Ethereum activated the Fusaka improve on December 3, 2025 to enhance the community’s information availability capability by way of BLOB parameter overrides that incrementally develop BLOB targets and maximums.
Two subsequent changes raised the objective from 6 blobs per block to 10 to 14, bringing the utmost cap to 21. The objective was to cut back layer 2 rollup prices by growing the throughput of BLOB information, that are compressed transaction bundles the place rollups are posted to Ethereum for safety and finality.
After three months of knowledge assortment, it turned clear that there was a niche between capability and utilization. Since Fusaka’s activation, MigaLabs has analyzed greater than 750,000 slots and located that the community has fallen in need of the goal variety of 14 blobs.
After the primary parameter adjustment, the median BLOB utilization truly decreased, and the miss fee elevated for blocks containing 16 or extra BLOBs, suggesting decreased reliability on the fringe of the brand new capability.
The conclusion of the report is simple: don’t improve blob parameters any additional till excessive blob miss charges normalize and the demand for headroom that has already been created materializes.
What did Fusaka change and when did it occur?
Ethereum’s pre-Fusaka baseline, established by way of EIP-7691, focused 6 BLOBs per block, with a most of 9 BLOBs. The Fusaka improve launched two consecutive BLOB parameter override changes.
The primary one was activated on December ninth and raised the goal to 10 and the utmost to fifteen. The second was activated on January 7, 2026, growing the goal to 14 and the utmost to 21.
These modifications don’t require a tough fork, and this mechanism permits Ethereum to dial in capability by way of shopper changes reasonably than protocol-level upgrades.
MigaLabs evaluation, which printed reproducible code and methodology, tracked blob utilization and community efficiency all through this migration.
The outcomes confirmed that regardless that the community capability elevated, the median variety of blobs per block decreased from 6 earlier than the primary override to 4 after. Blocks containing 16 or extra blobs are nonetheless very uncommon, occurring between 165 and 259 instances over the whole statement window, relying on the precise variety of blobs.
The community has unused headroom.
There may be one parameter battle. The report’s timeline textual content describes the primary override as growing the objective from 6 to 12, whereas the Ethereum Basis’s mainnet announcement and shopper documentation describe the adjustment as 6 to 10.
Makes use of Ethereum Basis parameters as a supply. Baseline is 6/9, 10/15 after first override, 14/21 after second override. Nonetheless, we deal with the dataset of experiences on noticed utilization and miss fee patterns as our empirical spine.
The upper the variety of blobs, the upper the miss fee.
Community reliability measured by way of lacking slots, that are blocks that can’t be propagated or verified accurately, reveals a transparent sample.
For low blob counts, the baseline miss fee is roughly 0.5%. When the block reaches 16 or extra blobs, the miss fee will increase from 0.77% to 1.79%. On the most capability of 21 blobs launched within the second override, the miss fee reaches 1.79%, greater than 3 times larger than the baseline.
The evaluation categorizes this into blob numbers from 10 to 21 and reveals a gradual degradation curve that accelerates when the blob goal worth of 14 is exceeded.
This drop is necessary as a result of it means that the community infrastructure, akin to validator {hardware}, community bandwidth, and authentication timing, is struggling to deal with blocks which have reached their capability.
As demand ultimately will increase to satisfy the 14 blob objective or method as much as 21 blobs, the elevated miss fee can result in vital finality delays and reorganization dangers. The report summarizes this as a stability boundary. Though the community can technically deal with excessive blob blocks, whether or not it does so constantly and reliably stays an open query.
Blob Economics: Why the bottom value flooring issues
Mr. Fusaka has not solely expanded manufacturing capability. Blob pricing was additionally modified by way of EIP-7918, introducing a flooring value to stop blob auctions from collapsing to one-way.
Previous to this transformation, if execution prices prevailed and demand for blobs remained low, the bottom value for blobs might drop till they successfully disappeared as a value sign. Layer 2 rollups pay blob charges to submit transaction information to Ethereum. These charges are believed to mirror the computational and community prices that BLOB imposes.
When costs drop to close zero, the financial suggestions loop breaks and rollups eat capability with out paying proportionately. In consequence, the community loses monitor of precise demand.
EIP-7918’s flooring value ties blob charges to execution prices, guaranteeing that value stays a significant sign even when demand is low.
This prevents the free rider drawback the place low-cost blobs encourage wasteful utilization and offers clearer information for future capability selections. If blob costs stay excessive regardless of elevated capability, then demand is actual. Even for those who fall to the ground, there’s headroom.
Early information from Hildobby’s Dune dashboard, which tracks Ethereum blobs, reveals that blob charges have stabilized since Fusaka, reasonably than persevering with the downward spiral seen earlier.
The typical variety of blobs per block confirms MigaLabs’ findings that utilization has not spiked sufficient to fill new capability. Blocks usually include fewer than the 14 blob goal, and the distribution stays closely skewed towards decrease counts.
What the info reveals about effectiveness
Fusaka has efficiently expanded its technical capabilities and confirmed that the Blob parameter override mechanism works with out the necessity for a controversial exhausting fork.
The minimal value flooring seems to be working as meant, stopping blob charges from turning into economically meaningless. Nevertheless, utilization lags behind capability, and new capability is turning into much less dependable on the edge.
The miss fee curve means that Ethereum’s present infrastructure comfortably handles the ten/15 parameters of the pre-Fusaka baseline and the primary override, however begins to pressure past 16 blobs.
This creates a threat profile. As layer 2 exercise spikes and blocks usually method ~21 blobs, the community can face elevated miss charges that compromise finality and reorganization tolerance.
Demand patterns present one other sign. Regardless of the rise in capability, the median blob utilization drops after the primary override, suggesting that Layer 2 rollups are usually not at present constrained by blob availability.
Both your transaction quantity just isn’t massive sufficient to require many blobs per block, or you might be optimizing compression and batching to suit inside current capability reasonably than increasing utilization.
Blobscan, a devoted blob explorer, reveals particular person rollups posting comparatively constant blob counts over time, reasonably than growing to benefit from new headroom.
Previous to Fusaka, the priority was that restricted blob capability would grow to be a bottleneck for layer 2 scaling and rollup fees would proceed to rise because the community competed for the provision of scarce information. Fusaka talked about capability constraints, however the bottleneck seems to be altering.
The rollup just isn’t filling the obtainable area. Because of this demand has not arrived but, or different components akin to sequencer economics, consumer exercise, and fragmentation between rollups are limiting progress greater than blob availability.
what occurs subsequent
Ethereum’s roadmap contains PeerDAS, a extra basic redesign of knowledge availability sampling that additional expands blob capability whereas enhancing decentralization and safety properties.
Nevertheless, Fusaka’s outcomes counsel that uncooked capability just isn’t a binding constraint presently.
The community has room to develop to 14/21 parameters earlier than additional enlargement is required, and a reliability curve with a excessive blob rely signifies that infrastructure upgrades might have to catch up earlier than capability will increase once more.
Miss fee information present clear boundary circumstances. If Ethereum pushes capability, regardless that the miss fee stays excessive for 16+ blob blocks, there’s a threat of system instability that might floor during times of excessive demand.
A safer method is to extend utilization towards the present objective, monitor whether or not the miss fee improves as purchasers optimize for larger blob masses, and regulate parameters solely when the community reveals that it could possibly reliably deal with edge circumstances.
The effectiveness of Fusaka depends upon the symptoms. We efficiently expanded capability and stabilized BLOB costs by way of a reserve flooring. It didn’t instantly improve utilization or resolve reliability points at full capability.
This improve has created room for future progress, however whether or not that progress will materialize is an open query that the info doesn’t but reply.
