Optimistic vs ZK Rollups

A quick concept primer on two types of rollups with different security models, the pros and cons, and resources for a deeper dive

Hi, folks. Sherpa here with another dive into some strange & exciting concepts that you may have heard referenced in the cryptosphere. There are a lot of terms that you’ll hear bandied about in relation to rollups, which is a nice catch-all for scaling solutions that allow you to transact & access dapps cheaply outside of Ethereum mainnet, but in some way confirm & tie validity to mainnet in order to provide security.

It all seems like a bunch of big words & ideas, but if you can zoom out a bit you’ll see that Ethereum essentially becomes a Bitcoin “layer 2” once bridged by wrapped BTC, because bridging by its very nature adds use & value to both chains. Without the security of Ethereum mainnet, most rollups would have no more inherent value than any Ethereum testnet, but by that same stroke, Ethereum wouldn’t have a means of releasing the pressure caused by increased dapp usage & gas cost.

Related: SNARKs, and PLONKs, and Merkle Trees – oh my!

That is as deep as I’ll go into roll-ups as a concept, but you’ll find my previous articles on “bridging the gap” over to the Polygon (Matic) network linked here. Let’s move on to two specific approaches on the how of rollups: zk & optimistic

ZK-Rollups

That all sounds pretty complex (spoiler alert: it is), but essentially the elevator pitch is that you’re compressing data in the event log of a contract, which keeps bloat down while still allowing the contract to confirm that everything is kosher. This builds on the concept of “zero knowledge proofs” that you may have heard mentioned elsewhere in crypto.

You can go as deep down the zkp rabbit hole as you please, but today I just wanted to present some of these concepts & try to give you a quick ELI5 & link the resources you’ll need to dive deeper. I will, of course, be exploring these concepts further as we look at some of the projects attempting to scale Ethereum via L2 solutions in future articles.

Optimistic Rollups

Optimistic rollups are a layer 2 scaling solution. A layer 2 solution is a network that takes some of the processing work off the main (or layer 1) Ethereum chain with the aim of improving performance. The goal of optimistic rollups is to decrease latency — the amount of time it takes to confirm a transaction, currently limited by Ethereum’s block time of about 13 seconds — and increase transaction throughput — the number of transactions that can be processed each second — thereby reducing gas fees.
Contrasting with sidechain solutions, optimistic rollups still use the primary Ethereum chain for consensus and security. – stakefish

As stakefish said above, an “optimistic rollup” is an approach to scaling Ethereum, but from another angle. So what’s the difference? One of the biggest strengths of optimistic rollups stems from the fact that they do not perform computation by default, which can lead to significant scalability gains – estimates suggest that optimistic rollups can offer up to 10-100x improvements in scalability. On the downside, the need to have a challenge period means that withdrawal periods are significantly longer than ZK rollups.

Another big advantage of optimistic rollups is that they are capable of executing smart contracts, whereas ZK rollups are mostly limited to simple transactions. -Limechain.tech

There we are; a “challenge period”. Ethereum.org puts it in even simpler terms. There are two types of rollups with different security models:

• Optimistic rollups: assumes transactions are valid by default and only runs computation, via a fraud proof, in the event of a challenge
• Zero-knowledge rollups: runs computation off-chain and submits a validity proof to the chain

As with anything in crypto, you just need to be aware of the pros & cons involved; there’s always the scaling trilemma to consider. Every solution “robs peter to pay paul” in some form or fashion.

Related: zkSync by Matter Labs