Which L2 fees changed after Fusaka?

TLDR

The L2 fee that changed after Fusaka is the L1 data?availability blob cost that rollups pay to post batches. Execution gas on L2s did not change.

1. A minimum blob base fee was introduced (? 1/15.258 of L1 base fee), replacing the near?zero floor that often stuck at 1 wei per a report.
2. Blob capacity is rising in stages to 10 and then 14 target blobs per block, improving throughput and fee stability for L2s per an Ethereum post.
3. Net effect. Short?term, blob fees rose due to the new floor. Over time, more blob space and PeerDAS should lower and smooth L2 data costs as outlined here.

Deep Dive

1. Blob Pricing Floor

Fusaka added a pricing floor to the L2 data component by bounding the blob base fee to a fraction of L1s execution base fee. This replaces the old regime where blob fees often sat at 1 wei.

1. The new rule sets blob base fee at least 1/15.258 of the L1 base fee, aligning cost with actual resource use and ending free data posting by rollups per a report.
2. Ethereum core messaging also highlights blob base fee tuning for better predictability under Fusaka in an Ethereum post.

2. Capacity and Predictability

Fusakas PeerDAS lets nodes sample blob data, enabling more blob space without raising hardware requirements, and Blob Parameter Only upgrades will steadily increase blob targets.

1. The schedule moves blob targets from 6 to 10 to 14 per block, with corresponding maximums, to expand throughput and temper congestion pricing over time per an Ethereum post.
2. The broader intent is lower, more stable rollup data costs as networks adapt to higher capacity and sampling as outlined here.
3. Several explainers note that fees should become more predictable and trend down as rollups optimize under the new model as summarized here.

3. Impact Across L2s

Only the blob data component changed structurally. The L2 execution gas model is unchanged, but overall user fees depend on both parts.

1. Commentators report an initial spike in blob fees due to the new floor, with reductions expected as capacity increments arrive and operators optimize per a report.
2. Multiple analyses frame the medium?term outcome as 40 to 60 percent lower L2 data costs, subject to chain conditions and rollout cadence as noted here.
3. Projects across the L2 ecosystem describe the goal as more blob space and lower fees for users as changes phase in see this overview.

Conclusion

Fusaka did not alter L2 execution gas. It re?priced the L1 data?availability side by adding a blob fee floor and set a path to expand blob capacity. Near term, some L2s saw higher blob fees. As PeerDAS and higher blob targets take effect, posting costs should stabilize and trend down, improving end?user fees where the data component dominates.