Ethereum Proposal Seeks to Adjust Validation Limits for Privacy Transactions

A proposal submitted by contributor AnkushinDaniil aims to modify EIP-8141, the design for Ethereum Frames transactions. The change would transform the existing validation maximum into a common floor, permitting nodes to propagate transactions within that limit while allowing capable nodes to accept more computationally expensive transactions locally.

The current draft of EIP-8141 limits signature checks and execution during the initial validation phase to 100,000 gas. This cap is designed to manage node workload and mitigate denial-of-service risks. However, a September 2 benchmark by mmjahanara indicates that an optimized Groth16 proof verifier requires 190,628 gas, with the cryptographic pairing check alone consuming 181,000 gas.

The benchmark models minimum costs of 211,828 gas for a single-note spend and 351,828 gas for an eight-note spend. These figures include verifier measurements, estimated overhead, and a 20,000-execution-gas charge per nullifier. While the benchmark author recommends a minimum of 250,000 gas for optimized transactions, this recommendation has not been adopted.

The proposal remains under review. Because the current nonce rules still restrict the public mempool to one pending frame transaction per sender, privacy designs face ongoing constraints. While EIP-8141 would allow transactions outside of public rules into local or private mempools, ensuring wider network support for privacy designs like Tornado Cash and RAILGUN would require accommodating higher proof costs across the network.

The benchmark model assumes specific application changes, including the optimization of verifiers, the compression of proof inputs, and the shifting of non-verification work into later frames. The model also notes that using SHA-256 compression increases the computational work required for users to generate proofs on their devices.