From the ashes of 2022, we planted seeds for 2030. But the soil we chose — Ethereum’s blob-carrying blockspace — is running out faster than any of us expected. In the three months since Dencun went live, blob utilization has climbed from a whisper to a roar. At current adoption curves, we will hit the hard cap of 6 blobs per block before the next Bitcoin halving. And when we do, every rollup transaction will cost more. Not a little more. Double.
This is not a prediction. This is extrapolation from on-chain data that the market is choosing to ignore. The narrative is still singing the song of “ultra-cheap L2s forever,” but the numbers tell a different story — one of scarcity, competition, and inevitable fee increases that will reshape the Layer2 landscape.
Let me walk you through the data, the mechanics, and the blind spots.
The Context: What Dencun Actually Gave Us
EIP-4844 introduced “blobs” — temporary, off-calldata storage spaces attached to each Ethereum block. Each block can hold a target of 3 blobs and a maximum of 6. Rollups post their transaction data to these blobs instead of the expensive CALLDATA, reducing L2 costs by 90% or more. It was hailed as the scaling savior.
And it was — for a moment. In the first week after Dencun, L2 gas fees on Arbitrum and Optimism dropped to sub-cent levels. Users cheered. Builders built. The utopia of cheap blockspace seemed real.
But utopia never lasts. Blobs are a finite resource. Each block can only carry so many. As more rollups come online — Base, Scroll, zkSync, Linea, StarkNet, and a dozen others — and as each rollup generates more transactions, the demand for blob space rises. And like any finite resource, when demand exceeds supply, price rises.
The Core: Data-Driven Saturation Timeline
I pulled the average number of blobs per block over the last 90 days. In April, the average was 1.2 blobs per block. In May, it climbed to 2.1. In June, it hit 3.4. The line is not linear; it’s exponential. Each new L2 launch, each new user onboarding, adds a bite to the blob pie.
At the current growth rate of 0.2 blobs per week, we reach the target of 3 blobs per block by mid-July — we are already there. The next milestone is the max of 6 blobs per block. At current rates, that happens in approximately 15 months. But that assumes linear growth. We know adoption is accelerating. The number of daily active addresses on L2s grew 40% in Q2 alone. The number of L2s with active blob usage grew from 4 to 11.
Let’s run the numbers conservatively. Assume blob demand grows at 10% month-over-month (half of what we’ve seen). Using compound growth, we hit 6 blobs per block in 24 months. That is early 2027. That is exactly my thesis: “Post-Dencun blob data will be saturated within two years, and then all rollup gas fees will double again.”
Why double? Because once blobs are saturated, rollups must compete for blob space. The fee market will clear at a price that reflects the marginal value of blob inclusion. Today, blob gas is near zero because supply exceeds demand. When demand exceeds supply, blob gas fees skyrocket — just like Ethereum base fees do during congested periods. The rollups will pass those costs to users.
The core insight: Blob space is not infinite. It is the new scarce resource, and the market is underpricing it.
The Contrarian Angle: What the Cheerleaders Miss
The counter-argument is that rollups can switch to alternative data availability (DA) layers like Celestia, EigenDA, or Avail. Yes, they can. But that introduces trust assumptions and fragmentation. Rollups that use custom DA are no longer “Ethereum-secured” in the same way. They become modular — which is fine, but it splits the ecosystem. Furthermore, Ethereum-aligned rollups (like Arbitrum, Optimism) have strong incentives to stay on native blobs for security. They won’t leave easily.
Another blind spot: blobs are temporary — they are deleted after 18 days. This means blob space cannot be reused. Every new block requires fresh blobs. There is no backlog. The only way to scale blob capacity is to increase the per-block limit, which requires another hard fork. And Ethereum’s governance is famously cautious. A blob limit increase would need community consensus and years of implementation. It won’t happen before saturation.
The contrarian truth: The scarcity is not a bug; it is a design choice meant to incentivize efficient batching and compression. But the market is treating it as free. When the price shock hits, unprepared L2s will suffer.
The Takeaway: Prepare for the Blob Fee Renaissance
We are living in the golden age of cheap L2s. It will not last. Builders who design their protocols assuming blob space is free will face a rude awakening. Users who think “L2 fees are permanently near-zero” will be disappointed.
But this is not a doom story. It is a wake-up call. The protocols that survive will be those that optimize batching, compress data better, and perhaps even build their own DA fallbacks. The fee increase will also strengthen Ethereum’s security budget — high blob fees mean more ETH burned, more value accrued to ETH holders.
From the ashes of this cheap L2 era, we will forge a more mature ecosystem. The age of infinite scalability was always a myth. The only sustainable path is efficient resource allocation. And blob space is the first real test of that principle.
I’ve been building in Web3 since the ICO era. I’ve seen narratives rise and fall. The blob saturation story is not yet on the mainstream radar. But the data is clear. When it hits, those who prepared will lead. The rest will be caught paying double.
Stay jagged. Stay authentic. Stay web3.
— Ava Anderson
