What is RISC-V? If you’d put that question to the cryptosphere a week ago, 90% of them would have responded with a shrug. Ask the same question today and there’s a good chance you’ll get some semblance of an informed response. Perhaps something to do with it being a superior VM whose open-source design and scalability makes it the solution web3 needs.
Credit for the RISC-V’s sudden moonwalk into the spotlight goes to Vitalik Buterin, whose blog post on April 20 proposing the tech replace Ethereum’s EVM has sparked intense debate. It’s also reignited interest in the projects that have been using RISC-V for years, notably Cartesi or that are on the verge of adopting it; Polkadot. Will Ethereum be added to the shortlist of RISC-V adopters or will its community give the upgrade the thumbs down?
Risking It All for Ethereum
It’s no secret that Ethereum is struggling. Not so much on the market – though price follows network performance (or is it the other way round?) – but from a developer perspective. Put simply, Ethereum is slower, more complex, and more expensive than most of the Layer 1s that have sprung up since it – the best of which look like sportscars to Ethereum’s Ford Model T: iconic at the time but now badly outdated.
Now, Ethereum’s most famous founder has concluded that the key to scaling Ethereum is to do away with the EVM altogether and replace it with something distinctly un-etherean: RISC-V. The instruction set architecture is open source and free for anyone to modify – and many have, from Polkadot, who are using it in their PolkaVM, to Cartesi, which has used it to power its modular dapp blockchain.
As for why Vitalik – known for his slow and steady approach rather than constantly chasing better tech – has planted his flag in RISC-V’s camp, it comes down to benchmarks. He estimates that RISC-V could improve execution efficiency on Ethereum by up to 100x, particularly for ZK proofs. Proof times could drop from five minutes to three seconds, enabling faster rollups and cheaper transactions. And those are the sort of efficiency gains that can’t be sniffed at.
Particularly at a point when the Ethereum community has begun to despair of ever regaining ground on the Suis and Solanas of the smart contract world.
What RISC-V Brings to the Party
The best indication of what RISC-V might actually do in a blockchain context can be found by scrutinizing Cartesi, which has been using the tech for seven years now. Cartesi’s rationale for doing so goes something like this: its goal is to bridge web3 with mainstream software development by avoiding the “reinventing the wheel” problem.
The EVM forces developers to rebuild basic operations due to its domain-specific design, limiting expressive power. RISC-V, in comparison, as an open standard, supports Linux and mature tooling. As a result, devs can use familiar languages such as C++ or Rust to build dapps – which makes it much easier to onboard regular tech teams and enterprises, which is how web3 mass adoption is won.
According to lead Cartesi dev Felipe Argento, “RISC-V is a free, open standard with a rich ecosystem supported by a global community and major corporations. RISC-V stands in a sweet spot among ISAs. It’s an open standard, supports Linux, and isn’t burdened by the complexity or obsolescence issues plaguing other ISAs. This makes RISC-V an ideal choice.” He’s also acknowledged Vitalik’s recent post on the topic, and stated that it would be a “pleasure” to contribute to its Ethereum integration.
Should Ethereum’s community give RISC-V the green light, Cartesi’s implementation will surely be studied as an indicator of what they can expect. But replacing the established EVM with a different VM on a live blockchain running tens of thousands of critical smart contracts is a huge undertaking – as Vitalik has been swift to admit.
Is the EVM’s Days Numbered?
For the last decade, smart contract blockchain have largely relied on bespoke VMs such as the EVM. While this enables architects to design their network precisely to their liking, this rigid approach mandates every prospective network developer devoting countless hours to mastering it before they can become proficient at dapp creation. Moreover, chain-specific VMs suffer from limited interoperability – not to mention limited tooling. All of which imposes a lot of friction on developers and limits cross-chain composability.
Adopting RISC-V as a shared, low-level execution fixes this. But implementing it while deprecating the EVM – or forcing the two technologies to co-exist – won’t be easy. The upside, however, arguably makes the torturous upgrade path worthwhile.
