Author: @yyy

Coinbase announced that it would enter the market and build an L2 based on OP Stack - @BuildOnBase. The sentiment of "Bullish on Optimism" was once again ignited. Taking this opportunity, this tweet will talk about the challenge period of Optimism.

As we all know, the withdrawal of Optimism’s second-layer funds to the Ethereum mainnet requires a 7-day challenge period.

So why exactly 7 days? 3 days? Not 5 days?

Before answering this question, let’s first understand why a challenge period is required.

Rollups improve the performance of the Ethereum mainnet by decoupling the execution layer from Ethereum. This means that transactions are executed on the second layer, but in order to inherit the security of the Ethereum main network, transactions occurring on the second layer must be verified on the main network.

Simply put, the Ethereum mainnet needs to confirm that the transaction submitted by the sequencer actually occurred on the second layer, that is, the validity of the transaction.

Rollup is divided into zk-Rollup and Optimistic Rollup due to different methods of verifying transaction validity.

Verification methods are mainly classified into two categories: zero-knowledge proof and fraud proof. Under the zero-knowledge proof mechanism, transactions have near-instant finality.

For ORUs based on fraud proof, the finality of the transaction depends on the length of the challenge period, which explains why it takes 7 days for Optimism funds to be withdrawn from the second layer to the main network.

Under the ORUs mechanism, it is optimistic that all Rollup transactions submitted by the sequencer are valid. Then during the challenge period, the verifier can initiate a fraud proof if it finds an invalid transaction.

If no one challenges after the challenge period, the transaction will be considered valid by default, and the transaction will be final at this time.

But does the challenge period really take seven days? Let’s first simplify the entire challenge process: the verifier discovers an invalid transaction by reconstructing the state root, and then issues a fraud proof. This series of actions takes time, which means that the challenge period cannot be 0.

We assume that the above challenge process is very fast, 10 blocks to complete (about 2 minutes).

Of course, it is impossible to have such a fast response, so if we slow down the time by 10 times, to 100 blocks (about 20 minutes), this time will be much shorter than 50,400 blocks (7 days).

The above discussion is based on the most ideal situation. What if someone does evil on the network? How long does it take to discover wrongdoing?

Or to put it another way, how long does the challenge period need to be extended at least to prevent evil behavior? The bad behavior here is to prevent fraudulent proofs from being received by validators by publishing a large number of transactions to spike gas over a period of time.

Here we make another assumption, assuming that there is a risk exposure of 2 billion US dollars on Optimism, and the perpetrator is willing to spend 1 million US dollars in gas tips (priority fees) per block to "front-run" transactions that contain fraud proofs.

Then, the maximum time that the network can be "hijacked" is 2000 blocks (6.67 hours). Of course, the previous examples are based on some very simple and crude assumptions.

It seems that no matter how you look at it, the challenge period does not need to be as long as 7 days. It is difficult to understand the non-linear relationship between the length of the challenge period and security.

Perhaps 7 is a very meaningful number for the founder of Optimism, just like how everyone guessed that Satoshi Nakamoto was a football enthusiast during the Bitcoin 4-year halving cycle.

So, can the challenge period for ORUs be shortened to 3-5 days without compromising security? Yes, but it’s not very optimized for user experience.

What about shortening the challenge period to 1-2 days? It seems that it can make sense to effectively optimize the user experience.

@0xMantle, an ORU incubated by BitDAO, can shorten the challenge period to 1-2 days. The optimization logic of the challenge period is based on the MPC scheme.

To put it simply, the sequencer does not directly submit the transaction batch to Ethereum, but first submits it to the MPC network, which is jointly signed by the MPC nodes. After confirming that the transaction is valid, it is submitted to Ethereum.

So the core differences are:

Optimism submits the batch containing the state root to Ethereum as the execution result;

Mantle submits batches containing state roots jointly signed by MPC nodes as execution results to Ethereum.

The purpose is to reduce the trust risk of Rollup execution results.

But in the final analysis, MPC verification is still an off-chain verification solution, which is better than nothing and provides an extra psychological comfort to users.

To use an inappropriate metaphor, the MPC verification scheme is like a decentralized zk scheme. The security of the former is based on economic assumptions, while the latter is based on cryptographic verification.

The above are some thoughts and discussions about the challenge period of ORUs. They are not very logical and should be used as suggestions. BTW, I still look forward to seeing ORUs (time-sharing levels) with significantly shortened challenge periods in the future.