A16z crypto: There is a theoretical lower limit of "5 rounds of delay" for blockchain anti censorship, and Ethereum is exploring new mechanisms According to a report by Golden Finance, on May 21st, a16z crypto researchers wrote an article pointing out the fundamental contradiction between blockchain's "censorship resistance" and "low latency". They pointed out that any Byzantine Fault Tolerant (BFT) blockchain protocol with censorship resistance requires at least 5 rounds of communication in cases where more than one-fifth of validators may commit wrongdoing, while traditional BFT consensus only requires a minimum of 3 rounds. In traditional BFT protocols, the block proposer holds both the power of block construction and consensus promotion, allowing for the exclusion of specific transactions for review. This is also the root cause of many MEV issues. To address this issue, Ethereum is studying FOCIL/EIP-7805, while Solana is researching mechanisms such as Constellation and MCP. The core idea is for validators to collect non negligible transaction "inclusion lists" in advance before the block is formally proposed. In systems with anti censorship guarantees, transactions will be included in blocks no later than 5 rounds of communication, making transaction confirmation time more predictable.