Technological debt is overwhelming, Ethereum chooses to "start over" with RISC-V.

Ethereum plans to replace EVM with RISC-V to address performance bottlenecks in ZK's future and position itself as the underlying trust layer for next-generation networks, realizing the vision of verifiable computing. This article is from an article written by jaehaerys.eth, collated, compiled and written by TechFlow. (Synopsis: Ethereum "changed RISC-V" to scare away developers? OG warning: ETH ecosystem will be redistributed, small projects will leave Solana) (background supplement: technology》What is RISC-V advocated by Vitalik? Why CKB-VM Why RISC-V? By embracing RISC-V, Ethereum not only solves its own scalability bottleneck, but also positions itself as the foundational trust layer for next-generation networks. Ethereum is preparing for the most important architectural transformation since its inception: replacing the EVM with RISC-V. The reason is simple – in a future with zero-knowledge (ZK) at its core, EVM has become a performance bottleneck: currently zkEVM relies on interpreters, resulting in 50–800x performance slowdowns; Precompiled modules make the protocol complex and risky; The 256-bit stack design is extremely inefficient in generating proofs. RISC-V solution: minimalist design ( about 47 basic instruction ) + mature LLVM ecosystem ( support Rust, C++, Go and other language ); (90% of projects that have become the de facto zkVM standard adopt ); Has a formal SAIL specification ( achieves strict verification compared to the vague yellow book ) →; The hardware attestation path (ASICs/FPGAs) has been ( in testing )SP1, Nervos, Cartesi, and so on. The migration process is divided into three phases: replacing RISC-V as a precompiled module ( low-risk testing ); Dual-VM era: EVM and RISC-V coexist and are fully interoperable; Reimplement the EVM (Rosetta policy ) within RISC-V. Ecosystem impacts: Optimistic rollups ( such as Arbitrum and Optimism) need to rebuild fraud proof mechanisms; Zero-knowledge rollups ( such as Polygon, zkSync, Scroll) will gain huge advantages → cheaper, faster, and simpler; Developers can use language libraries such as Rust, Go, and Python directly at the L1 level; Users will enjoy proof of about 100x lower cost → about 10,000 TPS( to the Gigagas L1 ). Eventually, Ethereum will evolve from a "smart contract virtual machine" to a minimalist, verifiable trust layer of the network, with the ultimate goal of "ZK-Snark." Ethereum's crossroads Vitalik Buterin once said, "The end includes ...... Let everything be ZK-Snark." The endgame of zero-knowledge proofs (ZK) is inevitable, and its core thesis is simple: Ethereum is reinventing itself from scratch based on zero-knowledge proofs. This marks the technical end of the protocol – a refactoring of L1 to reach its final form, powered by a high-performance zkVM powered by a core development team ( such as Succinct). Ending with this vision, Ethereum is at the most important architectural transformation juncture since its inception. This discussion is no longer about gradual upgrades, but about a complete refactoring of its computing cores – replacing the Ethereum Virtual Machine (EVM). This initiative is the cornerstone of the broader "lean Ethereum" (Lean Ethereum) vision. The Streamline Ethereum (Lean Ethereum) vision aims to systematically simplify the entire protocol, splitting it into three core modules: streamlining consensus (Lean Consensus) and streamlining data (Lean Data) and streamlining execution (Lean Execution). Among the core issues of streamlined execution, the most critical point is: as the engine driving the smart contract revolution, has EVM become a major bottleneck in Ethereum's future development? As Justin Drake of the Ethereum Foundation puts it, Ethereum's long-term goal has always been to "Snark everything" (Snarkify everything), a powerful tool that enhances the layers of the protocol. For a long time, however, this goal was more of an "unattainable blueprint" because achieving it required real-time proof of the concept of (real-time proving). Now, as real-time proofs become a reality, the theoretical inefficiencies of EVM have become a practical problem to be solved. This article will provide an in-depth analysis of the technical and strategic arguments for migrating Ethereum L1 to RISC-V instruction set architecture (ISA). This move is expected to not only unlock unprecedented scalability, but also simplify the protocol structure and align Ethereum with the future of verifiable computing. What exactly has changed? Before exploring the "why", it is first necessary to clarify the "what" is changing. The EVM ( Ethereum Virtual Machine ) is the operating environment for Ethereum smart contracts, known as the "world computer" that processes transactions and updates the state of the blockchain. Over the years, its design has been revolutionary, laying the groundwork for the birth of decentralized finance (DeFi) and NFT ecosystems. However, this custom architecture from nearly a decade ago has now accumulated a lot of technical debt. In contrast, RISC-V is not a product, but an open standard—a free, general-purpose processor design "alphabet." As Jeremy Bruestle highlighted at the Ethproofs conference, its key principles make it an excellent choice for this role: Minimalism: RISC-V's underlying instruction set is extremely simple, consisting of only about 40 to 47 instructions. As Jeremy puts it, this makes it "almost perfect for the use case of the ultra-minimalist general-purpose machine we need." Modular design: More complex functions are added with optional extensions. This feature is critical because it allows the core to remain simple while expanding functionality as needed without imposing unnecessary complexity on the underlying protocol. Open ecosystem: RISC-V has a large and mature toolchain support, including the LLVM compiler, which enables developers to use mainstream programming languages such as Rust, C++, and Go. As Justin Drake notes, "The tools around the compiler are very rich, and the compiler is extremely difficult to build...... So having these compiler toolchains is extremely valuable." RISC-V enables Ethereum to inherit these off-the-shelf tools for free. Interpreter overhead issues The impetus for replacing EVMs is not a single flaw, but rather a confluence of fundamental limitations that cannot be ignored in the context of a future with zero-knowledge proofs at its core. These limitations include performance bottlenecks in zero-knowledge proof systems and the risks posed by the increasing complexity that accumulates within the protocol. Interpreter overhead issues The most pressing driver of this transformation is the inherent inefficiency of EVM in zero-knowledge proof systems. With Ethereum...