I just saw the details of what Vitalik is planning for Ethereum, and honestly, it's quite ambitious. The guy is talking about two fundamental changes to the execution layer that could transform how the network operates.

First is the topic of the state tree. Currently, they use hexadecimal MPT with Keccak, but the idea is to migrate to a more efficient binary tree structure. It sounds technical, but the important thing is that this would reduce Merkle branches by 4 times and speed up proofs by between 3 and 100 times according to EIP-7864. Additionally, it significantly lowers the cost of accessing adjacent storage slots. That is, less overhead, more speed.

Next is the EVM. Vitalik proposes gradually replacing it with RISC-V. The advantage is that RISC-V offers better execution efficiency, superior compatibility with proofs, and a cleaner protocol design. It’s not a change overnight; they plan to do it in three phases: first with precompiles, then allowing users to deploy contracts with the new VM, and finally converting the EVM into a smart contract on the new architecture.

The interesting part is that all of this aims to address Ethereum’s historical bottlenecks. Better network efficiency, more robust proof capabilities, and laying the groundwork to scale into the future. If this is executed well, it could significantly change the game.