Go back

Michaël van de Poppe - 03 August 2022

Arbitrum vs Optimism

With the recent developments around the Ethereum hard-fork and the issue of scaling Ethereum at hand, you might have heard about layer 2’s as of late. Ethereum will need layer-2’s in order to scale and support workload associated with mainstream use of blockchains. Layer 2’s help Ethereum scale by off-loading workload from the base chain to the layer 2. 

Two of the most prominent layer 2’s are Arbitrum and Optimism. Both networks are defined as layer 2’s but differ in the way they function on the backend. In this article we will be going through both L2’s and break down their pros and cons. Strap in, it might get technical!

Source: https://l2beat.com/

What are L2’s?

Before we get into the specifics of Arbitrum and Optimism, we will have to get a basic understanding of layer 2’s in order to understand them as best as we can. 

In order to scale and support the workload associated with widespread blockchain use, Ethereum will require layer-2's. A layer 2 is an additional network or protocol that is constructed on top of an existing blockchain. A layer-2's major objective is to transfer data processing from the base chain to the layer-2.

Different types of L2’s

There are 3 forms of layer-2 scaling:

  • Rollups 

  • State Channels 

  • Plasma

Although each variety has advantages and disadvantages of its own, the rollup is the most useful and frequently used type. The term "full layer 2 schemes" refers to state channels and plasma, which attempt to transfer both data and processing off-chain. Vitalik Buterin claims that it is difficult to accomplish such safely for all applications due to "basic game theory concerns."


Rollups are ‘hybrid layer 2 schemes’. They move computation off-chain, but keep some data per transaction on-chain. This is done by bundling transactions in a batch and proving its validity through cryptography. The proof of the validity of a batch is called a fraud proof. 

Rollups also use certain types of compression to replace data with computation wherever possible, which improves efficiency. These characteristics result in a system where scalability is still limited by the data bandwidth of the base chain, but at a favorable ratio. 

There are also different types of rollups: optimistic rollups and ZK rollups. The difference lies in the method through which the rollup proves validity of the generated proof. There are complex tradeoffs between the two types of rollups. The main takeaway is that optimistic rollups are less complex, thus easier to implement, but also less efficient compared to ZK rollups. 

Fraud proofs

Because transactions are presumed to be genuine unless they are shown to be untrue, optimistic rollups are called "optimistic".

There is a window of opportunity when possibly incorrect transactions can be disputed by providing a fraud proof before optimistic rollup transactions are verified on Ethereum's main chain. 

In the case of ZK-rollups however, instead of assuming transactions are valid until proven otherwise, they use validity proofs to instantly prove if transactions are valid or not.

You can read more about fraud and validity proofs in this lovely article published by the StarkWare team: https://medium.com/starkware/validity-proofs-vs-fraud-proofs-4ef8b4d3d87a

This video also provides a solid explanation of layer 2’s with great illustrations: https://www.youtube.com/watch?v=7pWxCklcNsU

Now we have a basic understanding of layer 2 functionality we can start to discuss the differences between Arbitrum and Optimism, the two layer 2 solutions with the highest adoption at the moment. 


The Layer 2 solution Arbitrum expands the functionality of Ethereum's smart contracts. It adds more privacy features while also accelerating transaction speed and scalability overall. Optimistic Rollups are used by Arbitrum to store transactions sent to the Ethereum main chain before they are executed on a Layer 2 sidechain. This makes use of Ethereum and provides reliable results.

Source: https://defillama.com/chain/Arbitrum

Arbitrum is natively compatible with all ERC-20 tokens, but its main currency for paying fees and posting bonds is ETH. There is no Arbitrum token live as of 8/2/2022, but we can expect one soon enough.

Arbitrum also has numerous dApps which are currently live with many more planned to release in the coming months. 

Source: https://defillama.com/chain/Arbitrum


The Layer 2 network Optimism assists in scaling up the Ethereum ecosystem by utilizing optimistic rollups, just like Arbitrum. According to this Dune Analytics Dashboard, Optimism reduces Ethereum transaction fees (also known as gas fees) by 129 times. The ultimate goal of Optimism is to make transactions on the Ethereum network more sustainable. 

Source: https://defillama.com/chain/Optimism

There are already plenty of dApps available on Optimism. Even though there is less TVL compared to Arbitrum, it is still a substantial amount.

Source: https://defillama.com/chain/Optimism

Optimism launched its OP token on May 31. Holders of the OP token gain voting rights in The Optimism Collective, a two-tier governance system comprised of Token House and Citizens' House. Citizens' House will be operational later in 2022. Token House, which is already operational, governs Optimism-related technical decisions such as software upgrades. The Citizens' House makes funding decisions for public goods.

What are the differences?   

To begin, it is critical to understand that Rollups were invented by the Optimism team. As is common in the software industry, the Arbitrum team then tweaked its source code, resulting in two distinct layer 2 scalability solutions.

Fraud Proofs

The most significant distinction is that Optimism rollup employs single-round fraud proofs, whereas Arbitrum employs multi-round fraud proofs.

Why is this relevant?

Optimism's single-round fraud-proof (FP) transaction is dependent on L1 to complete the entire L2 transaction. Because of this the FP verification is instant. However, this imposes higher costs because on-chain L1 execution consumes more gas, and the L2 fee is limited by the L1 gas block.

Arbitrum, on the other hand, employs a fine-combing approach to FP verification. Arbitrum focuses on a single point of transaction disagreement by employing multi-round FP. As a result, network performance will improve. Similarly, because L2 transactions are not entirely executed on L1, the gas block limit becomes insignificant.

Dependency on Ethereum

Arbitrum and Optimism are both optimistic rollup protocols, relying on the majority of Ethereum validators to process transactions honestly. However, if Ethereum undergoes a significant consensus overhaul, re-executing L1 transactions will result in divergent final states. This is due to the fact that Optimism employs EVM (Ethereum Virtual Machine). 

Arbitrum has no such concerns because it has its own AVM (Arbitrum Virtual Machine). Although DApps from EVM must be translated to AVM, the process is automated. 

At last, Optimism is limited to a Solidity compiler. Arbitrum, on the other hand, supports all EVM programming languages, including Vyper, Solidity, Flint, YUL+, LLLL, and others.


Rollups are a powerful new layer-2 scaling paradigm that will be a cornerstone of Ethereum scaling in the near and medium term (and possibly long-term as well).

The Ethereum community is very excited about them because, unlike previous attempts at layer-2 scaling, they can support general-purpose EVM code, allowing existing applications to easily migrate over. They accomplish this by striking a key compromise: instead of going completely off-chain, they leave a small amount of data per transaction on-chain.

Despite the fact that the Optimism rollup was created first, the market has already spoken in favor of Arbitrum. Arbitrum is by far the busiest of the top five L2 networks. 

The majority of Ethereum's major DApps are already available within Arbitrum's ecosystem. You can connect to them all using the MetaMask wallet. The transaction fee difference between Ethereum's DApps and Arbitrum is massive, so it would be foolish not to use layer 2 solutions.

It seems like Optimism is behind the curve compared to Arbitrum, but you never know how the L2 wars will unfold in the future. Only time can tell.