Vitalik advocates completing self-contained heavy cryptographic computations on clients, with the chain verifying only proofs and results.
Article
Vitalik: Ethereum should move computationally intensive cryptographic computations to clients, with the chain only verifying proofs and results
PANews October 19 reported that Ethereum co-founder Vitalik Buterin further elaborated on the EIP-8288 roadmap, arguing that precompiled contracts and EVM are not ideal places to host most of Ethereum's cryptographic support, because both depend on data that must be included on-chain, typically consuming calldata space in blocks. He stated that even with PeerDAS and blocks-in-blobs, data throughput will begin to face practical pressure at around 8MB/s and above. Vitalik argued that Ethereum should be viewed as the “computer of the cryptographic world,” and content that need not be placed in the data and execution “main thread” should not occupy it. For self-contained computationally intensive tasks, computation can be completed on clients and generate STARK proofs, then be submitted through EIP-8288's recursive aggregation mechanism; the chain need only verify the computation...
PANews said that Ethereum co-founder Vitalik Buterin further elaborated on the EIP-8288 roadmap on X.
02
Vitalik believes that precompiled contracts and EVM are not ideal places to host most of Ethereum's cryptographic support.
03
The original text stated that precompiled contracts and EVM depend on data that must be included on-chain, typically consuming calldata space in blocks.
04
Vitalik stated that even with PeerDAS and blocks-in-blobs, data throughput will begin to face practical pressure at around 8MB/s and above.
05
Vitalik argued that self-contained computationally intensive tasks can be completed on clients and generate STARK proofs, then be submitted through EIP-8288's recursive aggregation mechanism.
06
The headline and summary state that the chain need only verify proofs and results.
AI-assisted interpretation
The following is analysis, separate from reported facts. Verify important claims independently.
This approach moves part of the computational workload from the chain to user-side clients: clients complete intensive computations and generate verifiable STARK proofs; Ethereum does not re-execute all computations on-chain, but instead verifies the submitted proofs and results.
Why it matters to readers
This proposal focuses on the use of on-chain data and execution resources, seeking to have the main thread process only content that must be placed on-chain.
Beginners can understand this as “solve problems off-chain; verify answers and proofs on-chain.” The material currently describes Vitalik's proposed technical approach and does not mean that the solution has already been deployed.
Risks and unknowns
The material is PANews' paraphrase of Vitalik's remarks and does not provide a link to the original X post.
The body text is truncated with an ellipsis, so the complete technical explanation and qualifications are unavailable.
The material does not state EIP-8288 whether it has been adopted, when it will be implemented, or whether client implementations already exist.
The material does not provide STARK complete data on generation costs, verification costs, security, or the feasibility of recursive aggregation.
The testing conditions and calculation basis for the approximately 8MB/s pressure threshold are not stated in the material.
Related Developments
Loading event timeline…
Related concepts
Precompiled contracts
This term is not in the glossary yet. Browse related concepts in the glossary.
