On March 23rd, Arbitrum token ARB, an Ethereum Layer2 Rollup solution project, started airdrop, and so far, Arbitrum TVL has reached $2.35B, far ahead of competitor Optimism's $891.41M TVL. This article will discuss Arbitrum's technology, ecosystem and tokens, and look into the reasons for its popularity and investment opportunities.

*This article is intended as a research discussion only and may contain biases and errors in facts, data, and opinions, and does not constitute any investment advice and does not assume legal responsibility.

♦ In response to the current situation of high transaction fees and slow transaction speed of Ethereum, various kinds of scaling solutions are blossoming. Among Layer2 scaling solutiions, Vitalik’s favorite one is Rollup - it executes transactions by packaging transaction data and transferring them off Ethereum, while the assets remain on-chain. Currently, the two mainstream Rollup solutions include Zero-Knowledge Rollups (zk-rollups) and Optimistic Rollups, and Arbitrum belongs to one of the Optimistic Rollups projects.

♦ Arbitrum consists of two chains: Arbitrum One and Arbitrum Nova. Core highlights of Arbitrum One include its fraud proof mechanism and strong EVM equivalence - users can continue to use Metamask on Arbitrum, and developers can deploy contracts in Ethereum libraries and tools, but its sequencer is currently run by Arbitrum’s development team and must be further decentralized in the future; Arbitrum Nova sacrifices security in favor of achieving lower gas fees for gaming and social applications.

♦ Arbitrum's token ARB is only used as a governance token, and after conducting a relative valuation analysis with its competitor Optimism, it is found that the value of Arbitrum project may have been overvalued in the long run, and the difference between the circulated market value and FDV is large, which means ARB token will face a large selling pressure after one year; in addition, from the indicators of MCap/TVL and FDV/TVL, the ARB token price still has a small upside trend in the short term.

♦ Arbitrum ecosystem currently has more than 170 projects, with Defi projects such as GMX and Radiant as the main ones. Except for Treasure, other game and NFT projects are still at extremely early stage, and NFT transactions volume are also very low.

♦ Compared with Optimism, Arbitrum has better data performance, a stronger team execution and faster development progress; however, the downtime during Arbitrum airdrop exposes its current performance still needs to be improved, and the AIP-1 proposal controversy has significantly reduced the community's confidence in Arbitrum.

1、Layer2 Overview 2、How Does Arbitrum Solve the Scaling Problem? 3. ARB Tokenomics and Valuation 4、Milestones 5、Team and Financing 6、Ecosystem Overview 7、Competitive Landscape 8、Opinion and Summary

Ethereum is one of the strongest Layer1 in the blockchain world today, but as the number of users increases and the volume of transactions proliferates, Ethereum currently allows only about 20-40 transactions per second (TPS); when the cap is reached, users are forced to pay even higher transaction fees to have their transactions executed as quickly as possible. This scalability problem arises because Ethereum requires that transactions must be verified by multiple nodes before completion and must reach consensus on the current state of the transaction, which requires each full node to keep a complete blockchain history ledger. And this has led to the cost of running and verifying nodes being too expensive, thus further weakening the decentralization property.

Various solutions have emerged for Ethereum scaling, including Layer1 and Layer2 scaling solutions. Layer1 scaling solutions such as sharding, directly modify Ethereum; Layer2 scaling solutions run on top of Ethereum. The following is a classification of common Layer2 solutions:

Among those Layer2 scaling solutions, Ethereum’s founder, Vitalik’s favorite one is the Rollup solution, and so far, Rollup is indeed the fastest growing and most popular, which executes transactions by packaging the transaction data and transferring it off Ethereum blockchain, while the assets remain on-chain. The transaction data is then sent back to Ethereum after processing is completed. Once Layer1 is strong enough, most transactions can be performed on Layer2, allowing for greater scalability and faster transactions with lower gas fees. The key to Rollup is that it must ensure it can inherit the security performance of Layer1. The two mainstream Rollup solutions include Zero-Knowledge Rollups (zk-rollups) and Optimistic Rollups, and the Arbitrum we discuss in this paper is one of the Optimistic Rollups. The two types of Rollups should be able to scale the TPS of Ethereum from 15 - 45 to 1000 - 4000. A brief description of the two factions is as follows.

Zero-knowledge rollups are implemented by leveraging a clever cryptography called zero-knowledge proof, hence the name ZK rollups. In this model, each batch of files sent to L1 includes a cryptographic proof called ZK-SNARK. When a transaction batch is submitted, this proof can be quickly verified by the L1 smart contract, and invalid batches can be rejected outright.

As its name "optimistic" suggests, Optimistic Rollups publish data to Layer1 and assume that the data is correct. If the posted data is indeed valid, it can continue to be "optimistic" and does not need to do anything additional. However, if an invalid transaction occurs, the system must be able to recognize it, restore it to its correct state and penalize the party that submitted the transaction. To achieve this, Optimistic Rollups has introduced a dispute resolution system that validates fraudulent proofs, detects fraudulent transactions, and discourages bad actors from submitting other invalid transactions or incorrect fraudulent proofs. A party able to submit batches of transactions to Layer1 must provide margin (usually ETH). Any other network participant who discovers an incorrect transaction can submit a proof of fraud. After the proof of fraud is submitted, the system enters dispute resolution mode. In this mode, the suspicious transaction will be executed on Ethereum again. If the execution proves that the transaction is indeed fraudulent, the party that submitted it will be punished, usually by cutting its margin. To prevent bad actors from disrupting the network with incorrect fraudulent proofs, the party submitting the fraudulent proof must also usually provide a margin that can be cut. The system works as expected and detects fraud even if only 1 honest party monitors the status of the rollup and submits fraud proofs when needed. Also, with the right incentives within the Rollup system, entering the dispute resolution process should be the special case, not the norm.

The shortcoming of Optimistic Rollups is that in order to implement their dispute resolution mechanism, all network participants must have enough time to submit proof of fraud before finalizing the Layer1 transactions. This time is typically quite long to ensure that even in the worst case scenario, fraudulent transactions can still be disputed, which can lead to rather long withdrawal times, with users having to wait up to a week or two to withdraw their funds back to Layer 1. However, there are already several projects (Hop protocol and Connext) that are working to improve this situation by providing fast "liquidity exists" to improve this situation, allowing users to withdraw funds back to Layer1 instantly, but for a small fee.

In contrast, ZK rollups do not have the problem of long withdrawals. The core difficulty is that it is much harder to create an EVM-equivalent ZK rollup due to the complexity of the technology, which makes it more difficult to scale a generic application without rewriting the application logic. However, ZKSync is making significant progress in this area and they may soon be able to launch a ZK rollup solution that is equivalent to EVM. Optimistic Rollups are a bit easier in terms of EVM equivalence. They still need to run their own version of EVM with some modifications, but 99% of the contracts can be transferred directly without any modifications. Moreover, ZK rollups are also much more computationally intensive than Optimistic rollups, which means that the nodes computing the ZK proofs must be high-spec machines and it is difficult for regular users to run them.

Arbitrum is an Optimistic Rollups solution that allows users to use dapps, deploy smart contracts, etc. on Arbitrum, with lower transaction fees and faster transactions, while using Ethereum to ensure the security of transactions. Currently Arbitrum runs two parallel chains based on Ethereum: Arbitrum One and Arbitrum Nova, whose core highlights are as follows.

The core feature of Arbitrum One is the ability to detect fraudulent transactions on Layer1. A validator can send stateRoot on the Arbitrum chain to Layer1's contract by running open source validation software, and while the average user is likely not interested in becoming a validator, the key point is that anyone can become a VALIDATOR and the chain will remain secure as long as there is an honest validator; that is, it only takes one non-malicious fraudster to catch any number of malicious ones. Together, these properties make the system "trustless" - users do not rely on any particular designated party to keep their money safe.

Such a "proof of fraud" mechanism is at the heart of Arbitrum's ability to reduce transaction costs; in addition, there are a number of other mechanisms designed by Arbitrum to help reduce the burden on the L1. First, Arbitrum transactions are submitted to the L1 in batches; typically, a batch (submitted in one L1 transaction) will contain hundreds of L2 transactions, and submitting them in batches amortizes the overhead of interacting with the L1, thus saving a lot of money compared to submitting a single transaction at a time; furthermore, transaction data is published in a compressed form on the L1 (decompressed only in the L2 environment), making the transactions take up less L1 space.

In Arbitrum, the experience is essentially the same as in Ethereum, except that transactions are faster and cheaper, as Arbitrum has had EVM equivalence as a feature at the forefront of its development from the beginning. Users can continue to use Ethereum wallets on Arbitrum, and developers can deploy contracts in Ethereum libraries and tools. In February, Arbitrum also announced the launch of Stylus, a next-generation programming environment that allows developers to use programs written in Rust, C and C++ through WebAssembly, which is expected to attract more Web2 developers. To achieve strong EVM equivalence, Arbitrum One uses and improves upon the fork of Geth by turning it into a Layer2 chain, meaning that most of the code running on Arbitrum is identical to that running on Ethereum.

In September 2022, Arbitrum One underwent a Nitro upgrade with the following major improvements:

• Higher throughput: 7 to 10 times higher than before;

• Advanced call data compression: driving down transaction costs for Arbitrum by reducing the amount of data posted to Ethereum;

• Ethereum L1 gas compatibility: EVM pricing and accounting is consistent with Ethereum;

• Additional L1 interoperability: including tighter synchronization with L1 block numbers;

• Geth tracing: broader debugging support.

With the introduction of Nitro, transactions are now processed in two stages. Nitro first places the transactions into a sequence that will be processed, then publishes the sequence and applies a deterministic state transition function to each transaction in the sequence. Currently, the sequencer is a centralized component operated by Offchain Labs, which will gradually transition to a distributed system as Arbitrum enters its later stages. While the centralization tradeoff in pursuit of lower gas costs is understandable in the short term, the creation of a decentralized system is imperative.

Arbitrum Nova is the AnyTrust chain launched by Arbitrum. Rollup and AnyTrust are similar in many ways, but there is one key difference: on Rollup, all data is published on L1 (allowing anyone to join as a validator without permission), while on Anychain, data are managed offchain by the External Data Availability Committee, whose members include Offchain Labs, Consenys, Google Cloud, P2P, Quicknode and Reddit. In case of a challenge, the AnyTrust chain reverts to Rollup. Arbitrum believes that AnyTrust may be a wise tradeoff for applications that require high transaction throughput and do not require Rollups to provide full decentralization (e.g., gaming, social applications). In other words, Nova reduces transaction costs at the cost of security.

The initial total supply of ARB is 10 billion, with a maximum inflation of 2% per year. The current price is stable at around $1.1, and the tokens are already listed on Binance, Coinbase and other first-tier exchanges.

As Arbitrum transitions to a decentralized autonomous organization (DAO), the token will be used exclusively for governance. Holders can use the token to vote on governance proposals to decide on key issues, including fees, token unlocking and inflation rates. Currently the gas fee is still paid using ETH on Arbitrum. 56% of the total is accounted for by the Arbitrum community and the airdrop on March 23rd has distributed 11.5% of the total supply to eligible users, with the other 44% going to investors and employees of Offchain Labs, the founding team of Arbitrum.

The token unlocking period is shown below, with institutional investors and teams starting to unlock after one year, followed by three years of linear monthly unlocks.

From the perspective of P/S Ratio, Arbitrum is higher than Optimism, which means that the value of Arbitrum may be overvalued in the long run. Although the ARB tokens held by investors will be unlocked after one year, there is no selling pressure in the short term. However, the difference between the circulated market value and FDV is large, and the latter will still face greater selling pressure. In addition, from the perspective of MCap/TVL and FDV/TVL, as the TVL of Arbitrum is far ahead of Optimism, both indicators are lower than OP, and the coin price still has a small upside trend in the short term.

The development team behind Arbitrum is offchain labs, whose core founders include Ed Felten, former White House CTO and professor of computer science at Princeton University; Steven Goldfeder, PhD in computer science at Princeton University, postdoctoral fellow at Cornell University, and former Google intern; and Harry Kalodner, also a computer science PhD at Princeton University. Among them, founder Ed Felten's experience and background are most notable. Back in 2014, before the birth of Ethereum, he decided to research blockchain scaling solutions after seeing the potential congestion in the Bitcoin network. Currently Offchain labs remains small and lean, with a team of 59 people, mainly engineering and business development employees, mostly from Princeton, CUHK, and Harvard.

Offchain labs raised $123.7M in total according to Crunbase's statistics.

• 2019.4: Seed Round, $3.7M, Pantera Capital led the investment, Coinbase Ventures participated in the investment

• 2021.4: Series A, $20M

• 2021.8: Series B, $100M, Lightspeed Venture Partners led the investment, with participation from Redpoint, Polychain, MCE, Alameda Research and others.

According to Messari's summary, the Arbitrum ecosystem is mainly divided into Defi system and other consumer applications.

The Defi ecosystem is based on DEX, lending and perpetual contracts and the venue is already quite crowded.

6.1.1 DEX According to Defi Llama's statistics, just a few months ago, DEX activity on Arbitrum surpassed Polygon and was second only to Ethereum, and the DEX protocols on Arbitrum are mostly similar to Ethereum. The project currently posing the biggest challenge to older DEXs such as Uniswap is Camelot (https://excalibur.exchange/). Each pool deployed on Camelot has customizable fee parameters, and a core feature of the product is its more targeted customizability; for example, USDC-WETH pools can charge 0.04% for transactions that purchase WETH from the pool, but only 0.02% for transactions that sell WETH into the pool. This directional feature allows the pool deployer to fine-tune the incentive structure for a given pair. In addition, these fee levers are dynamic, which allows the fee structure to change based on factors such as market conditions.

6.1.2 Lending Protocols Radiant (https://radiant.capital/) goes live on Arbitrum in July 2022 and is built on LayerZero to create a full-chain lending protocol. Radiant currently accounts for 57% of the TVL of the Arbitrum lending venue, a figure that is largely driven by the large number of RDNT incentives, which offers a one-click "cycle and lock" feature that allows users to re-deposit borrowed assets as collateral and then receive loans from the agreement up to five times. The longer the user pledges, users can earn more RNDT rewards. So beyond the basic double-digit lending APR, users can also receive more benefits from the agreement lending for more benefits, which so far looks like a very successful growth strategy for Radiant. Since the beginning of 2023, the price of RDNT has increased by 1,075%, which makes its incentives even more attractive.

6.1.3 Perpetual contracts Exchange hybrid models that combine AMM and order thinning are gaining popularity in Defi, and such platforms have been a key driver of early growth in the Arbitrum ecosystem. The current perpetual contracts venue holds about $565 million in TVL, the largest share of the Arbitrum Defi ecosystem. Of this, GMX (https://gmx.io/) accounts for over 80% of the TVL.

The key to the success of GMX, which was the first to go live with Arbitrum One in August 2021, is the dual token model: 30% of the protocol's revenue is allocated to stakeholders of GMX (the protocol's governance token) and 70% to stakeholders of GLP (the platform's liquidity token). Such a mechanism helps to build a complete ecosystem around the protocol, thus bringing greater liquidity to GMX.

Most of the tracks on the Arbitrum ecosystem other than Defi are still very early and have not yet reached the same level of volume as Defi applications.

6.2.1 Games Arbitrum's gaming track is represented by the Treasure protocol (https://treasure.lol/). Treasure is an ecosystem for game developers and aims to be the "Nintendo of the Web3". The games released on Treasure are directly interoperable and currently support 10+ games that can be traded using its native DEX MagicSwap and native NFT marketplace Trove, and its native governance token is MAGIC. The success of Treasure so far can be attributed mainly to its first-mover effect and the fact that there are currently no competitors of its kind in the Arbitrum ecosystem.

6.2.2 NFT Arbitrum's NFT ecosystem is still very early, with only 3,900 ETH of NFT transactions in the past month, compared to 1.3 million ETH of NFT transactions on Ethereum in the same period.

According to the CMC data, the current status of the Layer2 venue can be seen as follows:

The advantages and disadvantages of the two have been discussed in Part 1 of this paper. The core advantages of OP-Rollups including Arbitrum and Optimism are their EVM equivalence, which can be used directly on MetaMask, and their first-mover advantage, which has attracted many star projects such as Sushi, and may form an ecological barrier. In contrast, ZK-Rollup’s technology is quite difficult, does not have EVM equivalence , and the supporting ecosystem is weak, but the progress of ZK infrastructure is still worth noting; and from the fundamental technology point of view, if ZK path can be realized, it will kill OP-Rollup. From the investment point of view, we can still pay more attention to the OP-Rollups in the short term.

Here we focus on a multi-dimensional comparison between Optimism and Arbitrum, the representative projects in the OP family. While Arbitrum and Optimism both attempt to achieve the same goal, the designs differ. First, the dispute resolution model is designed in such a way that instead of re-running the entire transaction at Layer1 to verify whether the fraud proof is valid, Arbitrum proposes an interactive multi-round model that allows narrowing the scope of the dispute, possibly by executing only some instructions at L1 to check whether the suspicious transaction is valid, which also allows smart contracts deployed on Arbitrum to be larger than the the maximum contract size allowed; another major difference is the approach to deal with transaction sequencing and MEV. Arbitrum initially runs a sequencer responsible for transaction sequencing, but in the long run they want to decentralize it. Optimism prefers an alternative approach where the sequencing of transactions, and the resulting MEV, can be auctioned off to other parties over a period of time.

Offchain Labs made a presentation on Arbitrum's future development path in SmartCon 2022 speech. Around solving the problem of decentralization, the project will make upgrades to verifiers and sequencer, etc., and Odyssey will also restart this year. From the perspective of the Layer2 Rollup track, it is difficult for ZK-Rollups to catch up with OP-Rollups in the short term, and Arbitrum has crushed its competitors in all aspects of business performance, so its ecosystem projects are worth paying attention to in the short term. For example, ZKSync may soon be able to launch a ZK rollup solution with EVM equivalence. In addition, the new public chain Sui in the Move line. Mochain will keep an eye on the Layer2 venue and continue to output related research reports.

On March 23rd, Arbitrum token ARB, an Ethereum Layer2 Rollup solution project, started airdrop, and so far, Arbitrum TVL has reached $2.35B, far ahead of competitor Optimism's $891.41M TVL. This article will discuss Arbitrum's technology, ecosystem and tokens, and look into the reasons for its popularity and investment opportunities.

*This article is intended as a research discussion only and may contain biases and errors in facts, data, and opinions, and does not constitute any investment advice and does not assume legal responsibility.

♦ In response to the current situation of high transaction fees and slow transaction speed of Ethereum, various kinds of scaling solutions are blossoming. Among Layer2 scaling solutiions, Vitalik’s favorite one is Rollup - it executes transactions by packaging transaction data and transferring them off Ethereum, while the assets remain on-chain. Currently, the two mainstream Rollup solutions include Zero-Knowledge Rollups (zk-rollups) and Optimistic Rollups, and Arbitrum belongs to one of the Optimistic Rollups projects.

♦ Arbitrum consists of two chains: Arbitrum One and Arbitrum Nova. Core highlights of Arbitrum One include its fraud proof mechanism and strong EVM equivalence - users can continue to use Metamask on Arbitrum, and developers can deploy contracts in Ethereum libraries and tools, but its sequencer is currently run by Arbitrum’s development team and must be further decentralized in the future; Arbitrum Nova sacrifices security in favor of achieving lower gas fees for gaming and social applications.

♦ Arbitrum's token ARB is only used as a governance token, and after conducting a relative valuation analysis with its competitor Optimism, it is found that the value of Arbitrum project may have been overvalued in the long run, and the difference between the circulated market value and FDV is large, which means ARB token will face a large selling pressure after one year; in addition, from the indicators of MCap/TVL and FDV/TVL, the ARB token price still has a small upside trend in the short term.

♦ Arbitrum ecosystem currently has more than 170 projects, with Defi projects such as GMX and Radiant as the main ones. Except for Treasure, other game and NFT projects are still at extremely early stage, and NFT transactions volume are also very low.

♦ Compared with Optimism, Arbitrum has better data performance, a stronger team execution and faster development progress; however, the downtime during Arbitrum airdrop exposes its current performance still needs to be improved, and the AIP-1 proposal controversy has significantly reduced the community's confidence in Arbitrum.

1、Layer2 Overview 2、How Does Arbitrum Solve the Scaling Problem? 3. ARB Tokenomics and Valuation 4、Milestones 5、Team and Financing 6、Ecosystem Overview 7、Competitive Landscape 8、Opinion and Summary

Ethereum is one of the strongest Layer1 in the blockchain world today, but as the number of users increases and the volume of transactions proliferates, Ethereum currently allows only about 20-40 transactions per second (TPS); when the cap is reached, users are forced to pay even higher transaction fees to have their transactions executed as quickly as possible. This scalability problem arises because Ethereum requires that transactions must be verified by multiple nodes before completion and must reach consensus on the current state of the transaction, which requires each full node to keep a complete blockchain history ledger. And this has led to the cost of running and verifying nodes being too expensive, thus further weakening the decentralization property.

Various solutions have emerged for Ethereum scaling, including Layer1 and Layer2 scaling solutions. Layer1 scaling solutions such as sharding, directly modify Ethereum; Layer2 scaling solutions run on top of Ethereum. The following is a classification of common Layer2 solutions:

Among those Layer2 scaling solutions, Ethereum’s founder, Vitalik’s favorite one is the Rollup solution, and so far, Rollup is indeed the fastest growing and most popular, which executes transactions by packaging the transaction data and transferring it off Ethereum blockchain, while the assets remain on-chain. The transaction data is then sent back to Ethereum after processing is completed. Once Layer1 is strong enough, most transactions can be performed on Layer2, allowing for greater scalability and faster transactions with lower gas fees. The key to Rollup is that it must ensure it can inherit the security performance of Layer1. The two mainstream Rollup solutions include Zero-Knowledge Rollups (zk-rollups) and Optimistic Rollups, and the Arbitrum we discuss in this paper is one of the Optimistic Rollups. The two types of Rollups should be able to scale the TPS of Ethereum from 15 - 45 to 1000 - 4000. A brief description of the two factions is as follows.

Zero-knowledge rollups are implemented by leveraging a clever cryptography called zero-knowledge proof, hence the name ZK rollups. In this model, each batch of files sent to L1 includes a cryptographic proof called ZK-SNARK. When a transaction batch is submitted, this proof can be quickly verified by the L1 smart contract, and invalid batches can be rejected outright.

As its name "optimistic" suggests, Optimistic Rollups publish data to Layer1 and assume that the data is correct. If the posted data is indeed valid, it can continue to be "optimistic" and does not need to do anything additional. However, if an invalid transaction occurs, the system must be able to recognize it, restore it to its correct state and penalize the party that submitted the transaction. To achieve this, Optimistic Rollups has introduced a dispute resolution system that validates fraudulent proofs, detects fraudulent transactions, and discourages bad actors from submitting other invalid transactions or incorrect fraudulent proofs. A party able to submit batches of transactions to Layer1 must provide margin (usually ETH). Any other network participant who discovers an incorrect transaction can submit a proof of fraud. After the proof of fraud is submitted, the system enters dispute resolution mode. In this mode, the suspicious transaction will be executed on Ethereum again. If the execution proves that the transaction is indeed fraudulent, the party that submitted it will be punished, usually by cutting its margin. To prevent bad actors from disrupting the network with incorrect fraudulent proofs, the party submitting the fraudulent proof must also usually provide a margin that can be cut. The system works as expected and detects fraud even if only 1 honest party monitors the status of the rollup and submits fraud proofs when needed. Also, with the right incentives within the Rollup system, entering the dispute resolution process should be the special case, not the norm.

The shortcoming of Optimistic Rollups is that in order to implement their dispute resolution mechanism, all network participants must have enough time to submit proof of fraud before finalizing the Layer1 transactions. This time is typically quite long to ensure that even in the worst case scenario, fraudulent transactions can still be disputed, which can lead to rather long withdrawal times, with users having to wait up to a week or two to withdraw their funds back to Layer 1. However, there are already several projects (Hop protocol and Connext) that are working to improve this situation by providing fast "liquidity exists" to improve this situation, allowing users to withdraw funds back to Layer1 instantly, but for a small fee.

In contrast, ZK rollups do not have the problem of long withdrawals. The core difficulty is that it is much harder to create an EVM-equivalent ZK rollup due to the complexity of the technology, which makes it more difficult to scale a generic application without rewriting the application logic. However, ZKSync is making significant progress in this area and they may soon be able to launch a ZK rollup solution that is equivalent to EVM. Optimistic Rollups are a bit easier in terms of EVM equivalence. They still need to run their own version of EVM with some modifications, but 99% of the contracts can be transferred directly without any modifications. Moreover, ZK rollups are also much more computationally intensive than Optimistic rollups, which means that the nodes computing the ZK proofs must be high-spec machines and it is difficult for regular users to run them.

Arbitrum is an Optimistic Rollups solution that allows users to use dapps, deploy smart contracts, etc. on Arbitrum, with lower transaction fees and faster transactions, while using Ethereum to ensure the security of transactions. Currently Arbitrum runs two parallel chains based on Ethereum: Arbitrum One and Arbitrum Nova, whose core highlights are as follows.

The core feature of Arbitrum One is the ability to detect fraudulent transactions on Layer1. A validator can send stateRoot on the Arbitrum chain to Layer1's contract by running open source validation software, and while the average user is likely not interested in becoming a validator, the key point is that anyone can become a VALIDATOR and the chain will remain secure as long as there is an honest validator; that is, it only takes one non-malicious fraudster to catch any number of malicious ones. Together, these properties make the system "trustless" - users do not rely on any particular designated party to keep their money safe.

Such a "proof of fraud" mechanism is at the heart of Arbitrum's ability to reduce transaction costs; in addition, there are a number of other mechanisms designed by Arbitrum to help reduce the burden on the L1. First, Arbitrum transactions are submitted to the L1 in batches; typically, a batch (submitted in one L1 transaction) will contain hundreds of L2 transactions, and submitting them in batches amortizes the overhead of interacting with the L1, thus saving a lot of money compared to submitting a single transaction at a time; furthermore, transaction data is published in a compressed form on the L1 (decompressed only in the L2 environment), making the transactions take up less L1 space.

In Arbitrum, the experience is essentially the same as in Ethereum, except that transactions are faster and cheaper, as Arbitrum has had EVM equivalence as a feature at the forefront of its development from the beginning. Users can continue to use Ethereum wallets on Arbitrum, and developers can deploy contracts in Ethereum libraries and tools. In February, Arbitrum also announced the launch of Stylus, a next-generation programming environment that allows developers to use programs written in Rust, C and C++ through WebAssembly, which is expected to attract more Web2 developers. To achieve strong EVM equivalence, Arbitrum One uses and improves upon the fork of Geth by turning it into a Layer2 chain, meaning that most of the code running on Arbitrum is identical to that running on Ethereum.

In September 2022, Arbitrum One underwent a Nitro upgrade with the following major improvements:

• Higher throughput: 7 to 10 times higher than before;

• Advanced call data compression: driving down transaction costs for Arbitrum by reducing the amount of data posted to Ethereum;

• Ethereum L1 gas compatibility: EVM pricing and accounting is consistent with Ethereum;

• Additional L1 interoperability: including tighter synchronization with L1 block numbers;

• Geth tracing: broader debugging support.

With the introduction of Nitro, transactions are now processed in two stages. Nitro first places the transactions into a sequence that will be processed, then publishes the sequence and applies a deterministic state transition function to each transaction in the sequence. Currently, the sequencer is a centralized component operated by Offchain Labs, which will gradually transition to a distributed system as Arbitrum enters its later stages. While the centralization tradeoff in pursuit of lower gas costs is understandable in the short term, the creation of a decentralized system is imperative.

Arbitrum Nova is the AnyTrust chain launched by Arbitrum. Rollup and AnyTrust are similar in many ways, but there is one key difference: on Rollup, all data is published on L1 (allowing anyone to join as a validator without permission), while on Anychain, data are managed offchain by the External Data Availability Committee, whose members include Offchain Labs, Consenys, Google Cloud, P2P, Quicknode and Reddit. In case of a challenge, the AnyTrust chain reverts to Rollup. Arbitrum believes that AnyTrust may be a wise tradeoff for applications that require high transaction throughput and do not require Rollups to provide full decentralization (e.g., gaming, social applications). In other words, Nova reduces transaction costs at the cost of security.

The initial total supply of ARB is 10 billion, with a maximum inflation of 2% per year. The current price is stable at around $1.1, and the tokens are already listed on Binance, Coinbase and other first-tier exchanges.

As Arbitrum transitions to a decentralized autonomous organization (DAO), the token will be used exclusively for governance. Holders can use the token to vote on governance proposals to decide on key issues, including fees, token unlocking and inflation rates. Currently the gas fee is still paid using ETH on Arbitrum. 56% of the total is accounted for by the Arbitrum community and the airdrop on March 23rd has distributed 11.5% of the total supply to eligible users, with the other 44% going to investors and employees of Offchain Labs, the founding team of Arbitrum.

The token unlocking period is shown below, with institutional investors and teams starting to unlock after one year, followed by three years of linear monthly unlocks.

From the perspective of P/S Ratio, Arbitrum is higher than Optimism, which means that the value of Arbitrum may be overvalued in the long run. Although the ARB tokens held by investors will be unlocked after one year, there is no selling pressure in the short term. However, the difference between the circulated market value and FDV is large, and the latter will still face greater selling pressure. In addition, from the perspective of MCap/TVL and FDV/TVL, as the TVL of Arbitrum is far ahead of Optimism, both indicators are lower than OP, and the coin price still has a small upside trend in the short term.

The development team behind Arbitrum is offchain labs, whose core founders include Ed Felten, former White House CTO and professor of computer science at Princeton University; Steven Goldfeder, PhD in computer science at Princeton University, postdoctoral fellow at Cornell University, and former Google intern; and Harry Kalodner, also a computer science PhD at Princeton University. Among them, founder Ed Felten's experience and background are most notable. Back in 2014, before the birth of Ethereum, he decided to research blockchain scaling solutions after seeing the potential congestion in the Bitcoin network. Currently Offchain labs remains small and lean, with a team of 59 people, mainly engineering and business development employees, mostly from Princeton, CUHK, and Harvard.

Offchain labs raised $123.7M in total according to Crunbase's statistics.

• 2019.4: Seed Round, $3.7M, Pantera Capital led the investment, Coinbase Ventures participated in the investment

• 2021.4: Series A, $20M

• 2021.8: Series B, $100M, Lightspeed Venture Partners led the investment, with participation from Redpoint, Polychain, MCE, Alameda Research and others.

According to Messari's summary, the Arbitrum ecosystem is mainly divided into Defi system and other consumer applications.

The Defi ecosystem is based on DEX, lending and perpetual contracts and the venue is already quite crowded.

6.1.1 DEX According to Defi Llama's statistics, just a few months ago, DEX activity on Arbitrum surpassed Polygon and was second only to Ethereum, and the DEX protocols on Arbitrum are mostly similar to Ethereum. The project currently posing the biggest challenge to older DEXs such as Uniswap is Camelot (https://excalibur.exchange/). Each pool deployed on Camelot has customizable fee parameters, and a core feature of the product is its more targeted customizability; for example, USDC-WETH pools can charge 0.04% for transactions that purchase WETH from the pool, but only 0.02% for transactions that sell WETH into the pool. This directional feature allows the pool deployer to fine-tune the incentive structure for a given pair. In addition, these fee levers are dynamic, which allows the fee structure to change based on factors such as market conditions.

6.1.2 Lending Protocols Radiant (https://radiant.capital/) goes live on Arbitrum in July 2022 and is built on LayerZero to create a full-chain lending protocol. Radiant currently accounts for 57% of the TVL of the Arbitrum lending venue, a figure that is largely driven by the large number of RDNT incentives, which offers a one-click "cycle and lock" feature that allows users to re-deposit borrowed assets as collateral and then receive loans from the agreement up to five times. The longer the user pledges, users can earn more RNDT rewards. So beyond the basic double-digit lending APR, users can also receive more benefits from the agreement lending for more benefits, which so far looks like a very successful growth strategy for Radiant. Since the beginning of 2023, the price of RDNT has increased by 1,075%, which makes its incentives even more attractive.

6.1.3 Perpetual contracts Exchange hybrid models that combine AMM and order thinning are gaining popularity in Defi, and such platforms have been a key driver of early growth in the Arbitrum ecosystem. The current perpetual contracts venue holds about $565 million in TVL, the largest share of the Arbitrum Defi ecosystem. Of this, GMX (https://gmx.io/) accounts for over 80% of the TVL.

The key to the success of GMX, which was the first to go live with Arbitrum One in August 2021, is the dual token model: 30% of the protocol's revenue is allocated to stakeholders of GMX (the protocol's governance token) and 70% to stakeholders of GLP (the platform's liquidity token). Such a mechanism helps to build a complete ecosystem around the protocol, thus bringing greater liquidity to GMX.

Most of the tracks on the Arbitrum ecosystem other than Defi are still very early and have not yet reached the same level of volume as Defi applications.

6.2.1 Games Arbitrum's gaming track is represented by the Treasure protocol (https://treasure.lol/). Treasure is an ecosystem for game developers and aims to be the "Nintendo of the Web3". The games released on Treasure are directly interoperable and currently support 10+ games that can be traded using its native DEX MagicSwap and native NFT marketplace Trove, and its native governance token is MAGIC. The success of Treasure so far can be attributed mainly to its first-mover effect and the fact that there are currently no competitors of its kind in the Arbitrum ecosystem.

6.2.2 NFT Arbitrum's NFT ecosystem is still very early, with only 3,900 ETH of NFT transactions in the past month, compared to 1.3 million ETH of NFT transactions on Ethereum in the same period.

According to the CMC data, the current status of the Layer2 venue can be seen as follows:

The advantages and disadvantages of the two have been discussed in Part 1 of this paper. The core advantages of OP-Rollups including Arbitrum and Optimism are their EVM equivalence, which can be used directly on MetaMask, and their first-mover advantage, which has attracted many star projects such as Sushi, and may form an ecological barrier. In contrast, ZK-Rollup’s technology is quite difficult, does not have EVM equivalence , and the supporting ecosystem is weak, but the progress of ZK infrastructure is still worth noting; and from the fundamental technology point of view, if ZK path can be realized, it will kill OP-Rollup. From the investment point of view, we can still pay more attention to the OP-Rollups in the short term.

Here we focus on a multi-dimensional comparison between Optimism and Arbitrum, the representative projects in the OP family. While Arbitrum and Optimism both attempt to achieve the same goal, the designs differ. First, the dispute resolution model is designed in such a way that instead of re-running the entire transaction at Layer1 to verify whether the fraud proof is valid, Arbitrum proposes an interactive multi-round model that allows narrowing the scope of the dispute, possibly by executing only some instructions at L1 to check whether the suspicious transaction is valid, which also allows smart contracts deployed on Arbitrum to be larger than the the maximum contract size allowed; another major difference is the approach to deal with transaction sequencing and MEV. Arbitrum initially runs a sequencer responsible for transaction sequencing, but in the long run they want to decentralize it. Optimism prefers an alternative approach where the sequencing of transactions, and the resulting MEV, can be auctioned off to other parties over a period of time.

Offchain Labs made a presentation on Arbitrum's future development path in SmartCon 2022 speech. Around solving the problem of decentralization, the project will make upgrades to verifiers and sequencer, etc., and Odyssey will also restart this year. From the perspective of the Layer2 Rollup track, it is difficult for ZK-Rollups to catch up with OP-Rollups in the short term, and Arbitrum has crushed its competitors in all aspects of business performance, so its ecosystem projects are worth paying attention to in the short term. For example, ZKSync may soon be able to launch a ZK rollup solution with EVM equivalence. In addition, the new public chain Sui in the Move line. Mochain will keep an eye on the Layer2 venue and continue to output related research reports.