Optimistic and ZK Rollups Use Different Proofs to Settle Ethereum Batches

An Ethereum rollup’s proof system affects when direct withdrawals reach the main chain and how easily applications run on it. Optimistic and zero-knowledge (ZK) rollups use different methods to validate transaction batches.

Rollups execute transactions away from Ethereum and send transaction data back in batches. The data records activity for the network, while the rollup’s proof process determines how its proposed state is accepted.

An optimistic rollup posts a batch on the assumption that it follows the rules, then opens a defined period for challenges. During that window, a participant monitoring activity can contest an invalid state change with a fraud proof. If no challenge succeeds before the window closes, the state is accepted. This security model relies on at least one honest participant being able to watch the network and raise a valid challenge.

The challenge period affects direct withdrawals from a rollup to Ethereum. Arbitrum One sets its period at 6.4 days. That is a protocol challenge window, not necessarily the full time every user waits: total timing can vary with the rollup and the route used to withdraw.

A ZK rollup instead submits a validity proof with each batch to show transactions followed the rules. The rollup contract on Ethereum verifies the proof before accepting the updated state. That removes the need to wait out a fraud-challenge window, but a user’s withdrawal still depends on the rollup’s proof process and withdrawal system.

The name zero-knowledge does not mean activity is private. ZK rollups publish the data needed to reconstruct their state, in Ethereum formats such as calldata or blobs. Both rollup designs pay to publish data on Ethereum; ZK systems also generate and verify proofs, while optimistic systems use fraud proofs when a dispute occurs.

Software compatibility is another difference. Optimistic rollups have generally been more straightforward for developers building for the Ethereum Virtual Machine (EVM), which supports Ethereum applications. ZK systems have faced more difficulty proving general-purpose EVM computation, while zkEVM projects aim to support familiar contracts and developer tools.

The proof type therefore explains only part of a network’s user experience. Withdrawal routes, data costs and application compatibility also help determine how a specific rollup works.