Scalability is not computation. The primary bottleneck for Ethereum L2s like Arbitrum and Optimism is publishing transaction data to L1 for security, not executing transactions.
the-ethereum-roadmap-merge-surge-verge
Blog
Proto-Danksharding Without the Research Paper
Forget the academic jargon. Proto-danksharding (EIP-4844) is a pragmatic engineering fix that slashes L2 costs by 10-100x, making Ethereum's rollup-centric roadmap a tangible reality. This is the operational blueprint for The Surge.
introduction
THE BANDWIDTH BOTTLENECK
The Scalability Lie We've Been Sold
Proto-danksharding addresses the real constraint, which is data availability, not raw compute.
Proto-danksharding (EIP-4844) introduces data blobs. These are temporary data packets that provide cheap, high-throughput data availability without burdening Ethereum's execution layer.
This decouples data from execution. Rollups like zkSync and StarkNet will publish proofs and data blobs separately, slashing L1 fees by 10-100x.
The evidence is in calldata costs. Today, over 90% of an L2's L1 cost is data. Blobs reduce this cost by orders of magnitude, making sub-cent transactions viable.
key-trends
BEYOND CHEAP BLOBS
The Three Shifts Proto-Danksharding Enables
EIP-4844 isn't just about lower fees; it's a foundational upgrade that redefines what's possible on L2s.
01
The Problem: L2s Are Still Data-Starved
Rollups are bottlenecked by expensive, limited call data on Ethereum. This forces painful trade-offs between security, cost, and scalability.
- Security vs. Cost: Full data availability is secure but expensive, pushing L2s to use cheaper, less secure alternatives.
- Innovation Tax: High costs kill data-heavy use cases like on-chain gaming, high-frequency DEXs, and cheap social graphs.
~100x
Cheaper Data
~128 KB
Per Block
02
The Solution: A Dedicated Data Highway
Proto-danksharding introduces blob-carrying transactions, a new, cheap data channel separate from EVM execution.
- Cost Decoupling: Blob data is priced independently, preventing fee spikes from competing with DeFi transactions.
- Modular Foundation: This is the critical first step towards full danksharding, enabling a scalable data layer for hundreds of rollups.
- Immediate Relief: L2s like Arbitrum, Optimism, and zkSync can post data for pennies, passing ~10x fee reductions directly to users.
< $0.01
Per Tx Cost
Full DA
Security
03
The Shift: From Rollups to Hyperchains
With cheap, abundant data, the L2 landscape fragments into purpose-built execution layers.
- App-Specific Rollups: Games and social apps can afford their own chain with custom VMs, using shared security from Ethereum.
- Volition Becomes Viable: Hybrid data availability models, inspired by StarkEx, let users choose between on-chain (blob) and off-chain data per transaction.
- Interop Explosion: Cheap verification unlocks secure, lightweight bridges between these chains, enabling a mesh of EigenLayer AVSs, AltLayer rollups, and OP Stack chains.
1000+
Specialized Chains
Native Interop
Enabled
deep-dive
THE DATA LAYER
Blob Space: The New Commodity
EIP-4844 introduces blob-carrying transactions, creating a new, volatile market for temporary data storage on Ethereum.
Blobs are not blockspace. They are a separate resource for temporary data, priced by a dedicated fee market. This decouples execution costs from data availability, which is the primary scaling bottleneck for L2s like Arbitrum and Optimism.
Blob supply is inelastic. The protocol targets 3 blobs per block, but actual capacity adjusts slowly. This creates a volatile fee market where demand spikes from L2 batch submissions or novel data applications cause dramatic price swings.
Blob pricing is ephemeral. Data persists for ~18 days before being pruned, making blobs a pure commodity for throughput, not permanent storage. This design forces applications like EigenDA and Celestia to compete on cost for rollup sequencers.
Evidence: Post-EIP-4844, L2 transaction fees dropped 90%+. However, blob gas prices have shown 10x volatility within hours, proving the commodity's nascent and unpredictable market dynamics.
EIP-4844 IMPACT ANALYSIS
The Cost Matrix: Before and After Blobs
Quantifying the direct impact of Proto-Danksharding (EIP-4844) on Ethereum L2 transaction economics, comparing the pre-blob state to the post-blob reality.
| Cost & Performance Metric | Pre-Blob Era (Calldata) | Post-Blob Era (Blob-carrying Transactions) | Theoretical Full Danksharding |
|---|---|---|---|
Primary Data Storage Medium | Permanent Calldata | Transient Blob Data (≈18 days) | Persistent Blob Data (Sharded) |
Avg. Cost per L2 Batch (USD) | $200 - $2,000+ | $0.20 - $2.00 | < $0.10 |
Data Cost per Byte (Gas, Relative) | 16 gas (Calldata) | ~1 gas (Effective, via Blob Base Fee) | < 0.1 gas (Target) |
L2 User Tx Fee Reduction (Estimated) | Baseline (1x) | 10x - 100x | 100x - 1000x |
Throughput Target (Data per Block) | ~0.09 MB (Calldata limit) | ~0.75 MB (3 Blobs) | ~64 MB (64 Shards) |
Data Availability Guarantee | Full Ethereum Security (Permanent) | Full Ethereum Security (Transient) | Full Ethereum Security (Persistent + Distributed) |
Requires Consensus Change | |||
Status on Mainnet | Legacy (Pre-Mar 2024) | Live (Post-Mar 2024) | Roadmap (Post-2025) |
counter-argument
THE REALITY CHECK
The Skeptic's Corner: Is This Enough?
Proto-danksharding is a necessary but insufficient upgrade for Ethereum's long-term scaling.
Blob capacity is limited. EIP-4844 provides ~0.375 MB per slot, a 10x improvement over calldata. This is a stopgap, not a solution for mass adoption. Full danksharding targets 16-32 MB.
Rollup costs remain non-trivial. While cheaper, blob data still incurs a fee. For high-throughput chains like Arbitrum or Optimism, this is a variable cost that limits microtransaction viability.
The DA bottleneck shifts. The data availability problem moves from L1 to the blob market. Validators must now attest to blob availability, creating new coordination and latency challenges.
Evidence: Post-EIP-4844, L2 transaction fees dropped ~90%. However, a simple ETH transfer on Arbitrum still costs ~$0.05, far from the sub-cent target for global-scale applications.
protocol-spotlight
POST-EIP-4844 LANDSCAPE
Who Wins and Who Adapts?
Proto-danksharding (EIP-4844) redefines the L2 scaling playbook, creating new winners and forcing incumbents to pivot.
01
The L2s: From Cost-Crunch to Feature Wars
Blob data cuts L1 data posting costs by ~90%, shifting competition from pure cost to execution environment and developer UX.\n- Arbitrum & Optimism win on established ecosystems and $10B+ TVL.\n- zkSync & Starknet leverage native validity proofs for superior finality.\n- Base & Blast use cheap blobs to subsidize user growth and social apps.
-90%
Data Cost
$10B+
TVL Moats
02
The Rollup-as-a-Service (RaaS) Explosion
Cheap data availability makes launching a custom L2 viable for niche apps, fueling AltLayer, Caldera, and Conduit.\n- Enables app-specific chains for gaming and DeFi with custom gas tokens.\n- Creates a modular stack war between RaaS providers and Celestia/EigenDA for blob markets.\n- Risks fragmentation if interoperability (via LayerZero, Axelar) isn't solved.
100+
New Chains
Modular
Stack Focus
03
The Sequencer Shake-Up
Blobs expose the sequencer as the new profit center and centralization risk.\n- Shared sequencer networks like Espresso and Astria emerge to offer decentralization and MEV capture.\n- Forces incumbents like Arbitrum to formalize and potentially decentralize their sequencer sets.\n- Creates a new staking and delegation market for sequencer nodes.
New
Revenue Layer
MEV
Battleground
04
The Data Availability (DA) Layer War
Ethereum blobs are cheap but limited. Alternative DA layers like Celestia and EigenDA compete on cost and throughput.\n- Celestia wins on sovereign rollups and ~$0.01/tx DA costs.\n- EigenDA leverages Ethereum restaking (EigenLayer) for cryptoeconomic security.\n- Avail and Near DA target high-throughput use cases like gaming.
$0.01
Per Tx (Alt DA)
Restaking
Security Model
05
The Bridge & Liquidity Network Pivot
Cheap L2s multiply, making native bridging slow and expensive.\n- Intent-based bridges (Across, Socket) and liquidity networks (Connext, Circle CCTP) win by abstracting complexity.\n- LayerZero and Wormhole become critical messaging layers for omnichain apps.\n- Forces canonical bridges to improve UX or become legacy infrastructure.
Intent-Based
New Standard
Omnichain
App Enabler
06
The End-Game for Monolithic Chains
High-throughput L1s like Solana and Avalanche face pressure as Ethereum L2s match their TPS and undercut their cost.\n- Their advantage shifts to atomic composability and latency (~400ms vs L2's ~12s).\n- Must innovate on execution parallelism (Sealevel) and hardware optimization.\n- Risk becoming niche for ultra-low-latency trading (e.g., Jupiter, Drift).
~400ms
Finality Edge
Atomic Comp
Key Moat
future-outlook
THE EXECUTION
The Path to Full Danksharding
Proto-Danksharding (EIP-4844) is a production-ready data scaling layer that enables rollups to operate at a fraction of the cost, without waiting for the full theoretical construct.
Proto-Danksharding is the MVP. It introduces blob-carrying transactions and a separate fee market for data, decoupling L2 data posting costs from mainnet execution congestion. This is the core economic and architectural prerequisite for scaling.
The path is incremental, not binary. Full Danksharding requires complex data availability sampling and a new peer-to-peer network. EIP-4844 delivers 80% of the benefit by implementing the data format and fee market now, letting rollups like Arbitrum and Optimism slash costs immediately.
Evidence: Post-EIP-4844, Arbitrum transaction costs dropped over 90% for data-heavy operations. The separate blob gas market prevents L2 activity from spiking base layer gas fees for users of Uniswap or Aave.
takeaways
PROTO-DANKSHARDING
TL;DR for the Time-Poor Architect
EIP-4844 is not a scaling silver bullet; it's a foundational data availability upgrade that enables the real scaling to happen off-chain.
01
The Blob-Carrying Transaction
Introduces a new transaction type that carries large data 'blobs' (~128 KB each) which are cheap to post and automatically pruned after ~18 days.\n- Key Benefit: Decouples data cost from execution gas, enabling ~100x cheaper L2 data posting.\n- Key Benefit: Blobs are not accessible to the EVM, preventing bloating of historical state.
~128 KB
Per Blob
-100x
L2 Data Cost
02
The KZG Commitment Scheme
Uses Kate-Zaverucha-Goldberg polynomial commitments to cryptographically bind data to its commitment, enabling efficient verification.\n- Key Benefit: Allows light clients and L2s to verify data availability with a single group element (~48 bytes).\n- Key Benefit: Replaces fraud proofs for data availability, a prerequisite for future full Danksharding.
48 bytes
Proof Size
Trusted Setup
Requirement
03
The Fee Market Separation
Creates a separate fee market for blob data, distinct from standard gas for execution. This prevents congestion in one from spiking costs in the other.\n- Key Benefit: Predictable L2 costs; execution-heavy mainnet activity won't price out L2 data posting.\n- Key Benefit: Enables targeted EIP-1559 burning for blob data, creating a new deflationary pressure for ETH.
Separate
Fee Market
Targeted Burn
For Blobs
04
The L2 Scaling Catalyst
Proto-Danksharding's primary beneficiary is the rollup-centric roadmap. It directly lowers the largest cost component for Optimistic and ZK Rollups like Arbitrum, Optimism, and zkSync.\n- Key Benefit: Enables <$0.01 L2 transactions by reducing data publishing overhead to near-zero.\n- Key Benefit: Frees L2s to increase throughput without hitting mainnet data cost ceilings.
<$0.01
Target TX Cost
Rollup-Centric
Roadmap
05
The Node Operator Reality
Blobs increase the consensus-layer bandwidth and storage burden for nodes, but only transiently. This is the core trade-off.\n- Key Benefit: 18-day pruning prevents indefinite state bloat, a critical design constraint.\n- Key Benefit: Paves the way for PBS (Proposer-Builder Separation) and full Danksharding, where specialized actors handle data.
~18 days
Data Retention
PBS Path
Enabled
06
The Data Availability Bridge
This is not just an Ethereum upgrade. It's an interoperability primitive. Projects like Celestia and EigenDA pioneered the DA market; EIP-4844 brings a credible, secure DA layer to Ethereum L1.\n- Key Benefit: Establishes Ethereum as a cost-competitive DA layer, challenging external DA providers.\n- Key Benefit: Enables modular blockchain stacks (e.g., using Ethereum for DA, Arbitrum for execution) to be built natively.
DA Primitive
New
Modular Stacks
Enabled
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall