Highlights from the All Core Developers Consensus (ACDC) Call #174

Ethereum core developers used ACDC #174 to move several high-impact protocol threads forward, reinforcing a broader shift toward incremental scalability, disciplined fork scoping, & stronger censorship resistance guarantees. The call focused on three deeply connected tracks: interfork cell-level deltas, Glamsterdam ePBS Devnet-0, & the final phase of Hegotá headliner selection.

• Interfork Updates
• Glamsterdam Updates
• Hegota Updates

Interfork Updates

The first major update centred on cell-level deltas, an optimization designed to reduce unnecessary bandwidth usage during blob data propagation. Unlike previous scaling improvements that required coordinated fork upgrades, this change is explicitly interfork, meaning it can be deployed without a hard fork and progressively enabled across the network.

Today, consensus clients gossip full data columns during the data availability distribution phase, even when peers already possess partial data. Cell-level deltas replace this blunt approach with a more surgical mechanism, i.e., peers exchange bitmaps indicating which cells they are missing, and only those missing cells are transmitted.

This approach dramatically reduces redundant data transfer in realistic network conditions, where validators often already have partial blob data in their mempools. Crucially, the optimization is fully backward compatible.

Nodes negotiate support transparently via GossipSub capabilities, and a feature flag allows clients to enable or disable the behaviour without affecting peers that have not yet adopted it. Multiple consensus clients reported near-complete implementations.

1. Lighthouse is largely finished, with only minor spec updates & observability polish remaining.
2. Prysm is finalising review of partial column handling & integrating recent GossipSub changes related to masked requests.
3. Lodestar is in late-stage debugging & cleanup.

Some complexity remains for clients whose peer-to-peer stacks are maintained independently, particularly around end-to-end testing of partial messaging support. However, no fundamental protocol blockers were raised.

Rather than measuring success purely in bandwidth reduction, developers emphasised network topology safety as a primary concern. One key risk is unintentional peer clustering, where nodes supporting partial messaging might preferentially connect with each other, degrading overall network resilience.

Planned devnet testing includes mixed-support scenarios, private blob cases, empty mempools, & adversarial conditions where only a subset of peers adopt the optimization. Metrics dashboards are being expanded to track propagation patterns, peer scoring behaviour, & consistency across clients.

Glamsterdam Updates

The second major thread focused on Glamsterdam, specifically readiness for ePBS Devnet-0, the first live environment designed to test proposer-builder separation at the protocol level. Devnet-0 is intentionally minimal.

There are no external builders, no complex relay markets, & no MEV competition dynamics. Instead, validators self-build execution payloads but broadcast them separately from consensus blocks, validating the core architectural separation without introducing unnecessary risk.

Recent spec updates simplify this separation further, including structural changes to how commitments are embedded & how data is handled on the wire. Client teams were encouraged to target the latest alpha specs, with a shared expectation that Devnet-0 should be live within weeks rather than months.

This restraint is deliberate. Rather than rushing new EIPs into Glamsterdam, core developers appear focused on ensuring that the foundational mechanics of ePBS work reliably before layering additional complexity.

Several edge cases were highlighted during the call, including a deposit-related bug uncovered during review & the non-trivial complexity of fork-choice recovery in the presence of optimistic execution. Developers stressed that much of the real difficulty does not lie in block separation itself, but in handling error recovery, branch invalidation, & state rollback under failure conditions.

Testing infrastructure for these scenarios is being expanded now, rather than deferred until later devnets. Importantly, the Devnet-0 scope avoids external builders precisely to keep these variables manageable. Once the separation primitives are proven stable, subsequent devnets can safely introduce builder competition & market dynamics.

New tooling presented during the call allows real Ethereum blocks to be transformed into their post-ePBS representations, measuring wire size, compression behaviour, & overhead. Early results show that separating blocks into consensus & execution components introduces only negligible size variation, largely attributable to compression table differences rather than structural inefficiencies.

Hegota Updates

The final & most strategically significant discussion concerned Hegotá, the fork following Glamsterdam, & the selection of its consensus-layer headliner. With the proposal submission deadline closed, attention has shifted to evaluation & finalisation. While multiple ideas are on the table, one stands out clearly, i.e., FOCIL.

FOCIL enables multiple validators to participate directly in inclusion list construction, reducing reliance on any single proposer or builder while preserving MEV efficiency. In practical terms, it allows Ethereum to improve censorship resistance without forcing validators to sacrifice rewards or operate complex local infrastructure.

FOCIL is not a new idea. It has been iterated on for over two years, reviewed across clients, & prototyped to varying degrees by most major consensus implementations.

More importantly, it aligns cleanly with Ethereum’s longer-term roadmap around scaling, rollups, & zero-knowledge systems. Developers emphasised that transaction inclusion guarantees benefit everyone, i.e., end users, L2s, institutional participants, & even validators themselves.

Unlike purely performance-driven upgrades, FOCIL strengthens Ethereum’s core neutrality properties. From a governance standpoint, FOCIL is also unusually mature.

A proposal exploring partial reconstruction & incremental two-dimensional data availability was presented as a potential scalability-focused alternative. While the room leaned strongly toward FOCIL, several participants explicitly called for a final confidence check, emphasising that headliner decisions are deeply committing & difficult to reverse.

Perhaps the most important signal from this call was not any single technical proposal, but the process itself. Ethereum core developers are increasingly willing to slow down, defer decisions by a single call, & demand high confidence rather than momentum.

Cell-level deltas will likely ship before the next fork. Glamsterdam is being deliberately constrained.

Hegotá’s headliner will be finalised only after additional review, despite broad support already being evident.

If you find any issues in this blog or notice any missing information, please feel free to reach out at yash@etherworld.co for clarifications or updates.

To promote your Web3 articles, events, and projects, you may reach out anytime via EtherWorld PR for submissions and collaboration.

Related Articles

1. Highlights of Ethereum's All Core Devs Meeting (ACDC) #153
2. Highlights of Ethereum's All Core Devs Meeting (ACDE) #207
3. Highlights of Ethereum's All Core Devs Meeting (ACDC) #152
4. Highlights of Ethereum's All Core Devs Meeting (ACDE) #206
5. Highlights of Ethereum's All Core Devs Meeting (ACDE) #205

To follow blockchain news, track Ethereum protocol progress, and read our latest stories, subscribe to our weekly today.