The Ethereum Foundation launched the Platåberget public testnet on August 17, 2026, marking the first public testing phase for the Glamsterdam hard fork — described by core developers as the protocol's most significant overhaul since The Merge in September 2022. The testnet transitioned to Glamst...
"We don't friggin need more copypasta EVM chains." — Vitalik Buterin, Ethereum Co-Founder, February 2026
The Ethereum Foundation launched the Platåberget public testnet on August 17, 2026, marking the first public testing phase for the Glamsterdam hard fork — described by core developers as the protocol's most significant overhaul since The Merge in September 2022. The testnet transitioned to Glamsterdam rules on August 20. Mainnet activation, originally targeted for H1 2026, has slipped to Q4 2026 after the upgrade's scope expanded beyond initial estimates.
Glamsterdam's two headline changes — Enshrined Proposer-Builder Separation (ePBS, EIP-7732) and Block-Level Access Lists (BALs, EIP-7928) — restructure how Ethereum blocks are built and executed. The upgrade targets a gas limit increase from the current ~60 million to 200 million, enabling a theoretical throughput of 10,000 transactions per second. It is also projected to reduce L1 gas costs by approximately 78% for simple transfers and cut MEV extraction by up to 70%, according to protocol specifications. These changes carry direct implications for the $550 million-plus in annual MEV revenue currently extracted from Ethereum and the ~$200–500 million annual RPC and infrastructure services market.
The economic consequences are substantial. Two block builders — Titan and Quasar — currently construct approximately 73% of all Ethereum blocks, with the top four builders exceeding 95% market share. Titan alone has accumulated $112.6 million in block-building revenue through July 2026. Enshrining PBS into the protocol would eliminate the need for trusted relay infrastructure that currently intermediates 88% of block production, fundamentally reshaping the MEV supply chain.
On August 17, the Ethereum Foundation announced Platåberget, a purpose-built public testnet for the Glamsterdam fork. The testnet activated Glamsterdam rules on August 20, 2026. Unlike short-lived devnets used during earlier development phases, Platåberget is designed to run for several months, providing a stable environment for community testing before the upgrade progresses to Ethereum's longer-lived testnets, Sepolia and Hoodi, and eventually mainnet.
The testnet supports all major consensus layer clients — Lighthouse, Lodestar, Nimbus, Prysm, Teku, and Grandine — and all major execution layer clients — Besu, Geth, Erigon, Nethermind, Reth, Nimbus-EL, and Ethrex. Public validator deposits are enabled via the Dora explorer, and a faucet provides testnet ETH for experimentation.
The Ethereum Foundation noted in its announcement that "any tool that relies on a hardcapped maximum gas limit — think wallets, indexers and gas estimators — will break and needs to be updated." This warning signals a significant coordination effort across the Ethereum tooling ecosystem ahead of mainnet deployment.
Glamsterdam packages 10 EIPs across the consensus and execution layers. Two constitute structural changes to Ethereum's block production and execution architecture.
Currently, 80–90% of Ethereum block production depends on MEV-Boost, an off-chain relay system operated primarily by Flashbots. MEV-Boost separates block proposers (validators) from block builders (specialized firms that order transactions for profit), but it does so through trusted, off-chain intermediaries.
EIP-7732 moves this separation on-chain. Under ePBS, proposers commit to a block header and a separate builder constructs the execution payload, with the handoff enforced at the protocol level. A new Payload Timeliness Committee (PTC) replaces relay-based attestation. The data propagation window extends from 2 seconds to approximately 9 seconds, accommodating the two-phase block construction process.
The practical effect: trusted relays become optional. The protocol itself guarantees the integrity of the proposer-builder handoff, eliminating a single-point-of-failure dependency that currently intermediates approximately 88% of Ethereum blocks.
BALs introduce a mandatory record in every block that maps every account and storage slot the block touches, along with post-execution state values. This enables three capabilities:
| EIP | Change | |-----|--------| | EIP-8037 | Introduces a separate "state gas dimension" for state-creating operations, targeting average state growth of 120 GiB/year at 150M gas limit | | EIP-2780 | Reduces intrinsic transaction gas, making simple ETH transfers up to 71% cheaper | | EIP-8038 | Decomposes state-access costs alongside EIP-2780 | | EIP-7954 | Raises maximum deployed contract size from 24 KiB to 64 KiB; initcode cap from 48 KiB to 128 KiB | | EIP-7976 | Increases calldata floor cost | | EIP-7981 | Raises EIP-2930 access list cost | | EIP-7778 | Removes gas refunds from block accounting | | EIP-8024 | Introduces new stack-manipulation opcodes | | EIP-7843 | Exposes slot number via SLOTNUM opcode | | EIP-7708 | Adds ETH transfer event logging |
Ethereum's current MEV economy is concentrated among a small number of entities. According to MEV Watch dashboard data from July 2026:
| Builder | Block Share | |---------|------------| | Titan Builder | 54.7% (3,523 of 6,442 blocks in 24-hour sample) | | Quasar | 18.6% | | BuilderNet | 12.9% | | Eureka | 10.5% | | All others | ~3% |
Titan Builder alone has earned $112.6 million in block-building revenue through July 2026. In a single incident on July 8, Titan extracted $1.8 million (1,018 ETH) from one swap, where a trader swapping 1,126 ETH (~$2.01 million) on a low-liquidity pool received only ~$14,200 in return — a 99.3% loss.
Annual MEV extraction on Ethereum runs at approximately $550 million under normal market conditions, with daily extraction of $10–20 million on typical days and spikes to $40–50 million during volatility events. A further $1–2 billion in MEV infrastructure revenue is generated across relay operators, searcher firms, and builder networks.
ePBS is projected to reduce MEV extraction by up to 70%. If realized, this would cut approximately $385 million from annual MEV revenue. The practical consequences:
Glamsterdam's gas repricing package restructures costs through two mechanisms operating in tension:
Cost reductions: EIP-2780 decomposes intrinsic transaction gas, making simple ETH transfers up to 71% cheaper. The combined effect of all repricings is projected to cut average L1 fees by approximately 78.6%.
Cost increases: EIP-8037 introduces a separate state gas dimension that raises costs for operations creating new state — new accounts, storage slots, or contract deployments. This is designed to constrain state growth to approximately 120 GiB annually at the reference gas limit.
The net effect depends on transaction type:
A critical caveat applies to overall fee levels. A threefold jump in block capacity (60M to 200M gas) without a corresponding increase in demand would keep base fees low and throttle the EIP-1559 burn. In the near term, this could increase ETH's net issuance rate — currently running at approximately 0.8% annual inflation — by reducing the burn mechanism's effectiveness. Whether higher capacity attracts sufficient new demand to offset this remains an open question.
The Ethereum Foundation's explicit warning about breaking changes in gas limit handling signals meaningful costs for ecosystem participants:
Node operators face hardware considerations. Parallel execution under BALs requires concurrent disk reads, making storage IOPS a more critical bottleneck. The Ethereum Foundation recommends NVMe SSD audits for validator nodes. Lower network latency also becomes more critical under ePBS's extended propagation window.
Smart contracts themselves remain backward-compatible — no redeployment is required. However, contracts that depend on specific gas cost assumptions may behave differently after repricing.
The Glamsterdam upgrade has experienced multiple schedule adjustments:
Core developers have stated that "getting ePBS right outranks hitting any fixed date," leaving open the possibility of further delays.
The delay compresses the timeline for Hegotá, the follow-up upgrade. Developers are currently reviewing 66 EIPs for Hegotá, with a proposed-for-inclusion list expected by August 27, 2026, and client team preferences due by September 10. Developer Toni Wahrstätter noted that "only a fraction are likely to receive the implementation work, devnets, and testnets needed for a realistic 2027 release."
Hegotá's confirmed headliner is FOCIL (EIP-7805), a fork-choice inclusion list mechanism for censorship resistance that was originally proposed for Glamsterdam but deferred to reduce testnet complexity. Other items under consideration include shorter slot times (EIP-8198), frame transactions (EIP-8141), and privacy improvements.
Glamsterdam represents Ethereum's most consequential protocol change since The Merge. Where The Merge replaced proof-of-work with proof-of-stake — a consensus mechanism swap — Glamsterdam restructures how blocks are built, executed, and verified. It addresses two of Ethereum's most documented structural problems: the concentration of block production among a handful of off-chain builders, and the sequential execution bottleneck that limits L1 throughput.
The economic implications are direct. An estimated $550 million in annual MEV extraction and hundreds of millions in relay infrastructure revenue face restructuring. The 95%+ builder concentration that currently characterizes Ethereum's block market confronts a protocol-level intervention. Whether this redistributes value more broadly or simply shifts concentration to different chokepoints remains to be determined by the market's adaptation.
The capacity expansion carries its own tension. A 3.3x increase in the gas limit without commensurate demand growth would suppress the EIP-1559 burn, potentially increasing ETH's inflation rate in the near term — even as it scales the network for higher throughput. This dynamic mirrors the Dencun upgrade's effect on L1 fee revenue: technically successful, economically ambiguous for ETH holders.
The upgrade is not yet deployed. Months of testnet validation on Sepolia and Hoodi must follow the Platåberget phase. Breaking changes across the tooling ecosystem require broad coordination. The historical pattern of Ethereum hard forks suggests the actual mainnet date will be determined by client readiness and bug discovery, not calendar targets.