster without shortchanging the validation process.
Proof-of-work blockchains like Bitcoin and the current Ethereum sacrifice scalability, while proof-of-stake blockchains are less secure — but can be more scalable.
Buterin’s point is that by adding new blocks more quickly than Ethereum currently does, Ethereum 2.0 would give up too much security.
How Ethereum 2.0 will scale
Ethereum 2.0’s proof-of-stake consensus mechanism “attempts to give blocks a very high level of confirmation [meaning consensus of validity] and this requires thousands of signatures. … This process incurs latency and takes time.”
The way Ethereum 2.0 will get around the trilemma is by sharding — meaning creating new sidechains that run in parallel with the main blockchain, splitting the load. This lets it process more transactions — vastly more — without reducing the number of nodes or rushing the validation security process.
