Vitalik Buterin is shifting the Ethereum scaling conversation away from Layer 2 to the protocol’s core, arguing that the network’s biggest long-term constraints lie in its state tree and the Ethereum Virtual Machine. He says these bottlenecks account for more than 80% of the proving costs as zero-knowledge technology becomes central to Ethereum’s roadmap.
At the heart of the proposal is EIP-7864, which would replace Ethereum’s current hexary Merkle Patricia tree with a binary tree design. The change may sound subtle, but its implications are significant, as binary trees would produce Merkle proofs roughly four times shorter and dramatically reduce verification bandwidth. The new structure would group storage slots into pages, enabling applications that load related data to do so more efficiently, and many dApps could see more than 10,000 gas saved per transaction in some cases.
Buterin also suggested pairing the tree change with more efficient hash functions to accelerate proof generation, making the base layer more prover-friendly and allowing ZK applications to interact directly with Ethereum’s state instead of building parallel systems. Longer-term, he floated a move beyond the Ethereum Virtual Machine toward a RISC-V–based execution engine, which could reduce complexity and better align with modern zero-knowledge proving environments. In the near term, he proposed a vectorized math precompile described as a “GPU for the EVM” to speed cryptographic operations.
Yet not everyone is convinced Ethereum needs deeper-layer changes. Analyst DBCrypto criticized rising abstraction across the roadmap, including new frameworks aimed at addressing rollup fragmentation. Each additional layer, he argues, increases complexity, introduces new trust assumptions, and creates additional potential attack surfaces. The Ethereum Foundation’s Open Intents Framework has been proposed as another path to address L2 fragmentation and ongoing scalability discussions.
Leave a Reply