Derived from Loopring, Taiko may be one of the big winners of the next “adoption cycle.” The original article “Taiko: A New zkEVM on the Block” was written by umede.eth and translated by jk from Odaily Star.
According to Odaily Star, Taiko, an Ethereum second-layer network based on zkRollup, completed two rounds of seed funding totaling $22 million on June 8th. The first round of funding was $10 million, with Sequoia China leading the round, and ended in the third quarter of 2022. The second round of funding, $12 million, was led by Generative Ventures.
If you have even a passing interest in the Ethereum space and zero-knowledge proofs, you’ve probably heard of names like Polygon, Starkware, Scroll, and zkSync.
In fact, you may have not only heard of them but also seen them arguing on Twitter about various issues: what exactly is the definition of zkEVM, who has ever/ is currently/ will be the first to deploy zkEVM on the Ethereum mainnet, who is better at marketing, and other important and not-so-important issues.
- Conversation with Midjourney founder: How an 11-person team is changing the world, creating products used by millions
- Blockchain-based AI computing protocol Gensyn completes $43 million Series A financing, led by a16z.
- CEX cannot escape the fate of regulation: a detailed discussion on the dilemma and opportunities of DEX
No matter what your stance on their arguments is, all these projects are working hard to push Ethereum’s scalability.
However, there is a new player on the block, one that is relatively low-key but just as viable as the others. This new player is Taiko – a zkEVM that aims to be fully decentralized and equivalent to Ethereum in terms of zero-knowledge scaling. To my knowledge, no other zkEVM team is attempting this feat, so I think Taiko is worth special attention.
Let’s take a look at what Taiko is all about.
Taiko (which means “drum” in Japanese) was founded in early 2022 by Daniel Wang, Brecht Devos, and other well-known figures in the field. Before founding Taiko, both of them worked at Loopring, which was the first zero-knowledge rollup deployed on Ethereum.
I think it’s important to quickly understand the role Loopring played in the founding and development of Taiko. Initially, the zkEVM that Taiko was building had to be a part of Loopring’s efforts to popularize zero-knowledge proof-driven Ethereum to the masses.
However, at some point during the development phase, Wnag and his team realized that conflating the two could cause a lot of confusion, as Loopring is an application-specific extension while Taiko aims to be a general extension. As a result, Taiko separated from Loopring and developed independently.
Currently, Taiko consists of about 20 employees from countries such as Turkey, Austria, Barbados, China, India, the United States, Canada, and Ukraine.
Technical Details Type-1 zkEVM
The main feature that sets Taiko apart from competitors (or more precisely, peers) is its goal of becoming a Type-1 zkEVM. Type-1 zkEVM strives to be fully equivalent to Ethereum, which means that it does not make any changes to the Ethereum system in order to generate zero-knowledge proofs more easily.
Before diving deeper, it is worth noting that Taiko’s zkEVM is a branch of the Privacy and Scaling Explorations (PSE) team’s community efforts. Taiko uses this branch to test new things and contribute to PSE’s zkEVM so that any project developing a Type-1 zkEVM (which does not currently exist) can benefit from this research.
Type-1 zkEVM hopes to be fully compatible with Ethereum. In Taiko’s case, it intends to be equivalent to Ethereum at the opcode level, which means that hash functions, pre-compiled contracts, transactions and state trees, and other consistency logic will not be changed. However, as pointed out in Taiko’s white paper, certain Ethereum Improvement Proposals (EIPs) are currently disabled in the protocol, but this will change over time.
Becoming equivalent to Ethereum brings many benefits to developers. For example, developers can seamlessly migrate their smart contracts and decentralized applications (dApps) to Taiko. This is important because if a dApp like Uniswap wants to migrate to Taiko, it does not need to make any changes to its protocol. Otherwise, rewriting the code in a different programming language than Solidity or making other changes would immediately introduce multiple risks to the protocol’s security and the assets on it.
In addition, compatibility is further enhanced in Taiko’s case. The Taiko client is based on the battle-tested Ethereum client Go-Ethereum. This means that they are more familiar and easy to use for those who want to participate. From the end user’s perspective, the ability to use Uniswap on Taiko remains consistent, accessible, and user-friendly, and even improved.
Of course, pursuing a perfect compatibility path requires some trade-offs. The main challenge faced by the type-1 zkEVM, such as Taiko, is slow zero-knowledge proof generation. Since Ethereum did not consider the integration of zero-knowledge proofs in its design, many parts of the protocol require a lot of computation to generate zero-knowledge proofs.
However, proper protocol design can mitigate this issue. Let’s see how Taiko speeds up the generation of zero-knowledge proofs.
Speeding up Zero-Knowledge Proof Generation
Taiko speeds up the generation of zero-knowledge proofs in the following ways:
All proposal blocks on Taiko L2 are immediately verified because they are deterministic and cannot be revoked. “Deterministic” means that anyone can compute the chain state after execution. Only a single intrinsic validity test is required when submitting L2 blocks to Taiko L1 for them to be considered verified.
Since all Taiko L2 proposal blocks are deterministic, they can be verified in parallel, and proofs can be submitted in any order. For example, Prover #1 can verify Block #1 at the same time that Prover #2 verifies Block #2. This means that only Taiko L1 needs to wait longer to obtain proofs.
All of these can speed up the generation of zero-knowledge proofs to some extent. For end users, this means instant finality on L2 and faster bridging from L2 to L1.
Taiko’s zero-knowledge protocol deploys two smart contracts on the Ethereum mainnet (L1) and Taiko L2:
The L1 smart contract is used to propose, prove, and verify L2 blocks.
The L2 smart contract is currently used to prove the invalidity of proposal blocks and anchor them, which is a way for the protocol to enforce certain protocol behaviors using the programmability of the EVM.
The Taiko protocol uses zero-knowledge succinct non-interactive argument of knowledge (zkSNARK) proofs. At least in theory, zkSNARK is considered to lack scalability compared to zkSTARK and requires a trusted setup process. However, zkSNARK is lighter than zkSTARK, so verification time is shorter. Also, zkSNARK requires less gas, providing cheaper transactions.
To generate zkSNARK, Taiko uses the Polynomial-based zk-SNARKs with Standard CRS (PLONK) proof system. The advantage of PLONK is that it relies on standard cryptography and has small proof size. However, Taiko is also researching the possibility of combining PLONK with other proof systems such as Plonky 2, Hyperplonk, and Halo.
The Taiko network consists of three participants:
Proposer constructs Rollup blocks of L2 user’s transactions and submits it to Taiko L1 client. They decide which transactions to include in a block and how to order them.
Prover generates zkSNARK proof to confirm the validity of L2 transactions and blocks. They decide which blocks should be verified on-chain.
Node operators execute transactions from on-chain data to keep in sync with the chain’s state. Proposers and provers also run nodes but those who want to provide other services such as block explorers need to run a node as well.
Taiko L2 Node
Taiko’s L2 nodes get transaction data from Ethereum and execute those transactions on Taiko L2. As we mentioned before, Taiko’s L2 nodes are built on Go-Ethereum.
Read Taiko’s whitepaper for more detailed technical details.
Taiko has recently released its first testnet called Snæfellsjökull, which is the name of a volcano in western Iceland. Users can test Taiko by bridging between Ethereum A1 and Taiko A1, transferring tokens between different accounts, interacting with contracts, and running proposer nodes.
Snæfellsjökull volcano. Source: Icelandic Mag.
From a testing perspective, the testnet has been a huge success. In just over a week, the Taiko L2 network has processed over 1.6 million transactions, over 650,000 blocks, and 275,000 wallet addresses. One could say that Snæfellsjökull has erupted.
Source: https://l2 explorer.a 1.taiko.xyz.
The Taiko team plans to gradually phase out Snæfellsjökull in the coming months and replace it with a new testnet that will use zero-knowledge proof technology. Further testnets will improve on previous versions and implement a token economics model.
Taiko did not specify when it will launch its mainnet, but it is more likely to be at the end of 2023 or next year.
Although Taiko is completely different in technical design from other zkEVMs in the field, its pursuit is also worth discussing.
Taiko claims that it wants to be accessible, inclusive, open, permissionless, and decentralized. In fact, it hopes to go live using a completely decentralized proposer and validator set. This is different from other zkEVMs, which mostly choose to start from centralization and then gradually achieve decentralization.
Taiko believes that complete decentralization and the pursuit of the same goals as Ethereum are necessary conditions to become Ethereum’s equivalent. In addition, the team plans to launch a DAO at some point in the future. Co-founder Wang also said that Taiko hopes to operate as a non-profit organization.
Reading Taiko’s values, it is hard not to feel very avant-garde and almost revolutionary. This reminds me of the mission statement of zkSync. I hope Taiko can fulfill its promises and become the first fully decentralized type-1 zkEVM to land on the Ethereum mainnet.
Type-1 zkEVM has its advantages and disadvantages. But if its disadvantages can be ignored relative to the advantages it provides? If it is equivalent to Ethereum, which seems to be the highest standard, and with existing and potential ways to accelerate zero-knowledge proof generation, will Type-1 zkEVM really become the winner in the long run? There are still many unknowns, but at least Vitalik seems to imply that a type-1-like zkEVM is the right choice.
As for Taiko, I think it has found the right position at the right time. 2023 seems to be a year reserved for Build. If Taiko achieves its goals and launches the mainnet this year or early next year, it may become one of the big winners in the next adoption cycle.
Taiko is clearly unique and attempting an extraordinary feat. Whether it can succeed is another question, but based on Wang and his team’s achievements on Loopring, the odds seem to be in their favor. If Taiko wins, Ethereum will also be a winner.