There is more than one way to skin a cat, so the expression goes, and there’s certainly more than one way to scale Ethereum. Discussions about the blockchain, in fact, can barely get off the ground before the word ‘scalability’ is dropped like a clanger into the conversation. The question is, what’s the right way to scale?
“Optimistic Rollups or ZK Rollups?” is the all-too-common debate, one that typically and rapidly descends into the weeds amid esoteric arguments and counter-arguments related to withdrawal times, network congestion, fees, hardware, fraud detection – the list goes on.
While both solutions have advantages, the fact that users of Optimistic Rollups have to endure a week-long wait to extract funds remains a bone of contention for many: Vitalik Buterin is on record as saying OR withdrawal times should be reduced from 7 days to 1-2 days for Stage 1 rollups like Arbitrum One and Base.
Let’s take a look at the other points of consideration.
Ethereum’s Layer-2 Landscape Faces Challenges
Ethereum Layer-2 solutions exist to reduce congestion, cut fees, and boost the eponymous network’s throughput. Rollups, meanwhile, are a class of L2s that bundle transactions off-chain to lighten the load on the main chain, before posting their commitments back to the base layer.
The emergence of Optimistic Rollups and ZK Rollups, not to mention significant Layer-1 upgrades like The Merge and The Dencun upgrade (EIP-4844), have transformed Ethereum into a robust, multi-layered ecosystem. Ironically, though, the preponderance of solution has led to even fiercer debates about the best way forward. Not just for users (traders, LPs, validators, gamers, etc) but app developers.
Introduced in 2021, Optimistic Rollups operate on the premise that all batched transactions are valid, although there is a seven-day dispute window should anyone object (hence the delayed withdrawals). Although they constitute over 90% of the total value secured by rollups, critics have questioned whether their avoidance of heavy proving costs justifies the long withdrawals and reliance on external fraud detection.
First launched by Matter Labs in 2020, ZK Rollups by contrast offer superior speed, better privacy and strong cryptographic guarantees, but at a cost: computation is resource-intensive, necessitating the use of specialized hardware, and what’s more developer accessibility is limited.
Today’s Layer-2 landscape secures around $36 billion of value across chains like Arbitrum One, Base Chain, and OP Mainnet, although scaling bottlenecks, liquidity fragmentation, and trust tradeoffs continue to present challenges. Which brings us on to the topic of security…
The Question of Rollup Security
Economic security is central to the appeal of any blockchain or related solution. While ZK Rollups offer trustless security through mathematical validity proofs, ORs incentivize external parties to flag fraud, something many consider an Achilles’ heel.
Other solutions have started to emerge, however. Fraud-proof systems like Cartesi’s Dave, for example, allow any participant to challenge incorrect executions cheaply and reliably. Thus, Cartesi’s approach is to favor Optimistic Rollups reinforced by these interactive fraud proofs.
A permissionless dispute resolution algorithm, Dave was built to address the shortcomings of traditional fraud-proof systems, introducing an innovation whereby the resources required to defend against disputes grow logarithmically with the number of opponents. In other words, defending against on-chain challenges is affordable for a single honest node even when facing down multiple attackers.
Through Dave, users are empowered to validate rollups and in so doing, improve transaction security and reliability.
Of course, Zero-Knowledge Rollups have matured in recent years themselves, and particularly in the realm of security. On the other hand, the theoretical security assumptions behind some ZK proofs have been shown to be over-optimistic. The main impediment to further ZK Rollup adoption remains cost: they are simply too expensive for low-value transfers and high-frequency consumer transactions.
While Optimistic Rollups continue to have an oversized influence, it is true that ZK Rollups have stolen some of their shine in recent years. While some attribute this to hype and flash, others believe that ZK-based networks are better suited to machine-to-machine transactions that will define the AI-Fi age.
Optimistic Rolls, for their part, may need to pair usability upgrades with embedded, permissionless protection in order to maintain their advantage.
A Question of Priorities
Ultimately, faster, secure rollups – whether Optimistic, ZK, or some other implementation – reduce friction and compel both users and institutions to interact on-chain. While ZK Rollups win on finality speed, they are complex and off-putting for some devs; Optimistic Rollups, although quick to execute, are sluggish by comparison but support highly complex workloads and superior throughput.
Perhaps the long-standing Ethereum debate and scaling race may not come down to one or the other, but which rollups are better able to balance speed, cost, and developer usability. The fight goes on.
Top comments (0)