0x77a6632677a6…77a66323
Buterin Defends EIP-8037 as Ethereum Weighs Steeper State Costs
EIP-8037 would raise cold storage slot costs from 22,100 to 110,020 gas and contract deployment fees eightfold, as Buterin argues repricing should be routine, not crisis.

Outputs
EIP-8037 sets a cost-per-state-byte of approximately 1,530 gas, drafted October 2025 by Silva, Dietrichs and Brouwer.
A 24 KiB contract deployment would rise from about 4.95 million gas to 37.8 million gas.
Cold storage slot creation would climb from 22,100 gas to 110,020 gas.
Ethereum state grows about 110 GiB per year; EIP-8037 targets 60-120 GiB per year even at higher gas limits.
Glamsterdam testnet work has been delayed after at least 13 contracts were found with outdated gas assumptions.
Ethereum co-founder Vitalik Buterin has publicly defended EIP-8037, a proposal that would raise the cost of permanent state storage on Ethereum by as much as a factor of eight for certain operations, arguing that contracts treating gas constants as permanent guarantees are the real source of risk.
The proposal, drafted in October 2025 by Maria Silva, Ansgar Dietrichs and Jochem Brouwer, is slated for the Glamsterdam upgrade. It would introduce a fixed cost-per-state-byte, or CPSB, of approximately 1,530 gas per byte and separate state gas from ordinary execution gas, ending the practice of billing computation and permanent storage on the same meter.
The price changes are steep rather than marginal. Deploying a 24 KiB contract would rise from roughly 4.95 million gas to 37.8 million gas under the new schedule. Creating a new cold storage slot would climb from 22,100 gas to 110,020 gas — a fivefold increase on a single write. New account creation and contract code deployment would fall under the same framework.
Why does Ethereum want this now?
The arithmetic explains the urgency. Ethereum's state — the live record of account balances, contract code and contract storage that every node must hold — currently grows by about 110 GiB per year. That figure alone would fill a typical consumer laptop's storage in well under a decade.
Under a hypothetical tenfold increase in the gas limit and without EIP-8037, annual state growth could reach approximately 1,100 GiB, a pace no ordinary operator could sustain. The proposal is designed to let throughput rise while keeping state growth stabilized at roughly 60-120 GiB per year.
EIP-8037 sits alongside a companion proposal, EIP-8038, in the Glamsterdam roadmap. How the two interact remains unresolved, and the drafters have not yet confirmed the CPSB figure will survive further review. The promise of stabilized state comes with a caveat: none of these numbers is final until the upgrade's specification freezes.
Where has the friction already appeared?
Work on the Glamsterdam testnet has already run into delays. At least 13 contracts have been identified with outdated gas assumptions. Any contract that forwards a fixed amount of gas for a storage write, or that caps gas on a call, could fail outright once state operations become more expensive.
Buterin's defense lands precisely on that point. If teams treat gas costs as permanent constants, every future repricing becomes a crisis. If they design with flexibility in mind, repricing becomes routine maintenance. That is a design philosophy, not a guarantee — the thirteen affected contracts show how far the ecosystem currently sits from it.
Notably absent from the discussion so far is any compensation mechanism or migration window for affected deployers. The proposal prices the burden; it does not soften the transition.
What does this mean for developers and users?
For developers, the immediate task is an audit. Any application that deploys contracts, creates accounts, or writes new storage slots should check its assumptions against the proposed figures before Glamsterdam ships.
The steepest adjustment falls on specific patterns rather than the ecosystem at large:
- Factory contracts that spawn many new accounts
- Wallet deployment flows
- Applications that write large volumes of fresh storage
- Contracts that hard-code gas forwarding for storage writes
Projects deploying large contracts face the harshest arithmetic, since code deployment costs scale directly with the byte count now priced at roughly 1,530 gas each.
For users, the consequence is indirect but concrete: applications whose operating costs rise may pass those costs on, particularly account-abstraction wallet flows that depend on cheap account creation. The proposal itself does not change gas for reads or existing storage, only new state creation.
The open questions are specific and testable. Watch whether the Glamsterdam testnet clears its contract compatibility issues, whether the approximately 1,530-gas-per-byte CPSB figure holds through further review, and how EIP-8038 reshapes the final design. Until Glamsterdam's schedule firms up, developers have a closing but finite window to audit before the new cost schedule becomes consensus.
via cryptonews.net (Source)
More from Elena Vasquez
Show full bio
Staff writer covering regulation, policy and their impact on crypto and web3 industry at Mempool Brief.
1,525 articles