The encryption economy is not just a technological revolution

The encryption economy is far more than just a simple technological revolution. Many practitioners and enthusiasts find it difficult to adapt to this field, primarily because its cycles are too pronounced and the fluctuations are extremely intense. In a bull market, ordinary people do indeed have the opportunity to achieve tenfold or even hundredfold returns, but losses in a bear market can be equally severe. Regardless of who it is, once entering the encryption field, investment will become a required course.

The main reason for the current bear market is the capital tightening caused by the Federal Reserve's interest rate hikes. The crazy yields of decentralized finance ( DeFi ) are a thing of the past, and the yields of mainstream lending protocols are generally below 2%. In contrast, the yield on US Treasuries in the real world has exceeded 3%, prompting institutional investors and stablecoin projects to continuously transfer funds from the encryption field to purchase government bonds.

Since the 1980s, the Federal Reserve has experienced a total of 6 interest rate hike cycles, each lasting 1-3 years, with an average of up to 10 hikes. The slower the pace of rate hikes, the worse the effects, making it harder to curb inflation. For example, the 6th interest rate hike process from 2015 to 2018 was slow, and the prices of raw materials, led by crude oil, rose instead of falling. After October, crude oil prices remained strong, which may prompt the Federal Reserve to continue its aggressive rate hike policy. There is no doubt that in 2023, global stock markets and the encryption market will continue to face pressure from institutional capital withdrawal, the bottom of the bear market has not yet arrived, and it is not advisable to bottom-fish easily.

Optimistic about Ethereum zkRollup scaling solution

Putting aside the large cycle of the capital market, from the perspective of the transformation of encryption technology and the potential wave of Web 3.0 it may trigger, we are firmly optimistic about Ethereum and its zkRollup expansion solutions, especially the zkRollup solutions based on the new generation zkEVM.

Ethereum founder Vitalik Buterin stated at a conference on September 30: "After the merge, Ethereum's next step is scalability." Scalability is a core issue that hinders many cryptocurrencies and blockchain applications from becoming mainstream. It is well known that zkRollups can achieve scalability by bundling hundreds of transactions into a single execution task and verifying all transactions within that single task.

With Vitalik's strong support, Rollup has become the mainstream solution for Ethereum's scalability. Rollup solutions can be divided into Optimistic Rollup and zkRollup based on their technical types, and the main difference between the two lies in the method of ensuring transaction validity. Optimistic Rollup uses a fraud proof scheme, while zkRollup utilizes mathematical zero-knowledge proofs.

In the Optimistic Rollup network, there is a challenger role that can prove the existence of fraud in the data submitted to Ethereum, and then roll back invalid transactions through network consensus. In contrast, zkRollup uses zero-knowledge proof technology when batch processing transaction data, directly submitting proof to Ethereum based on the guarantee of the validity of transaction data, achieving immediate consistency of the final state.

Compared to Optimistic Rollup, zkRollup uses zero-knowledge proofs for mathematical verification, offering more technical advantages. Certain projects have been engaging in forward-looking exploration in this field for many years.

However, the EVM was not designed to support zero-knowledge proofs, which makes it very difficult to build a virtual machine that is compatible with Solidity and supports zero-knowledge proofs. To address this issue, some projects are developing virtual machines that support zero-knowledge proof computation and are compatible with Solidity, namely zkEVM. Unlike ordinary virtual machines, zkEVM can prove the correctness of execution, including the validity of inputs and outputs used during execution.

The Necessity of Redesigning zkEVM

Some projects compile bytecode into micro-operation code, use STARK to generate validity proofs for state transitions, and submit them to Ethereum for verification after using SNARK to validate the correctness of the proofs. Other projects have schemes that are somewhat similar but only use Halo 2 as their zero-knowledge proof method.

There are also projects that compile contract code written in Solidity into Yul(, an intermediate language that can be compiled into bytecode) for different virtual machines, and then recompile the Yul bytecode into a customized, circuit-compatible bytecode set specifically designed for its zkEVM.

Are these systems good enough to be put into production? Do we need to redesign a better zkEVM? zkRollup typically uses zero-knowledge protocols to prove and aggregate all transactions before publishing the aggregated proof to the chain. In principle, this means that the layer 1 chain can verify a short "proof" covering thousands of complex transactions, with no possibility of cheating. However, after the release of certain zkEVM testnets, it was found that the reality is not ideal—the process of generating proofs is very slow, taking dozens of minutes to execute a few transactions.

Some projects have optimized the structure of zkEVM and redesigned it to make it more efficient than all existing zkEVMs. The main reason is that it adopts a well-designed layered structure, compressing the redundant wasted space in the circuit and the size of the commitment polynomials, ultimately shortening the time required to generate proofs. At the same time, its Sequencer runs an Ethereum node, receiving user transactions, generating new states, and creating special zkEVM-friendly Traces. The Folder acts as a proof generator, obtaining this Trace from the Sequencer and using a large number of small tables ( instead of one large table ) to process it, which will greatly reduce redundancy and improve the speed of proof generation.

The Necessity of Faster Zero-Knowledge Proofs

SNARK is concise, but its efficiency is not as good as STARK. However, STARK seemed outdated when FOAKS emerged. Since speed is the bottleneck of zkEVM, comparing computational efficiency is of significant importance. STARK achieves quasi-linear proof and verification times, faster than SNARK, but noticeably slower than FOAKS. FOAKS is the world's first ZKP to achieve linear proof time and sub-linear verification time, reaching theoretical extremes. FOAKS is transparent and does not require any trusted setup, which means it maintains the highest level of security.

FOAKS is based on linearly time-encodable codes, making it the fastest among all existing ZKP schemes. Moreover, due to the use of recursive techniques, the proof size is reduced to 1/7 of existing schemes, allowing end users to enjoy Ethereum Layer 2 services for just one cent.

The Necessity of Independent Data Availability Layer

Currently, zkRollup mainly focuses on reducing the computational burden of validating transactions. This is particularly important for Ethereum, as validating the execution of complex smart contracts is very costly. However, Ethereum nodes still need to store the original transaction data simultaneously. This is not wise, as Ethereum is better suited to serve as a consensus layer rather than a storage layer, which means that scalability bottlenecks still exist—when nodes are impacted by insufficient bandwidth and storage ( rather than insufficient computation ).

This is why Ethereum needs a separate data availability layer to store these raw transaction data, to prevent the freezing of the entire smart contract due to zkRollup server or Ethereum node failures. More importantly, it decouples the costs of layer two from layer one, and further reduces the transaction costs of zkRollup based on zkEVM by over half.