Just when I was thinking about where Ethereum layer2 will go next, many overseas communities have already voiced "Make Ethereum L1 Great Again". Indeed, in the past, with the development strategy centered on Rollup-Centric, the market placed too much of Ethereum's lifeline on layer2. If L1 and L2 are separated, what will happen from the perspective of Ethereum's own performance optimization?
1) @VitalikButerin previously said that Ethereum is the overall roadmap of Rollup-Centric. He first used the Cancun upgrade EIP-4844 to reduce fees to incentivize layer2, and then proposed to increase the interoperability between layer2s, and even weakened or even changed the original Sharding strategy to support layer2.
Taking Rollup layer2 as the strategic focus can carry higher TPS, allowing more high-frequency trading application scenarios such as games to be implemented in the Ethereum ecosystem. This development strategy actually gives up the role of Ethereum Execution layer and only provides important functions such as security, consensus, and settlement. After all, due to the limitations of EVM's data storage structure, node client hardware performance, block size and speed, etc., developing layer2 as the execution layer is definitely better than Ethereum L1 itself.
But when the gas fee of L1 drops below 1Gwei and the development of L2 temporarily fails to meet expectations, some people will think about refocusing on the Ethereum mainnet as the execution layer and only letting layer2 provide icing on the cake. (Subtext: The price of layer2 itself is not good, but it cannot affect Ethereum itself.)
2) This makes sense. In fact, in addition to the Cancun upgrade that directly benefits layer2, the focus of Ethereum's subsequent roadmap, including Verge and Purge, is Ethereum itself.
For example: the Verge upgrade integrates the Verkle tree and uses a "stateless client" to optimize node storage costs, while the introduction of Fully SNARKed Ethereum and ASIC hardware acceleration plans are both aimed at optimizing storage costs, improving execution performance, and thereby increasing the scalability of Ethereum L1.
SNARKs can significantly reduce the amount of data stored and transmitted, and improve the efficiency of verification and complex calculations. ASICs can reduce the loss of SNARK execution and improve execution efficiency by improving hardware performance. It should be said that after providing a series of infrastructure guarantees for layer2, Ethereum's next upgrades will be more focused on optimizing the performance of its own chain components.
3) However, the Ethereum L1 Execution layer can carry more transactions and the strategy of developing Layer2 does not conflict. To be precise, layer2 and layer1 are running in parallel.
On the Layer2 side, @arbitrum, @Starknet, @MetisL2, @megaeth_labs, etc. are all continuing to cultivate the ecosystem based on their respective leading advantages and development priorities. I won’t go into details here, but it’s far from being a pessimistic view, so please be patient.
On layer 1, in addition to Ethereum’s next few major upgrades around its own performance, technological breakthroughs in two key areas, ZK underlying framework protocol and ZK hardware acceleration, will also directly help Make L1 Great Again:
For example: @ProjectZKM uses the ZK general protocol based on the zkMIPs microprocessor instruction architecture. In addition to bringing the Bridgeless Native asset cross-chain solution to the Bitcoin ecosystem as mentioned above, the ZK general protocol allows Ethereum to access the liquidity of more homogeneous or heterogeneous chains as a "settlement layer". This means that other chains can interact with the Ethereum chain through the ZK protocol, and then input more key settlement transactions and liquidity to the Ethereum mainnet, which naturally strengthens Ethereum's "Execution" performance;
Another example: @cysic_xyz is committed to the large-scale adoption of ZK technology through chip-level hardware acceleration solutions such as FPGA, GPU and ASIC. As a benchmark project in the ZK emerging and hardware acceleration track, Cysic's goal is to fully ZK the underlying layer of Ethereum in the future. After Ethereum is fully SNARKed, the corresponding Ethereum nodes will also enhance hardware performance, and hardware acceleration is an inevitable choice.
The end.
In short, whether focusing on promoting the L2-centric strategy or further optimizing L1 performance, it is not an antagonistic relationship of one gaining at the expense of the other, but rather a long-term coexisting competitive and cooperative relationship.
In the short term, layer2 is still the main theme of Ethereum expansion, while in the long term, Ethereum, based on ZK underlying and hardware acceleration, as a secure consensus settlement layer, can also carry and achieve expansion capabilities that are full of imagination. Ethereum does not need Great Again, Ethereum has always been Great.
