A team of five Ethereum consensus researchers, including Francesco D’Amato and Ben Edgington, set up a Core EIP project Wednesday (October 21) called Decoupled consensuswhich would remove Gasper, the current consensus design of the beacon chain, and split validator voting into two separate jobs.
The result, if delivered, would be faster finality, stronger reorganization defenses, and a cleaner on-ramp to post-quantum cryptography.
-
Two jobs, not one: Today, each validator certificate serves a dual purpose: to vote at the head of the chain and to count toward finality. Under the drafta committee of approximately 512 validators sampled from each slot would manage principal votes, while each active validator votes on finality in separate rounds that no longer wait for epoch boundaries.
Do you like this article?
Subscribe to Bankless or log in
-
The final target: The initial objective of the EIP is to divide the time to finality at least by 4. Blocks currently need two epochs to complete (~13 minutes at best), so for today’s numbers that’s around three minutes. Longer term, Edgington wrote on the FE consensus blog in July, this goal “in a few dozen seconds” appears to be a useful objective.
-
Defense of reorganization: LMD-GHOST, Gasper’s choice half, never expires old votes, so attackers can sit on votes and strategically drop them to influence the channel. The new design only counts recent votes, adds expiring locks to protect confirmed blocks, and replaces nominator boosting with a vote-sharing mechanism called view-merge.
-
The quantum angle: The project does not choose a post-quantum signature scheme. Instead, it separates the vote paths so that each can change how votes are signed and grouped later. This is important because PQ signatures are much larger than current BLS signatures, and limiting votes per head of list to approximately 512 signers, while giving final votes more time to propagate and aggregate, leaves room for this additional weight.
-
The first days: Many elements are still to be determined, including committee selection, awards, and full security analysis, and the authors say they are primarily looking for an EIP number at the moment (one reviewer suggested putting it under the name EIP-8453). The draft also indicates that a small group is an easier target in the event of a denial of service. Its expedition requires a consensus layer hard fork, and Ethlabs researcher Barnabé Monnot has I fixed it for myself*the upgrade after Hegota, early 2028.
About the Author
