Ethereum node operators now have access to an active public network called Platåberget to rehearse the coordinated execution- and consensus-client upgrade known as Glamsterdam. Maintained by ethPandaOps and launched on Aug. 13, the network features an open, permissionless validator set designed to test client pairs, validators, builders, and supporting infrastructure until the upgrade reaches mainnet.
While decentralized application, smart contract, and EVM testing is directed to Sepolia, Platåberget focuses specifically on client software, staking setups, builders, and infrastructure. Node operators must utilize compatible releases for both the execution client, which processes transactions, and the consensus client, which tracks validators and agrees on the chain.
The largest architectural change tested on the network is EIP-7732, or enshrined proposer-builder separation (ePBS). This Review-stage proposal separates consensus-block validation from execution-payload validation, introduces staked builders, and creates a Payload Timeliness Committee to report whether a builder revealed its payload on time. This design expands the roughly two-second critical path by giving the next proposer six seconds to validate a payload and other validators nine seconds.
Additionally, Review-stage EIP-7928 introduces block-level access lists that record the accounts and storage locations touched by a block. This map allows clients to parallelize disk reads, transaction validation, state-root calculation, and state reconstruction, though realized throughput will still depend on individual client implementation.
Other proposals scheduled and being tested include EIPs 2780, 8037, and 8038, which remain in Review. Platåberget's public page references gas adjustments, such as an 18,000-gas example for a plain ETH transfer, while current EIP-2780 reference cases total 21,000 execution gas for a transfer to an existing externally owned account. The state-creation proposal introduces a charge of 183,600 state gas for creating a new account, and the state-access proposal raises storage writes from 2,800 to 10,000 gas. These adjustments require development teams to recheck hardcoded subcall gas, cached estimates, and workflows involving account or storage creation.
