Ethereum’s Glamsterdam upgrade is approaching mainnet, and it could be the most significant change to the network since the transition to proof-of-stake. With proposer-builder separation moving into the core protocol and a massive gas limit increase on the table, the upgrade promises cheaper transactions and a fundamentally different way blocks are constructed.
By Amir Hassan | August 13, 2026
The Hook: Two Big Changes in One Upgrade
Glamsterdam cleared its final devnet stage in June 2026, and Ethereum core developers are now preparing for the transition to public testnets like Holesky and Hoodi. While no locked mainnet date has been announced yet, trackers point to a window ranging from late August through December 2026, consistent with the two-to-four-month testnet seasoning that past Ethereum forks have required.
The upgrade carries two headline changes that matter even if you are not a developer. First, proposer-builder separation — often abbreviated as PBS — would move the rules for block construction directly into Ethereum’s core protocol. Second, block-level access lists would let validators process unrelated transactions in parallel, potentially speeding up confirmation times and reducing fees for certain transaction types.
On-Chain Evidence: What the Technical Changes Actually Do
To understand why PBS matters, think of Ethereum block production like a factory assembly line. Right now, the same entity both proposes a block (the proposer) and builds its contents (the builder). This creates inefficiencies and potential censorship risks, because whoever builds the block decides which transactions get included and in what order.
Proposer-builder separation splits this into two roles. Specialized builders compete to create the most valuable block, and the proposer simply selects the best one. Think of it like opening more checkout lanes at a store — instead of one cashier processing every transaction, multiple cashiers work in parallel and the fastest one wins your business.
Block-level access lists work similarly. They let validators identify which parts of the network state a transaction will touch, allowing unrelated transactions to be processed simultaneously. If transaction A touches your stablecoin balance and transaction B touches a completely different smart contract, they no longer have to wait in line for each other.
The gas limit floor target of 200 million is another significant change. For context, raising the gas limit means each block can hold more transactions, which could ease congestion during high-activity periods. However, as analysts have noted, actual fees still depend on network demand rather than capacity alone — a bigger highway does not eliminate traffic jams during rush hour.
The Core Conflict: Speed vs. Stability
Ethereum has historically taken a deliberately conservative approach to upgrades, and for good reason. The network secures hundreds of billions of dollars in value — rushing a change could have catastrophic consequences. Glamsterdam’s PBS implementation has been tested on devnets, but moving to public testnets introduces variables that simulated environments cannot fully capture.
Meanwhile, Solana is taking the opposite approach. Its Agave v4.2 client update, scheduled for mainnet activation on August 17, halves slot times to 200 milliseconds and implements the Alpenglow consensus rewrite. Solana’s aggressive timeline contrasts sharply with Ethereum’s cautious roadmap, creating a competitive dynamic where speed of execution meets robustness of process.
For Layer-2 networks, which now handle an estimated 95 percent or more of all Ethereum transaction volume, Glamsterdam could be transformative. The upgrade addresses bottlenecks at the base layer that directly affect how L2 solutions settle transactions. Lower base-layer fees and faster confirmation could cascade through the entire ecosystem of rollups, sidechains, and scaling solutions.
Market Implications: What This Means for ETH Holders
If you hold Ethereum (ETH), currently trading around 1,886 dollars, the Glamsterdam upgrade is fundamentally bullish for the network’s long-term value proposition. Cheaper and faster transactions attract more users and more applications, which drives more demand for ETH as the gas that powers the network.
Here is what to watch:
- Testnet performance — Watch for reports from Holesky and Hoodi testnets for any issues with PBS or access lists
- Mainnet date announcement — Ethereum core developers typically lock dates only after testnets run cleanly for several weeks
- Layer-2 fee changes — Even before Glamsterdam launches, anticipation could affect L2 token prices as markets price in future improvements
- Validator readiness — Validator operators need to upgrade their clients before the fork, and adoption rates signal network health
The Verdict: Evolution, Not Revolution
Glamsterdam may not generate the headlines of a halving or an ETF approval, but its impact on Ethereum’s daily operation could be profound. By moving block construction into the protocol and enabling parallel transaction processing, Ethereum is addressing two of its most persistent criticisms: high fees and slow confirmation times.
The upgrade also reinforces Ethereum’s philosophy: evolve carefully, test thoroughly, and ship only when the network is ready. In a crypto industry often driven by hype cycles and rushing to market, that patience is itself a competitive advantage — one that could pay dividends for ETH holders over the long term.
The cryptocurrency market remains highly volatile. This article is for informational purposes only and does not constitute financial advice.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. Cryptocurrency investments carry risk; always do your own research.
PBS finally moving to core protocol is massive. no more relying on MEV-Boost middleware that breaks every other upgrade. about time devs handled this at consensus level
The MEV extraction impact is the real story here. If PBS works as designed, validators wont need to trust relay operators anymore. That alone fixes one of Ethereums biggest trust assumptions.
Yuki Tanaka PBS eliminating relay trust is huge but the implementation details matter. if builders can still censor txs through block construction the problem just moves one layer down
PBS finally moving to core protocol. about time, the MEV mafia has been eating retail alive for years
A 200 million gas limit sounds great until you realize state bloat will make running a full node even harder for hobbyists. There is a reason L1s with huge block limits end up centralized.
stefan acting like gas fees being cheap is a bad thing lmao. some of us actually use the chain instead of just running nodes
cheap gas is lovely until you ask who pays for it. hobbyists running the infra so swaps cost cents, someone always eats the bill
same argument got made when the limit went to 36M and hobby nodes survived. but state expiry better ship with 200M or the raspberry pi gang is done for
hobby nodes survived 36M because ssd prices cooperated. 200M without state expiry and full nodes become a datacenter hobby real fast
ssd prices wont save you this time. state growth is quadratic and ram requirements quietly kill more hobby nodes than disk
everyone argues about ssds while ram quietly kills more hobby nodes. you can nvme your way out of disk, you cant ram your way out of memory
200 million gas limit is insane. current limit is around 36 million right? thats almost 6x. state bloat gonna be wild
@beata parallel processing via access lists is the key piece. without that the gas bump would nuke nodes
200M gas limit with access list parallelization is ambitious but hobby node operators are gonna get crushed. history rhymes
200M gas limit with access list parallelization is the only way this works without nuking node hardware requirements. Stefan is right to worry about hobbyists getting priced out
PBS moving to consensus layer is the most underrated part of Glamsterdam. MEV relays have been a trust bottleneck since Merge
enshrined PBS also breaks the relay cartel. ofac-compliant relays censoring blocks was tolerable while optional, less tolerable when most blocks flow through two operators
two relays deciding what gets into most blocks was never sustainable. enshrined pbs is late but at least it arrives before the gas bump makes every block fatter
gus gets it. 2 relays controlling most blocks is the softest censorship vector ethereum has. enshrined pbs at least moves the chokepoint on chain where its auditable
two relays routing most blocks was never sustainable. enshrined pbs at least makes the chokepoint visible, auditable beats hidden
august through december is basically we ship when we ship, and honestly fine. letting holesky and hoodi cook beats rushing mainnet into a fork mess
no locked mainnet date but holesky and hoodi are up next. if enshrined pbs and the 200M gas limit land together thats the biggest single upgrade since the merge
enshrined pbs plus 200M in one fork is doing a lot at once. if the fork schedule slips past december i wouldnt even be mad
if 200M gas lands without state expiry the full node count falls off a cliff. access list parallelization better be load bearing
late august through december is the most honest timeline ethereum has ever published. let holesky and hoodi eat the bugs first
200M gas limit with pbs in the same upgrade, ethereum devs really said lets do both arguable things at once. holesky better find every bug