Offchain Labs Critiques RISC-V Proposal for Ethereum
Offchain Labs, the core developer behind the Ethereum Layer 2 solution, Arbitrum, recently published a comprehensive technical piece that challenges Vitalik Buterin’s suggestion to shift Ethereum’s execution layer to RISC-V. They contend that WebAssembly (WASM) would provide greater long-term advantages.
In a post dated November 20th, four researchers from Offchain Labs expressed that while RISC-V might be proficient in generating zero-knowledge (ZK) proofs, it falls short as a means for delivering and storing smart contracts on Ethereum.
Back in April, Buterin proposed on Ethereum Magicians the idea of replacing the Ethereum virtual machine’s bytecode with an open-source RISC-V instruction set architecture (ISA). He claimed this shift could potentially cut on-chain ZK proof costs by up to 100 times.
“We appreciate these objectives, but we’re skeptical of Vitalik’s assumption that a single ISA can effectively cater to both ZK proofing and smart contract deployment,” the researchers—Mario Alvarez, Matteo Campanelli, Tsahi Zidenberg, and Daniel Lumi—stated.
Delivery and Proofing: Different Concerns
The crux of their argument revolves around differentiating between a “Delivery ISA” (dISA), which pertains to uploading contracts on-chain, and a “Proof ISA” (pISA) for ZK virtual machines. They assert that these do not need to be identical.
Offchain Labs has even created a prototype to illustrate their point. Arbitrum blocks include WASM-based Stylus smart contracts, which undergo ZK proofing by first compiling to RISC-V and then validating the execution of RISC-V.
“Using a RISC-V-based ZK-VM as a backend allows real-world blocks to be ZK-proofed on a blockchain utilizing WASM as dISA,” the post notes.
Questions About RISC-V’s Future
The research group raises concerns about whether RISC-V represents the final evolution of ZK-VMs, pointing out the fast-paced changes in the demonstration environment. For instance, the shift from 32-bit to 64-bit RISC-V demonstrates this unpredictability.
They warn that adopting RISC-V for Layer 1 could restrict Ethereum to a specific proof technology, especially as better alternatives become available. In contrast, WASM-based ZK-VMs, like Ligero’s Ligetron, have already shown advantages that hardware-specific ISAs might not be able to match.
Additionally, the researchers highlight that the costs for ZK proofs have sharply decreased to approximately $0.025 per Ethereum block and continue to decline. Thus, focusing solely on proof efficiency optimizations is becoming less critical. “Even if Layer 1 required several ZK certificates per block, that cost would pale in comparison to potential gas fees and miner extractable value (MEV) that builders could receive,” they noted.
WASM’s Advantages
The team underscores that WASM’s structured framework allows for easy code modifications and optimizations without disrupting existing contracts. While WASM operates effectively on standard hardware, most Ethereum nodes are not equipped with RISC-V CPUs, necessitating emulation.
WASM’s validation features ensure type safety, mitigating potential vulnerabilities, and it has a well-established tool ecosystem that has undergone rigorous testing across billions of execution scenarios.
“We believe WASM could evolve into a type of Internet protocol for smart contracts, bridging the gap between various source languages used for writing smart contracts and the different backends employed for executing and proving them,” the researchers concluded.
