Abstract#
This article compares the main Ethereum Layer 2 solutions in terms of user experience, team and funding, community development, and on-chain data.
Optimistic rollups have gained a significant advantage by integrating with the Ethereum ecosystem, attracting users and funds.
The progress of zk rollups has been relatively slow due to the high difficulty of implementation. Its long-term technical advantages, such as reduced withdrawal time and transaction fees, are not significantly better than optimistic rollups.
There are three challenges for zk rollups: user experience, migration cost of the existing Ethereum ecosystem, and difficulty in developing new projects.
In my opinion, zk rollups have limited opportunities to compete in the general-purpose L2 field under the current landscape. However, if they can seize the next bull market trigger point, there is a chance for them to surpass the competition.
Information Compilation#
I have only had fragmentary knowledge about the progress of the four major rollups in Ethereum L2 and have not organized it systematically. This week, I have compiled some information and gained a framework understanding of these rollups. Let's start with the images:
Rollups Overview#
In a nutshell, rollups are about off-chain computation with on-chain data. They store data on L1 so that anyone can detect fraudulent transactions, withdrawals, and reconstruct transaction history. The data stored on L1 is highly compressed, resulting in a significant reduction in gas fees for individual transactions. Each batch of transactions submitted to L1 forms a rollup chain.
There are two main types of rollup technologies: optimistic rollups and zk rollups.
Optimistic rollups do not validate the data when uploading it to L1 but allow others to submit fraud proofs within a time window to prove the presence of erroneous transactions in the batch. Once erroneous transactions are proven, the nodes that submitted the batch will be penalized, and the proposer of the fraud proof will be rewarded, and the transaction records will be rolled back to before this batch. Therefore, it is called "optimistic" because it assumes that not everyone is malicious, and at least one good actor is monitoring and correcting data errors to ensure the correctness of the rollup chain based on this premise.
Zk rollups already come with validity proofs when uploading data to L1, so they do not require a time window.
For detailed differences, refer to Vitalik's article: https://vitalik.ca/general/2021/01/05/rollup.html
Comparison of the Four Rollups#
Arbitrum and Optimism are optimistic rollups.
zksync and StarkNet are zk rollups.
Let's compare them in terms of technical advantages and disadvantages, user experience, team and funding, community development, and on-chain data.
Technical Advantages and Disadvantages#
Due to the existence of the challenge period in optimistic rollups, the withdrawal time is longer, usually 7 days, while zk rollups only need to wait for a batch. However, most optimistic rollups chains now provide third-party withdrawal services for instant withdrawals, so there is not much difference in practical experience between the two technologies.
In addition, the advantages of optimistic rollups are low off-chain computation costs and low EVM compatibility difficulty. The advantages of zk rollups are low on-chain transaction gas fees.
However, zk rollups have a major drawback, which is the high technical difficulty. On the other hand, optimistic rollups require fewer modifications and have simpler technical implementations. This will be discussed in detail later.
User Experience#
In conclusion, Arbitrum and Optimism provide a user experience consistent with Ethereum, as they can be used with MetaMask, which aligns with the habits of ordinary users. zksync and StarkNet, on the other hand, have their own issues and provide a mediocre user experience.
With zksync, users can only view assets and transaction records using a web wallet and cannot view them in MetaMask. When making transactions, MetaMask only needs to sign instead of sending transactions, so it feels like authorizing a website to access wallet information on Zapper. However, after signing, the coins have already been sent, and users need to go back to the web wallet to view balance changes and transaction history, which feels very strange.
StarkNet requires the installation of the Braavos wallet plugin for transactions, which provides a better experience compared to zksync.
In terms of transaction fees and confirmation speed, StarkNet is expensive and slow. The swap fee is around 0.5 USD, and the confirmation speed varies, sometimes taking 1 minute and sometimes requiring 3 minutes. It takes more than 1 hour for bridge transactions. The other three solutions have similar transaction fees, which are relatively cheap and provide instant confirmations.
Team and Funding#
Among the four projects, StarkNet stands out with its well-known founders who have academic backgrounds and entrepreneurial experience. It has raised over 200 million USD and has a valuation of 8 billion USD, with investment from both blockchain and traditional venture capital firms such as Sequoia and Intel.
Arbitrum has three founders with good educational and professional backgrounds. It has received investments from around 10 institutions, mainly from within the blockchain industry, and has raised around 100 million USD.
Optimism, formerly known as the Plasma Group in the Ethereum community, has a close relationship with Ethereum. Although the founders do not have impressive resumes, they have the inherent Ethereum community genes and gained support from leading DeFi projects. It has raised 175 million USD with investments from a16z and Paradigm.
zksync is relatively mysterious. There is limited information about its public funding, with the most recent investment being 50 million USD from a16z, and its valuation is unknown. The team consists of about 40 people.
Overall, all four projects have sufficient funding, so there are no immediate funding issues.
Community Development#
Let's look at the B-side and C-side separately.
For the B-side, we can look at the number of protocols. Arbitrum and Optimism have 113 and 71 projects, respectively, with a significant portion being migrated from Ethereum. zksync and StarkNet have 16 and 27 projects, respectively, but some of them are not very active.
Arbitrum and Optimism have gained a significant advantage by integrating with the Ethereum ecosystem, attracting users and funds.
This clearly shows the difficulty of zk rollups in reusing the existing Ethereum ecosystem due to the underlying technical issues. zksync has uncertainties in its development progress due to the need to develop zkEVM to be compatible with Ethereum smart contracts. StarkNet has developed its own programming language, Cairo, which means starting from scratch in terms of infrastructure. This results in high costs for ecosystem development on the B-side, including high integration costs for existing wallets, DApps, and information service websites. Additionally, new ecosystem developers cannot leverage existing knowledge and need to learn new technologies.
On the C-side, Arbitrum has attracted many users to its ecosystem through airdrops and Odyssey events, achieving good results with a Discord user count of 223,000. StarkNet, on the other hand, has the fewest user followers and less attention.
On-Chain Data#
The lock-up and transaction volumes of optimism and arbitrum are comparable, while StarkNet has very little on-chain activity and minimal transaction volume.
Expectations#
Vitalik's view is that zk rollups will win in the long run due to their shorter withdrawal time and the expectation of significantly reduced transaction fees in the future. However, it does not mean that the best technology will always dominate the market. I remember in one of my university courses, the professor mentioned the format war of music files: mp3 was not the best audio format on the market, as wma had a much higher compression ratio, but mp3 still became the mainstream.
There are three challenges for zk rollups: user experience, migration cost of the existing Ethereum ecosystem, and difficulty in developing new projects. Currently, progress is not rapid. If we want to bet on general-purpose L2 solutions, optimistic rollups have a better chance of winning.
Among the currently successful chains, except for Solana, all have rapidly accumulated users by copying the Ethereum ecosystem and relying on high yield expectations during the bull market. Solana succeeded due to a combination of SBF's funding and Solana's own technology, which is difficult to replicate.
In my opinion, zk rollups have limited opportunities to compete in the general-purpose L2 field under the current landscape. However, if they can seize the opportunity and optimize C-side products, complete initial B-side application development, and establish user incentive mechanisms, there is a possibility of surpassing the competition.