Ethereum Pectra-era gas fee myths vs facts

The myth of unused capacity

I view the current blob market as a massive surplus of unused capacity. Since the Pectra upgrade on May 7, 2025, the network target for blobs per block rose to 6, while the maximum reached 9 per EIP-7691. This increase reduces the scarcity of blobspace and increases data availability capacity. Despite this expansion, the average number of blobs per block stays 33% below that new target of 6. Rollups purchased 25,600 blobs daily in the five full days after Pectra went live, yet this only uses 40% of the available 8.17GB daily capacity. Because demand remains well below the updated target rate, blobs cost rollups almost nothing. The median cost per blob object since Pectra activation sits at just $0.00000000035. This surplus creates a buffer for rollups like Base, which acts as the biggest beneficiary in terms of absolute net income after onchain costs. While rollups purchase 20.8% more blobs daily than they did pre-upgrade, they still only utilize 3.3GB of the 8.17GB available. This increase in blob volume has pushed the amount of data that consensus layer nodes must maintain between pruning to a new high estimate of 44.6GB. You know the basics, so I will skip the historical context of the Dencun upgrade.

The myth of zero-fee blobs

The claim that blobs cost zero fees is false. While blob object fees dropped by nearly 100% following Pectra, Type-3 transactions still require execution layer fees. I find the distinction between blob object fees and execution fees necessary for understanding the true cost. Rollup submitters must pay mainnet gas fees for execution and priority. On average, a blob pays between $0.50 and $3.00 in execution fees. In the 60 days leading into the Pectra update, rollups paid a daily average of $20,660 in total blob-related costs, including both object and execution fees. This compares to a much lower $11,015 daily average after the upgrade. Efficiency varies by how much data a rollup packs into a single transaction. Base and OP Mainnet remain extremely financially efficient because they post data to an EOA which requires only 21,000 mainnet gas for execution regardless of blob count. In contrast, L2s that post more complex data like Arbitrum or StarkNet use significantly more mainnet gas and tend to submit fewer blobs per transaction. The current configuration, where the minimum blobspace base fee is set to 1 wei, requires at least 30 minutes of fully saturated blocks for blobspace fees to reach $0.01 per blob and to begin to influence pricing dynamics.

Stabilizing the market

EIP-7918 aims to fix the instability when demand surges. When the blob base fee stays at 1 wei, the market lacks a proper price signal. During the LayerZero airdrop on June 20, it took six hours for the network to reach equilibrium after a surge in demand. This slow response creates a cold-start problem where the protocol cannot control the equilibrium quantity of blobs consumed. The new proposal introduces a reserve price to ensure blob fees stay relevant relative to L1 execution costs. Specifically, the function will not subtract target gas from excess gas if the price of a blob falls below the price of BLOB_BASE_COST execution gas. This prevents the base fee from dropping so low that it loses its ability to regulate demand. This mechanism ensures the blob fee’s share of the total price remains at least BLOB_BASE_COST divided by the sum of BLOB_BASE_COST and execution cost. I wonder if this reserve price will eventually trigger political pushback from well-funded L2 operators.