Original text: (Understanding the debate around based rollups)
Author: Donovan Choy (Blockworks)
Translation: Zombit
Ethereum loves rollups, and recently based rollups (layer-based rollups) are becoming popular.
Why are based rollups special? The key lies in their sequencer.
Current Layer-2 solutions typically use a trusted centralized sequencer to order users' transactions before sending them to Layer-1 for settlement. Based rollups, on the other hand, delegate this execution responsibility to Ethereum Layer-1 validators, a mechanism known as 'underlying sequencing.'
This approach is favored for two main reasons: anti-censorship and interoperability.
1. Anti-censorship
Using Layer-1 as a sequencer ensures the same liveness guarantees as Ethereum Layer-1 blocks and avoids censorship issues that may arise from a centralized sequencer.
2. Better interoperability
Another advantage of based rollups is enhanced interoperability. In recent months, supporters like Justin Drake have emphasized this, calling it 'synchronous composability.' This means that transactions on Ethereum can be synchronously ordered (or bridged) across different Layer-2s as if all transactions were running on the same chain.
In short, smart contracts on based rollups based on synchronous composability can call any contract on Layer-1 with almost immediate certainty within a single block, as if they were all running on the same chain.
The fragmented state
However, the current fragmented state of Rollups results in asynchronous transactions between Arbitrum and Optimism, leading to the issue of fee uncertainty. Fee uncertainty arises because gas fees are calculated in different time slots rather than uniformly calculated within a single Ethereum block's 12-second time slot.
Potential cost savings
In addition to restoring seamless interoperability for Ethereum, based rollups can also bring significant cost savings. Ahmad Mazen Bitar, technical product lead of the Nethermind team, explained:
A user might want to exchange tokens on Layer-1 but also wants to take advantage of deep liquidity pools on Layer-2. With synchronous composability, they will be able to send a single transaction from Layer-1 to Layer-2, execute it, and then return to Layer-1.
Existing based rollups
The largest based rollup currently operational is Taiko, which has seen explosive growth in total value locked (TVL) and daily transaction volume this month.
Other early based rollups are also in trial runs, such as Surge by the Nethermind team and UniFi by the Puffer Finance team, both of which are forks of Taiko.
Challenges and technical dependencies
However, based rollups also have their drawbacks. Since execution (sequencing responsibility) is handed back to Layer-1 block validators, this means the performance of based rollups is limited by the 12-second block time of Layer-1.
Therefore, realizing the advantages of synchronous composability is not easy. Based rollups must complete instant proofs within a single 12-second time slot; otherwise, they cannot quickly execute composable transactions.
To achieve this fast proof, more technical dependencies need to be introduced. However, Brecht Devos, co-founder and CTO of Taiko, is optimistic about technological advancements.
Taiko recently enabled two zero-knowledge proofs developed by Risc Zero and Succinct Labs on its rollup, becoming the first multi-proof based rollup that does not rely on a single trusted party, based on Intel's SGX trusted execution environment (TEE).
'With more TEE, faster and cheaper zkVMs, and available AVS (verification accelerators), proof technology is advancing rapidly. We believe that proof technology with sub-time slot delays will soon be achieved,' Devos said in an interview with Blockworks.
The impact of MEV
Another commonly recognized disadvantage is that based rollups may lose MEV (maximum extractable value) as a main source of revenue due to a lack of centralized sequencers. However, Devos mentioned that there are still other clever alternatives.
Devos stated that on Taiko, 'MEV can be captured by auctioning 'execution tickets' to Layer-1 block proposers.' Therefore, although based rollups default to giving sequencing rights to Layer-1 validators, this is not the only option.
Matthew Edelen, co-founder of Spire Labs, shares a similar view. He explained in a recent Bell Curve podcast:
An auction is not necessarily the only way to allocate sequencing rights. You can allocate 99% of the sequencing rights through an auction, then give the last 1% to a friend or independent staker, which can help present a good image on L2Beat.
Moreover, in the long run, the importance of MEV may gradually decline. This view is based on a simple cost-benefit analysis: currently, most blockchain revenue comes from fees generated by network congestion, rather than MEV revenue. As better MEV solutions emerge, the proportion of MEV revenue is gradually decreasing.
As Justin Drake explained in The Rollup podcast:
The current ratio of congestion fees to contention fees is about 80:20. About 80% of the revenue on Layer-1 comes from congestion fees—approximately 3,200 ETH daily since EIP-1559, while MEV has been around 800 ETH daily since the merge. I believe this ratio will become even more skewed, possibly changing from 80:20 to 99:1.
Conclusion
The advantages of based rollups have returned the user experience of Ethereum to its original state.
There is an irony here, as these advantages are features that blockchain has had since day one. From the inception of the Bitcoin network, synchronous composability and underlying ordered transactions have been the norm.
The separation of execution layer responsibilities is a phenomenon that has emerged in recent years due to the rollup-centric roadmap (or multi-chain model adopted by Polkadot, Cosmos, and Avalanche). Based rollups are prepared to return to the past model.
Source
