jakiro
Source: OKX Ventures
Important points
-
Pectra (Prague + Electra) is the largest hard fork bundling upgrade since The Merge, with 11 Ethereum Improvement Proposals (EIPs) designed to synchronize reducing data costs, simplify smart account operations and optimize validator operations.
-
EIP-7691 doubled the Blob target capacity to 6 and increased the hard cap to 9.This reduced the Gas charges for L1 data availability (DA) to below about 1 Gwei, and the redemption charges on major L2s had been briefly reduced to below about $0.02.
-
Every externally owned account (EOA) can now be converted into a smart account via a single transaction (EIP-7702), allowing Gas fee sponsorship, stablecoin payments, and batch user experience without contract migration.
-
The validator cap is increased to 2,048 ETH (EIP-7251), which significantly cuts DevOps costs for large operators without significantly affecting decentralization.
-
PeerDAS will be the next step in Fusaka upgrades, which is much more important to Roll-up than any single EIP in Pectra.
OKX VenturesViewpoint: Pectra itself is not a “turning point in modular data availability (DA)”; it is the last key puzzle that allows L2 to attract mass market users while Ethereum maintains its settlement dominance.The winners of the future will be those (a) externalizing all user experience friction through 7702-style accounts, (b) using the Blob’s favorable benefits to drive new data-intensive verticals (on-chain AI, order book DEX, content storage), and (c) doubling down on the stack that proves compression technology before PeerDAS.
OKX Ventures held a Twitter Space event on May 13, inviting guests who are directly involved in code writing or infrastructure operations after the fork to dive into these changes.
In this event, OKX Ventures and those who have to deliver code or run hardware after the fork goes live, took a scrutiny of these changes:
-
Derek Lee – Offchain Labs 核心协议产品经理
-
Leonardo Lerer – Core Product Manager, StarkWare
-
Qi Zhou – Co-founder of EthStorage
-
Pok Kopp – Co-founder of Ether.fi
-
Esme Zheng – OKX Ventures Investor
1. In the first week, who changed direction and who felt the shock wave?
Arbitrum – DerekNo code is required, but every Arbitrum instance must upgrade its embedded Geth client.The only EIP he really cares about is 7691: Double the Blob target capacity means more low-cost space for DEX and game users.
PectraAfter Progress:
-
BoLD’s licenseless fraud certificate has entered the test network in April and plans to go online “before the fourth quarter”.
-
Stylus compiles hundreds of Rust/C contracts per day—the llama.cpp reasoning and real-time chess engine are examples Derek shows.
-
TimeBoost has replaced the “first price auction” mechanism in the Arbitrum sorter memory pool, improving the fairness of transaction inclusion.
Starknet – LeoPectra mainly adjusted Starknet’s Blob cost model; Cairo 1.x’s fee schedule remains the same.The bigger shift lies in StarkWare’s own roadmap: The reduction in L1 data availability (DA) costs reduces urgency for “voluntary choice” (Volition, or selective off-chain data) and redirects attention to stateless client research.
Decentralization milestones:
-
Staking v2 (proof of reward by block) will be released this quarter.
-
Sorter as a Service (v0.14) – a 3f + 1 Byzantine Fault Tolerance (BFT) cluster – will be available in 2025; cheaper blobs may allow them to shorten the proof generation cycle without increasing fees.
EthStorage – Qi ZhouHigher Blob activity is a boon – his storage layer now plans to provide continuous incentives when Ethereum drops data in two weeks.Pectra also forced a full fleet-wide Geth upgrade; a group of operators using v1.13.8 was frozen in the mid-epoch.
Shockwave:
-
Node operator:Within 48 hours, 732 validators increased the staking to more than 32 ETH.
-
Roll-upuser:As the Blob Gas fee fell back to about 1 Gwei, the redemption fee on the main L2 dropped below two cents, despite the memory pool fluctuations.
-
基础设施开发者：The fast point version of Go-Ethereum is released, keeping all developers running archive nodes highly alert.
OKX VenturesViewpoint: Cheap Blobs, one-time accounts and larger validators are not headline features; they are the basic “pipelines” that will allow the next headline feature to be successfully implemented.PeerDAS will soon verify this conclusion.
two,EIPIn-depth analysis
Large-scale smart account(EIP-7702)
EIP-7702 allows any externally owned account (EOA) to behave as a smart account in a single transaction, inheriting batch call, Gas sponsorship, and stablecoin payment capabilities without contract migration.
Starknet has been in the AA (Account Abstraction) world since the Genesis Block.Leo gives an example: Productivity application FocusTree quietly deploys an account for each mobile user and begins minting an achievement NFT backed by sponsorship fees.The user didn’t even notice that they were operating on the chain.
EthStorage will use pay-masters to fund users’ first ten transactions—for example, deploying personal websites with one click through Blob.
Arbitrum already has GMX, Camelot, and Plays integrated 7702 processes through third-party AA providers.Derek expects the first direct indicator improvement to be transaction success rate and failed redemption refunds, followed by a second wave of Web2 native user import channels.
OKX Ventures Perspective: 7702 Eliminates the ultimate self-hosted user experience barrier.Areas that can be invested are those payers who perform Gas arbitrage between different tokens and chains, and secure middleware that enforces spending restrictions and fraud rules at the AA layer.
Cheaper data(EIP-7691)
Arbitrum: Every Roll-up competes for the same blob pool; doubled capacity just “provides breathing space before congestion pricing takes effect.”
Starknet only publishes status differences; cheaper blobs don’t unlock new features, but reduces the total cost per transaction and helps avoid a shift to “voluntary choice” (selective off-chain data).
EthStorage estimates that the new capacity (about 3 TB every 12 days) finally makes it possible to fix static sites less than 100 MB of them—including vitalik.ca—to be stored completely in the blob.It should be noted that the Gas fees for on-chain transactions are now a bottleneck and usually exceed the long-term storage costs.Qi is pushing block-level access lists and higher Gas restrictions to mitigate this constraint.
OKX Ventures’ view: The focus of data is redirecting to the chain; the recent potential market (TAM) is non-financial blob native content (AI reasoning weights, game assets, social graphs), and if there is a search incentive, they can tolerate blob expiration.Long-term tail search and proof markets will become critical infrastructure.
Verifier cap2,048 ETH (EIP-7251)
Data Points: So far, only 732 / 1,000,000 validators have increased stakes – no centralized panic has occurred.
Arbitrum & Starknet: Pure operational victory—lower DevOps costs, minimal impact on users.Theoretically, the risk of confiscation for each validator has increased, but Leo believes it is a “interesting academic question” rather than a practical obstacle.
EthStorage: allows running hot/cold validator copies without increasing infrastructure spending; makes large-scale blob proofs more reliable.Qi highlights the operational advantages of staking services: fewer equipment, the same benefits, and running hot standby validators without double the hardware budget.
OKX Ventures Insight: Hardware cost savings will flow to the restake and shared security markets rather than directly to end users; EigenLayer-style protocols will absorb the released liquidity.More idle ETH → More security capacity.
three,PectraPost-roadmap signal
PeerDAS (FusakaHard fork, EIP-7623)
Each speaker uses the same adjective:Existence.Derek calls it “critical and non-negotiable” – Offchain Labs already has Prism engineers involved in the specification discussion because Arbitrum’s fraudulent proof throughput is ultimately limited by DA bandwidth.Leo sees PeerDAS as “the foundation of the entire L2 roadmap”; once the sampling goes live, he can increase the proof frequency of Starknet without worrying about blob fees.Qi has written an economic foundation for that world – EthStorage incentivizes payments to peers to continue to maintain its availability after the protocol forgets to sample blobs.
VerkleTrees and historical data expire
Qi believes that these two paths are complementary.Verkle 树缩小见证数据大小并启用无状态客户端；历史数据过期通过丢弃旧区块来减少全节点一半的磁盘空间。He estimates that it will save about 50% of storage space, but that is limited to the existence of a search market – Portal + EthStorage – paying someone to save cold data.Leo is less concerned with savings, but more concerned with the functionality unlocked by the Verkle tree: a phone that can verify Ethereum in an out-of-sync state.He wanted to observe how the Beacon client handled stateless mode before porting the idea to Starknet.
SSZObject transaction(EIP-6404series)
Qi is leading the first public test network.The promises are: smaller witness data, faster decoding, and perfect alignment with the Verkle tree object hash model.Derek is currently agnostic (“Roll-up can absorb any format”), while Starknet is just monitoring – Cairo has already serialized with its own field element layout.
Contract size128 KB (EIP-7907)
If you ever split a monomer application into 24 KB pieces, you will understand the pain.Qi and Curve’s development teams are leading the patch, but the barrier is not a consensus – it is to target very large deployment transactions, requiring a DDoS-proof Gas meter.
Block-level access list
Qi’s benchmarks show that parallel preload status can reduce Geth’s IO wait time by about 70%.This in turn proves the rationality of increasing Gas restrictions, thus directly reducing the release cost of the blob.He is collecting mainnet tracking data to prove this trade-off.
OKX Ventures Insight: Pectra is a “comfortable” release; the next twelve months will focus on data availability, stateless verification, and ultimately provide developers with space to deploy large, complex contracts without the need to make workarounds.
4. Macro picture——How will all this expand the moat?
The core indicators worth paying attention to areSettled value.Leo expects that the compound growth curve of L2 throughput will surpass all other key performance indicators (KPIs), and he believes: “Every Roll-up will bring its own exponential growth.” If this statement holds, then all these settlement layers anchored by L2-Ethereum——The entire flywheel effect will be captured.
Developer retention rates seem to be finally getting healthy.Qi’s touchstone lies in the Web2 team’sAAwalletandSubsidyBlobtrial status – both did not exist in a credible form six months ago.
The security budget is about to be repriced.Derek notes that validator integration plus EigenLayer-likeRe-mortgageThe mechanism will shift the income from idle pledge to active verification services.Ether, which was once idle in personal 32 ETH hot wallets, is transforming into oracles, bridges and DSPsRemote proof income.
Is Pectra a “fulcrum”?This statement is biased.The panelists agreed that even if Blob utilization is only 40%, the large Roll-up will not turn to Celestia or EigenDA.Ethereum’s **Data Availability (DA)** remains the most cost-effective trust premium in the crypto space.Continue to expand the basic layer, and the entrusted market will solve the remaining problems on its own.
Derek’s core view is: “It’s not a complete turn, it’sDouble the investment and strive to increase Ethereum’s speed and ensureRoll-upComfort.”
Qi added: Killer User Experience Upgrade will still be prototyped on L1 first—as is true for 7702, and PeerDAS sampling will follow this path—and then cascading down to L2.
OKX Ventures Views:Highly modularRoll-upStackandContinuously expandingL1Capacity curveThe synergistic effect between them forms a flywheel that cannot be matched by any single chain.Pectra is not “save Ethereum”; it just clears the last user experience divide before PeerDAS boosts DA by 25 times.
5. Hot topics:Pectrawhether“enough”?
Roll-upofDAchoose: Unanimously agreedno—Arbitrum, Starknet and EthStorage are not planning to switch to external DAs, even if Blob utilization remains below 60%.
Non-technical leverage: Derek believes that attention should be continuedL1ExtendedandRoll-upsupport; No circulation game is required.Qi means that the L1 extension experiment (block-level access list, larger Gas restrictions) will originate from Ethereum, andBackpropagation toL2.
six,OKX VenturesFuture investment strategies
Smart Account InfrastructureWill be the default consumer portal.EIP-7702, combined with the third-party payer mechanism, will compress the two-year gap between encrypted native contract design and large-scale user guidance.OKX Ventures prioritizes investment inModular Payer Liquidity Network,Intent relayas well asAARisk Scoring Engine, enabling any Web2 app to provide “stablecoin payment of Gas fees” and “one-click registration” capabilities from day one.
BlobNative contentIt’s the next blank area.Can guarantee that the blob searches after the agreement expiresStorage excitation layerVery promising.The cheap blobs, coupled with the storage incentive layer, will create a medium where games, AI models and social media can be built entirely on Ethereum security.
Re-pledWill absorb the funds saved by the validator.EIP-7251 freed up hardware budgets and activated idle Ether.Able to convert these collateral into a measurable protocol—e.g.Oracle proof,Bridge Verification,Shared sorter collection——The excess returns will be obtained.OKX Ventures expects “investment-grade” re-staking targets to be scarce and is actively funding teams with the clearest risk-adjusted accounting.
PeerDASThat’s the real turning point.When the sampled DA is landed, Roll-up can dilute the marginal DA cost by 10 times without giving up Ethereum trust.Currently under constructionData sampling client,Proof compression circuitorDAmarketThe team will master the core tool layer after Fusaka is activated.OKX Ventures is actively looking for those who canPre-order layoutFusakaTools, Proof of Compression and Data Availability Market Projects.
