
Gavin Wood
1.3K posts

Gavin Wood
@gavofyork
Founded Polkadot, Kusama, Ethereum, Parity, Web3 Foundation. Building Polkadot. All things Web 3.0












New @VitalikButerin blog post on replacing the EVM with RISC-V in the long-term. I am a huge fan of this direction for Ethereum's execution layer. Today, RISC-V zkVMs like SP1 are the clear endgame solution for "ZK-ifying" Ethereum and quickly becoming the de-facto solution for ZK EVM. But as Vitalik cites in this post, our research at Succinct shows that the EVM is an extremely inefficient ISA for ZK proving. This proposal to migrate the execution layer directly to RISC-V can be viewed as migrating to "native" code for a ZK world as opposed to living in an interpreted world that imposes between 100-1000x overhead. By replacing EVM with RISC-V for the execution layer, we can up the gas limit on L1 by orders of magnitude (assuming optimization of other bottlenecks like a more zkVM friendly state merkleization format), while preserving verifiability. Link to full blog post below.



New @VitalikButerin blog post on replacing the EVM with RISC-V in the long-term. I am a huge fan of this direction for Ethereum's execution layer. Today, RISC-V zkVMs like SP1 are the clear endgame solution for "ZK-ifying" Ethereum and quickly becoming the de-facto solution for ZK EVM. But as Vitalik cites in this post, our research at Succinct shows that the EVM is an extremely inefficient ISA for ZK proving. This proposal to migrate the execution layer directly to RISC-V can be viewed as migrating to "native" code for a ZK world as opposed to living in an interpreted world that imposes between 100-1000x overhead. By replacing EVM with RISC-V for the execution layer, we can up the gas limit on L1 by orders of magnitude (assuming optimization of other bottlenecks like a more zkVM friendly state merkleization format), while preserving verifiability. Link to full blog post below.



Let’s be clear: criticism is welcome, defamation is not. There’s a clear difference between honest disagreement and spreading false narratives. We’ve never had an issue with people raising concerns — but we will speak up when there’s deliberate misinformation. We’re addressing facts and setting the record straight, while others are being emotional and trying to blur the focus. That’s not being dramatic. That’s setting boundaries.

I see a similar shift in this “industry”. Though there are some whose direction has remained largely unchanged for a decade. Lessons no doubt learnt along the way. But core principles unwavering: a desire not just to do great engineering for joyous products useful to many, but to achieve true decentralisation and build systems that stand the test of time.



