Ethereum just tripled its block size... on a testnet.
Glamsterdam, the biggest protocol overhaul since the Merge, activated on Sepolia on Oct 6. The headline: validators can now produce blocks up to 200M gas, up from 60M. It also brings ePBS, which moves the proposer-builder handoff into the protocol itself instead of relying on outside relays.
Before you price this in, three things:
- It's a testnet, and the 200M limit is opt-in. Mainnet's gas limit hasn't changed, no mainnet date has been set, and the next stop is another testnet, Hoodi.
- The rollout needed a last-minute fix. Hours before activation, Prysm shipped v7.2.1 because older versions would quietly keep producing 60M blocks.
- Bigger blocks have a cost that doesn't show up in TPS charts. At 200M gas with current pricing, state growth could exceed 380 GiB a year, enough to push consumer hardware past its comfort zone in under 12 months. Developers are working on a state-byte pricing fix targeting about 120 GiB a year, but that's still a proposal.
That last point is the real debate. Ethereum's edge has always been that ordinary people can run nodes. Scale L1 too fast and you quietly trade that away.
The fair counterpoint: parallel validation means bigger blocks don't take proportionally longer to verify, and testing on Sepolia first is exactly how this should be done.
Should Ethereum scale the L1 hard, or keep blocks small and leave the volume to L2s?
Follow for protocol news without the hype, and drop a like if this helped.
#Ethereum #Glamsterdam
Glamsterdam, the biggest protocol overhaul since the Merge, activated on Sepolia on Oct 6. The headline: validators can now produce blocks up to 200M gas, up from 60M. It also brings ePBS, which moves the proposer-builder handoff into the protocol itself instead of relying on outside relays.
Before you price this in, three things:
- It's a testnet, and the 200M limit is opt-in. Mainnet's gas limit hasn't changed, no mainnet date has been set, and the next stop is another testnet, Hoodi.
- The rollout needed a last-minute fix. Hours before activation, Prysm shipped v7.2.1 because older versions would quietly keep producing 60M blocks.
- Bigger blocks have a cost that doesn't show up in TPS charts. At 200M gas with current pricing, state growth could exceed 380 GiB a year, enough to push consumer hardware past its comfort zone in under 12 months. Developers are working on a state-byte pricing fix targeting about 120 GiB a year, but that's still a proposal.
That last point is the real debate. Ethereum's edge has always been that ordinary people can run nodes. Scale L1 too fast and you quietly trade that away.
The fair counterpoint: parallel validation means bigger blocks don't take proportionally longer to verify, and testing on Sepolia first is exactly how this should be done.
Should Ethereum scale the L1 hard, or keep blocks small and leave the volume to L2s?
Follow for protocol news without the hype, and drop a like if this helped.
#Ethereum #Glamsterdam
