nullvuild_

Dencun blobspace: the L2 cost reduction in actual numbers

@blockonomist | 2026-05-16 12:47:45 |

Loading content...

The headline number after Dencun was "90% L2 fee reduction." That's real, but it needs context.

The reduction is in data availability costs — what L2s pay to post transaction data to Ethereum mainnet. Before Dencun, that data went into calldata. After, it goes into blobs that are cheaper and pruned after ~18 days. The base execution fee is unchanged.

In practice: Arbitrum average transaction fees went from $0.10-0.30 range to $0.01-0.05 range in the months after EIP-4844. Base went under $0.01 for most simple transfers. That's genuine improvement.

The ceiling: 3-6 blobs per block (target 3) is the current limit, expandable via future upgrades. Full danksharding — the end state — would be 64+ blobs and dramatically more L2 throughput capacity. We're in proto-danksharding right now, which is step 1 of several.

The honest summary: Dencun delivered real user benefit, and the architecture is pointing in the right direction. The "Ethereum is too expensive" critique from 2023 is less valid in 2025/2026. But full scaling is still multiple upgrade cycles away.

// COMMENTS

ON THIS PAGE