Rollups and validity-proof systems that scale Ethereum without compromising its security. The bet that the base layer should optimise for finality and the execution layer for throughput.
L2s are Ethereum's scaling answer and they finally work. The endgame thesis Vitalik articulated in 2020 — rollup-centric Ethereum — became reality through Arbitrum and Optimism's optimistic rollups, then zkSync, Scroll, and Linea's validity proofs, then Base, Coinbase's L2 built on the OP Stack that hit a billion in TVL faster than any chain in history. EIP-4844 in March 2024 introduced blobs and dropped costs from dollars to cents. The L2Beat dashboard is the de facto leaderboard nobody admits to checking.
The cultural fault line runs between optimistic and ZK. Optimistic rollups assume validity and let anyone fraud-proof; the seven-day withdrawal window is the price. ZK rollups prove validity cryptographically and settle faster but the prover stack is hard. Both camps are converging — Arbitrum is researching Stylus and BoLD, Optimism shipped fault proofs in 2024, zkSync ships native account abstraction, Scroll bytecode-equivalent zkEVM, Linea via Consensys. The Superchain (OP Stack) and Orbit (Arbitrum) are platform plays for app-specific rollups.
The honest critique is sequencer centralization. Almost every major L2 runs a single sequencer the foundation operates. Decentralizing it is on every roadmap and shipped on almost none. The Base team — Jesse Pollak's group — is open about treating it as a multi-year problem. Until then, L2s are a credible-neutrality bet on the foundations that run them. Fast and cheap is the feature. Trustless is the asymptote.
Rollups are a powerful new layer-2 scaling paradigm, and are expected to be a cornerstone of Ethereum scaling in the short and medium-term future (and possibly long-term as well). They are based on the principle of moving computation off-chain while keeping data on-chain — and proving the correctness of that off-chain computation either through fraud proofs or validity proofs.Vitalik on rollups →
Vitalik's 2020 rollup-centric roadmap committed Ethereum to scaling through L2s rather than at the base layer. Five years in, the architecture is legible. Optimistic and ZK rollups inherit security from L1 by posting state and proofs. EIP-4844 cut data costs by 10-100x. Stack frameworks let teams ship a rollup in weeks. The remaining work is decentralizing sequencers and finishing fault-proof systems on chains that still ship with training wheels. Six pieces below.
Optimistic rollups assume submitted state is valid unless someone proves it isn't. Arbitrum (Offchain Labs) and Optimism shipped mainnets in 2021. Transactions execute off-chain, batches post to L1, and a fraud-proof window — typically seven days — lets challengers prove invalid state transitions before withdrawals finalize. The seven-day window is the trust assumption: anyone watching can challenge, but users wait a week to bridge home unless they pay a third-party for instant exits. Cheap, EVM-equivalent, security inherited from L1.
ZK rollups post a validity proof — a succinct cryptographic certificate that the off-chain execution was correct — alongside the state diff. zkSync Era, Scroll, Linea, Polygon zkEVM all shipped EVM-compatible ZK rollups across 2023. No fraud-proof window: once the proof verifies on L1, withdrawals are final. The trade-off is proving cost and EVM equivalence — generating proofs is expensive, and matching every Ethereum opcode bit-for-bit is hard. The endgame for L2 security; the current cost of getting there is real.
Proto-danksharding shipped March 2024 in the Dencun upgrade. Blobs are a new transaction type carrying ~128KB of data that's available to consensus for ~18 days, then pruned. Crucially, blob data isn't accessible from the EVM — it exists for rollups to post their batches to. Blob fees price separately from gas, with their own EIP-1559-style market. Result: L2 transaction costs dropped 10-100x overnight. Step toward full danksharding, where blob count scales with validator-side data availability sampling.
Fault proofs are the mechanism that turns 'optimistic' into 'actually secure.' A challenger contests an invalid state root by submitting a proof that the L1 contract verifies on-chain. Optimism's Cannon (FPVM) shipped fault proofs in 2024, removing the multisig training wheels that Stage-0 rollups carried. Arbitrum's BoLD followed. Until a rollup ships permissionless fault proofs, withdrawals depend on a security council that can override the system. The L2Beat 'stages' framework grades chains exactly on this.
OP Stack, Arbitrum Orbit, ZK Stack (Matter Labs), and Polygon CDK turned rollup deployment into a configuration file. Base, Worldchain, Unichain, Ink all ship on OP Stack and inherit upgrades. Orbit chains settle to Arbitrum. ZK Stack chains share a hyperbridge. The thesis: L2s become commodity infrastructure, value accrues to the stack with the strongest network effects and the largest interop guarantees. The Superchain (OP Stack chains coordinating upgrades) is the most legible expression of the bet.
Every major rollup currently runs a single sequencer — usually operated by the rollup's company — that orders transactions and posts batches. That sequencer can censor, extract MEV, or go down. Espresso, Astria, Radius, and rollup-native designs (Metis, Fuel) are racing to ship shared or decentralized sequencer sets. The hard part isn't the consensus — it's preserving fast pre-confirmations users expect while removing the single point of failure. Until this ships at scale, L2s are partially trusted by construction.
Conviction is stated as conviction; you decide what to do with it. Tiers below — Core, Conviction, Watch, Speculative — reflect how much of FRQNCY's attention each project currently earns, not a recommendation to buy.
Conviction is theatre without practice. Five steps that turn the thesis above into something the body actually does, not just something the mind agrees with.
Arbitrum, Base, zkSync. Time the deposits and withdrawals. The seven-day optimistic window is real.
Stage 0, 1, 2. Sequencer, prover, upgrade keys. Most L2s are still Stage 0 and the page tells you why.
Spin up an OP Stack chain locally with op-stack-getting-started. The whole stack fits in a laptop.
Native paymaster flows. Notice the moment the unit of account stops being ETH.
GMX on Arbitrum, Aevo on its own L2. Feel how block time and finality affect execution quality.
The Crypto hub is the index of all sectors and the freedom-technology frame they share. The Fund is what happens when the same conviction gets put to work on behalf of the network.
Scaling Ethereum is a credibility test for the rollup thesis.
Cheap blockspace is necessary but not sufficient.
The endgame is many chains, one settlement layer, shared security.
Decentralize the sequencer.
FRQNCY · Crypto