4D analysis of the essence of the cross-chain protocol: Web3’s TCP/IP will change the competitive landscape of public chains

core point

1) Since there are trade-offs in the design of the blockchain, according to the principle of division of labor, the multi-chain ecosystem will inevitably exist for a long time. Specifically, there are seven obvious divisions of labor, namely security chain, performance chain, function chain, bitcoin chain, alliance chain, capital chain and application chain.

2) If each chain is an isolated trusted computer, the cross-chain protocol is the trusted communication protocol of these trusted computers, that is to say, the cross-chain protocol is the TCP/IP of Web3 , which has great potential.

3) The cross-chain track barriers are high, and the concentration will be high in the long run . The reason for the low concentration right now is that the track is still in its early days.

4) The core competitive elements of the asset cross-chain track are brand and product experience ; the core competitive elements of the information cross-chain track are BD capabilities , security and product experience .

5) The comparison of the core competitive elements of the existing head cross-chain projects is as follows:



Rather than the name cross-chain bridge, I prefer to call it a cross-chain communication protocol. From an abstract point of view, each public chain is logically a permissionless, world-shared, isolated trusted computer (for example, Ethereum is the Ethereum Virtual Machine EVM, which is trusted, immutable, and logically unified). A virtual machine), and the cross-chain bridge is the communication protocol between these isolated computers. Therefore, if the public chain is the world computer of Web3, then the cross-chain communication protocol is the TCP/IP protocol of Web3 (the difference is that the cross-chain protocol has one more trust layer than the communication protocol of TCP/IP ), and its function is to These isolated trusted computers are grouped together to form the Internet of Trusted Computers .


After reading this, you may have many doubts, such as why there must be multiple trusted computers shared by the world? Why can’t these computers be interconnected with real TCP/IP, but need to use a cross-chain communication protocol? What is the trust layer? How valuable is TCP/IP for Web3? And what we care most about: With so many cross-chain protocol projects, who has a better chance of winning? I will answer these questions one by one below.

The discourse of this paper is as follows:

The first part deconstructs cross-chain issues and introduces the concept of trust layer to help you quickly understand the nature of 90% of existing cross-chain projects.

The second part starts from the demand side and evaluates the value of cross-chain protocols: first, it demonstrates the inevitability of multi-chain ecology, then expounds the meaning of multi-chain composability, and clarifies the ecological niche of cross-chain communication protocols in this trusted computer Internet , and finally we can roughly estimate the value of the cross-chain communication protocol and the market space of this track.

The third part starts from the underlying logic and analyzes the competition pattern of the existing head cross-chain communication protocol projects: first, analyze the barriers and core competition elements of this track, and infer the future industry concentration of this track. Based on this framework, we analyze the existing head projects and technical paths, and make judgments on their competitiveness.

This article pays more attention to the analysis and deduction based on the underlying logic, not the translation of technical documents for cross-chain projects, and friends without technical background can read it with confidence. Are you ready? Let’s go!


1. Analysis of the operation mode and technical path of the cross-chain protocol

The function of the cross-chain communication protocol is to ensure that the state information on one trusted computer (such as Ethereum) can be transmitted to another trusted computer (such as Solana) without being tampered with. Specifically, for the trusted communication protocol of chain A → chain B, three points need to be guaranteed in the process of transmitting the state information of chain A:

  • 1) Source chain side: The state obtained by the protocol in the A chain is indeed the final state of the A chain.
  • 2) Transmission side: The status information of the protocol has not been tampered with during the information transmission process.
  • 3) Destination chain end: Only the status information transmitted by the protocol can be received by the B chain.

Let’s take a relatively simple asset cross-chain as an example. If Xiaoming wants to transfer 1,000 USDC from Ethereum to Solana through a cross-chain bridge, the cross-chain process is: First, Xiaoming needs to transfer 1,000 USDC to the cross-chain The bridge is on the Ethereum contract, and then the cross-chain bridge transmits the information that “Xiao Ming has paid in Ethereum” to the cross-chain bridge contract on Solana, and then the contract transfers 1,000 coins that were originally stored on the Solana cross-chain bridge protocol. USDC calls Xiao Ming’s address in Solana. Then, the cross-chain bridge protocol needs to guarantee three points:

  • 1) Source chain side: The protocol confirms the status of “Xiao Ming has made a payment in Ethereum” on Ethereum, and this status has been finalized.
  • 2) Transmission side: During the information transmission process of the protocol, the status information of “Xiao Ming has paid in Ethereum” has not been tampered with.
  • 3) Destination chain end: The B chain knows that the status information of “Xiao Ming has made a payment in Ethereum” is indeed sent and received by the protocol.

So, can the function of trusted communication protocol be accomplished by directly using network communication protocol (such as TCP/IP)? The answer is no, because the blockchain is actually a trusted computer composed of many nodes (each node can be considered a normal computer). If you want to directly use TCP/IP for information transmission, each node of the A chain must send the entire block to all the nodes of the B chain, that is, m*n blocks are sent at the end (where m is the A chain The number of nodes in the chain, n is the number of nodes in the B chain), this cost is unaffordable for the blockchain. Therefore, we need a TCP/IP protocol belonging to a trusted computer, that is, a cross-chain protocol.


The essence of the cross-chain protocol is realized through the trust layer. The design of the trust layer determines the trade-off of the security, cost, delay and other attributes of the information transmission of the cross-chain project, which is also the place where innovation occurs in the cross-chain field.

The most direct way to build the trust layer is to let the smart contracts of the two chains trust an intermediate subject (this subject may be composed of multiple nodes in a decentralized manner), this subject confirms the state information obtained from the A chain, and transmits it to the chain. to the B chain. Specifically, as shown in the figure above, if an event T occurs in the A chain and needs to be transmitted to the B chain, the cross-chain protocol will operate as follows: 1) The trust layer observes the latest block of the A chain and finds that the information of the event T needs to be transmitted, Then the event T is downloaded to the trust layer; 2) The trust layer transmits the information of the event T to the B chain; 3) The B link receives the information of the event T transmitted by the trust layer , and knows that the event T happened on the A chain.

Under this scheme, cross-chain users must trust that the trust layer will not do evil. In the absence of code loopholes in the protocol, as long as the trust layer is honest, cross-chain information can be securely transmitted. If the trust layer does evil, the trust layer can intercept information or even publish false information, thereby causing damage to the user’s property.

The way this trust layer is designed and its variants can cover 90% of existing cross-chain solutions. Specifically, this trusted subject can be of the following categories (in descending order of centralization):

  • 1) Centralized institutions: Binance Bridge is typical. The user’s payment operation in the A chain will be observed and confirmed by the centralized server of Binance, and then the user will be given the corresponding withdrawal permission in the B chain. Its advantages are obvious: low cost and low latency; but the disadvantages are also obvious: users need to trust Binance (Binance can completely refuse to send users money in B chain after A chain receives money).
  • 2) External committees: The Ronin bridge is typical. The user’s payment operation in the A chain will be observed and confirmed by the members of the external committee, and then the operation will be signed and sent to the B chain to tell the B chain that “this committee confirms that the user in the A chain has completed the payment”. When the B chain receives enough signatures (for example, 5 signatures in the 9-member committee), the B chain will believe the majority of the committee members and think that the user has indeed made a payment in the A chain, and the information has been passed. The advantages are also: low cost and low latency; but the disadvantages are also: users need to trust the majority of the members of this committee (the committee is usually small), and also hope that their private keys will not be stolen.
  • 3) Pos chain: Axelar is typical. The user’s payment operation in the A chain will be observed and confirmed by the nodes of the Pos chain. After obtaining a majority consensus within the Pos chain, the state information will be transmitted to the B chain through the multi-signature method. The advantage is that the POS chain is more decentralized than the committee; the disadvantage is that the POS chain requires more time and cost to reach a consensus, so the delay and cost are higher.
  • 4) Oracles: Layerzero and CCIP (Chainlink) are typical. The user’s payment operation in the A chain will be observed and confirmed by the decentralized oracle node, and then the status information will be transmitted to the B chain through the multi-signature method. The advantages are: Chainlink is more secure; the disadvantages are: it is still not decentralized enough (Chainlink nodes are now whitelisted), and Chainlink needs to be trusted.
  • 5) No intermediate trust principal (light client scheme): Rainbow Bridge is typical. First, chain B will run a light node of chain A at the smart contract level. Then the user’s payment operation in the A chain will be directly observed and confirmed by all the nodes in the B chain (through the light node), and then a consensus is reached on the B chain. Its advantages are: very high security, as long as the chain itself operates normally, it cannot do evil; its disadvantages are: because it needs to run a light node of another chain on one chain, its operation cost is extremely high.

Note that the above solution descriptions are highly simplified. The author mainly aims to explain the nature of its operation mechanism and the source of its security. For detailed information, please read its technical documents or other technical articles.

In addition to adding intermediate trust subjects, there are some clever ways to build a trust layer, such as Nomad’s Optimistic scheme, Connect’s hash lock scheme, etc. In addition, some project parties try to use the zk method to build the trust layer, but the essence is only to optimize the verification efficiency of multi-signature, not a new solution.

Note that although this section has been taking asset cross-chain as an example, the above solutions (except hash locks) can help any information on the chain to cross-chain: as long as the protocol can guarantee that a certain state occurs in the A chain, then the B chain This state can be read and continued to execute tasks, enabling multi-chain interoperability.


Multi-chain ecology will last forever: the seven divisions of blockchain

In 1776, Adam Smith’s “The Wealth of Nations” first put forward the idea of ​​division of labor: each employee specialises himself, does what he has an advantage in the same task, and then cooperates to share the results of his labor. Combined, this approach can be far more efficient than doing the entire task by one person.

Similarly, the design of each chain must have its own strengths and weaknesses. Therefore, the specific requirements of the application should be handed over to the chain with more advantages in this respect. In contrast, a single chain completes all tasks. Inefficient behavior. The following are 7 types of blockchain jobs that I think are more common:

  • Safety chain:

    A typical example is Ethereum. The security chain mainly refers to the chain that has been tested by history, is decentralized, has high security and high reliability. The blockchain acts as a trust machine, and Ethereum has the lowest trust cost at the cost of lower throughput and higher latency. In the long run, Ethereum is suitable as a “treasury” to carry high-value and low-frequency operations.

  • Performance chain:

    Typical examples are Solana, Aptos and Sui. The performance chain mainly refers to the chain with higher throughput and lower latency, and the cost is generally more centralized and less reliable. Compared with Ethereum, the trust cost of the performance chain is higher, and the difference is that governments may hold btc or eth as currency reserves in the future, but obviously it is unlikely to hold Sol, which has a high trust cost. Note here that layer2 (mainly refers to rollup) cannot be regarded as a performance chain, because on the one hand, the performance of layer2 is still very limited. On the other hand, even if the throughput of layer2 can be greatly improved, its final delay cannot be improved, so it is difficult to carry Some scenarios that require high frequency operation.

  • Function chain:

    A typical example is Arweave. Functional chains mainly refer to chains designed for special functions, such as storage, at the expense of generality. In terms of storage function, Arweave does have a huge advantage over other general smart contract chains.

  • Bitcoin chain:

    There is only one chain, and that is the chain that carries Bitcoin. Bitcoin has its special historical position, has the strongest consensus among all cryptocurrencies, and has the most friendly regulatory attitude towards it. Obviously, this division of labor is irreplaceable.

  • Alliance chain:

    Typical examples are the future Chinese Web3 chain and Central Bank Digital Currency (CBDC) chain. The alliance chain is a chain supported by the government and may be deployed vigorously. The price is that it is controlled by the government to a certain extent and the cost of trust is high. First of all, we don’t have to rush to exclude the consortium chain. As a trust machine, the consortium chain is also a kind of blockchain, but its trust cost is high. Compared with an opaque single centralized server, the consortium chain is obviously more trustworthy. Secondly, the alliance chain has a very special advantage: it is loved by the government and can get a lot of government resources. For example, if Web3 happens in the world, China will most likely not be left behind, so Web3 with Chinese characteristics will appear, that is, digital RMB + domestic alliance chain (just like the Internet with Chinese characteristics with walls). CBDC has a similar logic. Therefore, some governments do not intend to decentralize, but want to digitize information, increase composability, and enhance trust scenarios, which will be satisfied by governments in the form of alliance chains.

  • Capital chain:

    A typical example is the BSC chain. The capital chain mainly refers to the chain made by big capital itself. Similar to the alliance chain, the advantage of the capital chain is that there is capital behind it, but the disadvantage is that it is more centralized and the cost of trust is high.

  • Application chain:

    A typical example is Axie’s Ronin chain. The application chain mainly refers to the chain made by the head application itself. The advantage is that the value of the chain can be captured, and the performance will not be affected by other Dapps. The cost is lower security, higher trust cost, and lower composability.


    The author believes that the first five divisions of labor are very likely to exist for a long time, because they all have strong advantages and are irreplaceable. The advantages and irreplaceability of the last two divisions of labor are relatively small, and may disappear in the long run.


Complementary combination of multi-chain division of labor

After a precise division of labor, we now have seven trusted computers with their own strengths. Assuming that the cross-chain protocol technology is mature enough to allow the contracts of these chains to fully enjoy interoperability, how can these seven machines cooperate? Here are some complementary examples:

  • Safety chain + performance chain:

    This combination is very complementary. Users can put a large number of assets and high-value operations on the security chain, and then cross over to the performance chain whenever high-frequency operations are required, and then cross back to the security chain after the operation is completed. For example, when a user wants to play poker on the chain, the user can transfer a certain amount of money to the poker application on the performance chain (similar to buying chips), and enjoy the high throughput, low latency and low gas brought by the performance chain The fee is flopped, played, and bet on the chain. After the game is over, the money won (or lost) is transferred back to the security chain for storage. Scenarios such as cross-chain gaming, cross-chain DAO voting, cross-chain NFT, and cross-chain SBT are similar.

  • Function chain + general chain:

    Take the storage chain as an example for now. After Ethereum implements proto-danksharding, the historical data to be stored by each Ethereum full node will increase rapidly. Ethereum responded with EIP4444: full nodes will delete historical data older than a year. But obviously these data are very precious, and we can only count on at least one relevant agency to keep them. But when the trusted communication protocol technology matures, it is obviously a safer and more natural method to let Ethereum automatically and trustlessly store these precious data on the storage chain through smart contracts.

  • Bitcoin Chain + Universal Chain:

    Bitcoin is still the largest asset class in the crypto world, and introducing this part of the asset into the defi in the universal chain can unlock a lot of value in Bitcoin.

  • Alliance chain + public chain:

    The public chain will carry most of the Web3 applications, but some regions or scenarios restricted by the government may have to use the alliance chain. Connecting the public chain and the alliance chain can make them communicate with each other and improve efficiency. For example, the digital renminbi of Web3 with Chinese characteristics obviously needs to be exchanged for foreign exchange, and the cross-chain agreement between the Chinese alliance chain and the public chain can provide a channel for the digital renminbi to be exchanged with overseas currencies.

  • All chain aggregates:

    Liquidity has been severely fragmented in the capital war of the public chain, which itself is a very inefficient behavior of capital. Cross-chain protocols can reunite these fragmented liquidity and improve the capital efficiency of Dex, lending and other defi protocols.

Cross-chain communication protocol: the cornerstone of multi-chain composability

The better the cross-chain communication protocol, the lower the cost of cross-chain interoperability, and the more likely it is that contracts on different chains are deployed on the same chain. The possible costs of cross-chain communication include: cost, delay, security, etc. The existing cross-chain bridges still have many shortcomings. The huge cross-chain cost prevents some obvious multi-chain applications from appearing, such as the defi ecology that introduces a large amount of BTC into the smart contract chain (now WBTC only helps 1% of the BTC cross-chain). to Ethereum, because it is too insecure), or aggregate the entire chain liquidity.

Further, the development of cross-chain communication protocols is likely to change the competitive landscape of public chains. Two important barriers to public chains are: 1) network effects formed by composability between public chain ecological projects, and 2) liquidity. These are also the biggest barriers to Ethereum right now. The development of cross-chain communication protocols is likely to weaken this part of the barrier, because: 1) projects on other chains can enjoy the composability of high-quality projects on Ethereum through cross-chain protocols, and 2) other chains can obtain frictionless access to /Capture liquidity on Ethereum. This will make the public chain need to rely more on the characteristics of the product itself to win in the future, which may force the future public chain to become more and more specialized, do a good job in the division of labor, and cooperate with other chains to carry the various needs of a Dapp.

In general, the cross-chain communication protocol is the TCP/IP protocol of Web3, which can help trusted computers (blockchains) with their own advantages to be combined together to form an Internet of trusted computers (blockchains), making the blockchain Chains can divide labor and cooperate and exert great value , just like the Internet allows computers all over the world to gather together and cooperate with each other. The current cross-chain protocol is not yet mature, and the huge cross-chain cost has prevented many obvious multi-chain applications that can improve efficiency. As the technology matures, more and more multi-chain applications will appear. As an important part of multi-chain applications, cross-chain protocols will capture more and more value, and the market space is huge.


Competitive Landscape

Industry barriers are high, and concentration is high in the long run

In the long run, because the barriers of cross-chain protocols are high and the demand is relatively homogeneous, it can be expected that the concentration of cross-chain bridges will become higher in the long run, instead of being as decentralized as they are now. (Barriers, also known as moats, are defined as a company’s sustainable competitive advantage against competitors’ attacks. The higher the industry barriers, the easier it is to monopolize, otherwise it tends to be more perfect competition.)

Specifically, the cross-chain communication protocol has the following strong barriers:

1) Scale effect : First of all, as a software, the marginal cost of each additional user is 0, and the customization cost of each additional chain/project is also lower, so the marginal cost is decreasing, and there is an economy of scale effect; secondly, the brand is a cross-chain The agreement is a very important competitive element (explained in detail in the next section), and the brand is also a competitive advantage for a fixed investment project to be used for life, so the larger the project, the lower the marketing cost per unit user, so it also has the effect of scale .

2) Conversion cost : The cost of replacing the cross-chain protocol in the middle of a multi-chain project is relatively large, so the conversion cost is relatively large.

At the same time, the cross-chain communication protocol track is different from catering, fashion, NFT and other consumer tracks, and its needs are highly homogeneous:

1) Whether it is 2B or 2C, the cross-chain bridge makes rational money, not emotional money.

2) There is only one cross-chain information transmission required by users (as well as cross-chain asset information transmission), and the requirements are highly homogenized , but for different users, there are certain differences in the preferences of attributes such as security, cost, and delay.

A similar example is the oracle track. Its barriers are also scale effects and switching costs, and the demand is also homogenized. The result is a very high degree of concentration – Chainlink monopolizes the entire market. The author believes that the reason for the low concentration of cross-chain projects is that this track is still in its early stage (by contrast, the oracle track is relatively mature), the protocol side is still iterating on new solutions, and the multi-chain application side Applications have not yet emerged. However, if analyzed from the first principles of the cross-chain track, the cross-chain track has a high concentration in the long run.

Core Competitive Elements

To analyze the competitive landscape of a track, we must first analyze the core competitive elements of the track (for example, in a cross-chain track, is security more important? Or is reusability more important? Or is TVL more important? Which one is Core competitive elements?), otherwise, it will fall into a vicious circle where project A and project B have their own advantages and disadvantages and cannot be compared.

For the demand side, cross-chain projects may generally provide two services: asset cross-chain and information cross-chain. Asset cross-chain mainly refers to token cross-chain. For example, a user wants to transfer his 1000 USDC in the A chain to his account in the B chain. It can be seen that the asset cross-chain service is to C (consumer-oriented). Information cross-chain mainly refers to providing arbitrary information cross-chain services for multi-chain projects, such as the multi-chain Dex and multi-chain games mentioned above. It can be seen that the asset cross-chain service is to B (project-oriented). The difference in the demand side makes the core competitive elements of asset cross-chain and information cross-chain very different. Therefore, the author will analyze these two tracks separately below, even though most star cross-chain projects will provide both services at the same time. There are many competitive elements in the cross-chain track. Below, I will list the core competitive elements in descending order of importance, and temporarily ignore other less important competitive elements in order to focus on important things.

Asset cross-chain (to C)

1) Brand : Undoubtedly, in the dark forest, users care about security the most when it comes to operations involving money. Specifically, users are not going to cross-chain bridge with an asset they have never heard of and discovered from a source they don’t trust, no matter how cheap and silky it claims to be (you’ll be tweeting with someone you don’t know) Give you Amway’s cross-chain bridge?). However, it would be a big mistake to take security as the core competitive factor of asset cross-chain, which is also a common mistake made by many project parties and investors with technical background.

First, let’s talk about economics. Economists divide goods into three categories: search goods, experience goods, and trust goods:

  • Search items are items whose quality can be judged by observation before users buy, typically furniture such as tables, chairs, benches, and the like.
  • Experience products are products that users can use for a period of time to judge the quality after they have purchased them, typically cars.
  • Trusted products are products whose quality cannot be judged after users have purchased and used them, typically health care products.

The more difficult it is for users to judge the quality of products, the greater the role of brand marketing. The obvious situation in real life is that the marketing expenses of health care products are much greater than that of cars, and the marketing expenses of automobiles are much higher than that of furniture such as tables, chairs, benches, etc.

For security, cross-chain bridges are trusted items , because most users cannot understand the technical solutions of cross-chain protocols and verify the security of cross-chain bridges. At this time, users can only trust their brand: choose to use a bridge that is officially endorsed by the public chain, backed by big capital, promoted by big KOLs, and frequently heard of. Therefore, although users actually care about the security of the cross-chain bridge, due to information asymmetry, in fact, the core element of the project party to attract users is the brand. It’s as if consumers of health supplements actually hope that there are magic formulas in the supplements that can make their bodies better, but in fact consumers are only attracted to the health supplements that understand the marketing and build the brand best.

But this does not mean that the security of to C’s cross-chain bridge is completely unimportant, but that the security of the project itself is for the brand: the higher the security of the bridge itself, the public chain adoption, capital assistance, and Kol’s promotion The greater the probability, the better the project party can occupy the user’s mind when doing PR with an actually high-security solution.

2) Product experience : Product experience mainly includes cross-chain costs and delays. Since asset cross-chain itself is a low-frequency operation, slightly higher costs and delays can be tolerated. Of course, the lower the cost and the smoother the use, the better.

Information cross-chain (to B)

1) BD capability : Information cross-chain is a to B business, and it is very important for the project to get enough top benchmark customers: when there are enough benchmark customers, long-tail projects will automatically follow, just like Chainlink . In the case that each product has its own trade-off, the BD capability has become the key to whether the cross-chain project can win enough head projects in the early stage to enhance the brand power and income.

2) Security : If there are loopholes in the third-party cross-chain protocol, the multi-chain project itself will suffer a devastating blow (for example, the assets of the full-chain Dex may be stolen due to the loopholes in the cross-chain protocol), so security is important for the project party. Say it’s important. At the same time, unlike ordinary users, the project party has a certain ability to verify the security of the cross-chain protocol, and will not make a decision just by looking at the brand.

3) Product experience : including cost and delay. The Dapps carried by the information cross-chain protocol generally have relatively low-frequency cross-chain operations (for example, the full-chain dex only needs to cross the chain once to provide liquidity, and the cross-chain DAO vote only needs to cross the chain twice to release and withdraw the voting results), so the slightly higher The cost and delay are bearable, of course, the lower the cost and the smoother the user experience, the better.

Competitive landscape of cross-chain projects

Due to the limited space, the author only selected the following five projects/technical paths to analyze the competitive landscape. Please forgive me if there are any omissions. The following is a summary of the core competitive elements of the five project/technology paths:


(T1 refers to the first echelon, so T1>T2>T3>T4)

external committee

The external committee plan is to arrange several external members to act as the trust layer (friends who forget the trust layer model can go back and read Chapter 1). The external committee scheme has become the cross-chain scheme with the highest market share due to its simple development, high versatility, low cost, and low latency. Until now, the official bridges of many chains still use external committee + multi-signature schemes, such as Solana’s Warmhole. However, the external committee scheme is less secure due to its centralization. The author believes that the external committee program will decline in the medium and long term for three reasons:

1) Without the competitiveness of the whole chain, it is difficult to become the head: when the external committee bridge is used as the official bridge, its brand is better, and the users of the public chain are also willing to use it. However, when it goes out of the current public chain as a third-party bridge, its brand power will plummet, because the external committee is indeed a backward and unsafe solution, it is difficult to obtain a good endorsement, and it is difficult to use this solution to brainwash users. Based on our discussion on the medium and long-term industry concentration, after the track matures, there is a high probability that all public chains will be monopolized by a few leading cross-chain protocols, instead of each public chain doing its own thing. bridge. Therefore, there is a high probability that external committee bridges that cannot “go abroad” and have no full-chain competitiveness will decline in the future.

2) As the industry matures, external committees are no longer the only option: In the past, an important reason why people chose to develop external committee cross-chain bridges was that the solution was mature and the development was simple. By contrast, developing an “oracle bridge” or “light client bridge” is much more difficult. However, when the head third-party bridge appears, the new public chain can connect to a secure third-party bridge, instead of having to devote little energy to developing a rough external committee bridge.

3) For information cross-chain, security is very important, and it is extremely difficult for the external committee scheme to be adopted by multi-chain applications. In fact, there is no external committee project that has made a relevant layout in the information cross-chain track.

Summary: For the full-chain and cross-chain communication protocol, the external committee solution has poor brand power, but its product experience such as cost and delay is better. In general, the external committee program will likely decline after other technologies mature.


Simply put, Layerzero uses Chainlink as a trust layer, and the security is basically the same as that of Chainlink. For specific technical details, please refer to the layerzero white paper or other technical articles. Layerzero is currently valued at US$1 billion, and the mainnet has been launched. Stargate, a multi-chain asset cross-chain bridge made by its team, currently has a TVL of US$500 million.

Elements of asset cross-chain competition

  • 1) Brand (T2): There are two main reasons why the Layerzero brand is better: 1) Well-known institutions such as A16z, Sequoia, FTX, Binance, Multicoin are all Layerzero’s owners, and the endorsement and publicity of star institutions can bring Layerzero Obvious brand advantage; 2) Layerzero’s security is mainly guaranteed by Chainlink, so Layerzero can use its security and Chainlink’s reputation to do better publicity to build a brand.
  • 2) Product experience (T2): Using Chainlink as the trust layer is a compromise choice, and the delay and cost are lower while ensuring security, so the product experience is better.


Analysis on the elements of information cross-chain competition

  • 1) BD capability (T2): The information cross-chain track is still in the very early stage, and all projects have basically not been integrated with head projects. The only applications on Layerzero that are out of the circle temporarily are the cross-chain bridge Stargate and multi-chain NFT Gh0stly Gh0sts. However, as far as the current situation is concerned, the large amount of capital behind Layerzero can provide greater assistance to Layerzero’s future BD.
  • 2) Security (T2): The security of Layerzero is basically the same as that of Chainlink, and its security is higher than most cross-chain protocols.
  • 3) Product experience (T2): The product experience is the same as above.


Summary: Layerzero chooses Chainlink oracles as its trust layer, which strikes a good balance between security and product experience. Its brand and BD capabilities are also at a high level because of its strong capital lineup, which deserves attention.

CCIP (Chainlink)

CCIP (Cross-Chain Interoperability Protocol) is the next core function being developed by the oracle giant Chainlink, which aims to connect all public chains and achieve cross-chain information across the entire chain. CCIP also uses Chainlink as a trust layer, and the security is basically the same as that of Chainlink.

However, the development of CCIP is slow: Sergey Nazarov, the co-founder of Chainlink, announced the official launch of the CCIP project in September 2021, and no products have been launched until now. The author estimates that the main reason for the slow development progress is that Chainlink attaches great importance to the security of CCIP. After all, if there is a loophole in CCIP, Chainlink’s Price Feed and other businesses will also be questioned. According to the estimation of the CCIP project volume, the author believes that the CCIP has a high probability of going online next year.

Chainlink currently has a FDV of 7.7 billion US dollars. It is an oracle giant that passes through bulls and bears. At present, its main function is to provide the oracle Price Feed service for the Defi protocol. In fact, it is very suitable and natural for Chainlink to provide cross-chain communication services: the role of the oracle machine is to reliably transmit the information of Web2 to the isolated blockchain of Web3, which is similar to the information on an isolated blockchain. Trusted transmission to another isolated blockchain makes no difference (Chainlink’s name is the connection of chains).

Elements of asset cross-chain competition

1) Brand (T1): Chainlink is an oracle giant that has crossed the bull and bear market. It has monopolized the oracle market. Almost all serious projects that require oracles will use Chainlink. Currently, $20 billion in the entire crypto world is protected by Chainlink, and 1,000+ projects have partnered with Chainlink. These have accumulated a very good brand for Chainlink, and Chainlink has become a representative of security in the hearts of users, even more secure than Solana, BSC and other public chains that are often questioned as centralized. Even the current Chainlink is actually somewhat centralized.

2) Product experience (T2): CCIP also uses Chainlink oracles as a trust layer, which ensures low latency and cost while ensuring security, and provides better product experience. Compared with Layerzero, the product experience of CCIP may be slightly better, because Chainlink oracles may be partially optimized for CCIP, but the overall difference should be small.

Analysis on the elements of information cross-chain competition

1) BD capability (T1): Chainlink, as an oracle giant that crosses bulls and bears, has a very strong ecosystem, and has now reached cooperation with 1,000+ projects. Compared with start-up projects, Chainlink has much stronger ecological cooperation resources, which makes it very likely to obtain the cooperation of star projects, quickly establish star benchmark customers after the launch of CCIP, and cause a large number of new multi-chain projects to follow suit. Monopolize the cross-chain communication track in the same way as the oracle track.

2) Security (T2): Chainlink’s security will be slightly better than Layerzero, but the overall difference is not significant. There are two main reasons for the slight improvement: First, Chainlink has an independent anti-fraud network to ensure the correct operation of the oracle and CCIP, and second, Chainlink may make some special optimizations for CCIP.

3) Product experience (T2): The product experience is the same as above.

Summary: CCIP chose its own Chainlink oracle as the trust layer, which has made a good balance between security and product experience. More importantly, Chainlink, which launched CCIP, has accumulated brand and ecological resources that other startups cannot catch up in the first two cycles. The only problem is that CCIP has not been launched yet, and we need to continue to observe the future development of CCIP.


Axelar utilizes its own POS chain as a trust layer with security equivalent to its POS chain. Axelar is currently valued at $1 billion, and Axelar’s ​​TVL on Ethereum is only $80 million.

Elements of asset cross-chain competition

1) Brand (T3): The Axelar brand is average, as can be seen from its TVL. The author believes that there are two reasons: 1) its POS chain solution is indeed not a solution with good security, 2) its top major investors are only Polychain and Dragonfly, and the publicity resources and endorsements are inferior to Layerzero, and they cannot be compared with CCIP. Compare.

2) Product experience (T2): Now the Axelar chain has fewer nodes, so the POS chain can reach a consensus faster and at a lower cost.

Analysis on the elements of information cross-chain competition

1) BD capability (T3): Only Polychain and Dragonfly are the top big investors, and BD resources are relatively limited.

2) Security (T3): Now the number of nodes in the Axelar chain is relatively small, which is relatively centralized and actually less secure.

3) Product experience (T2): The product experience is the same as above.

Summary: Although Axelar has the same valuation as Layerzero, its brand, BD capabilities and security are not as good as Layerzero, and its public chain narrative is actually of little significance. It is necessary to continue to observe whether its competitive elements can develop in the future.

light client

The light client (client and node are synonymous) scheme is the most secure cross-chain scheme. It does not require a trust layer, so it can be trustless at the cost of its extremely high operating cost. To put it simply, if there is information from the A chain to be transmitted to the B chain, first of all, the smart contract of the B chain needs to run a light node of the A chain (actually, each node of the B chain must run a light node of the A chain) , which means that the B chain will continuously synchronize the block header of the A chain. When there is information that needs to be cross-chain, the light client solution does not require a trust layer. Only a relayer transmits the information of the A chain to the B chain, and the B chain can use the A chain light node running on the B chain to communicate with the A chain. information to verify. Note that we do not need to trust the relayer, because if it transmits false information maliciously, when the B chain verifies the information, it will find that the false information does not match the block header of the A chain, thus rejecting the information.

The biggest problem with light client solutions is that they are too expensive to operate. Assuming that to cross-chain from Solana to Ethereum, Ethereum must run a Solana light node in its contract layer (actually every full node of Ethereum must run a Solana light node), which will cost a lot of money The cost of gas fee is mentioned in the Layerzero white paper that running a light node on Ethereum costs millions of dollars per day. In this regard, the solution of Rainbow Bridge (Near’s official bridge) is to submit block headers every 16 hours (instead of 15 seconds), which leads to users’ cross-chain delay of up to 16 hours, which has a great impact. product experience.

Summary: The light client solution is the best security solution at present. However, due to its high operating costs, no mature multi-chain light client projects have emerged at present.

Finally, if you disagree with my point of view or have new and interesting cross-chain solutions, welcome to discuss~

Author: Billy, senior analyst of A&T, CS of Peking University, the youngest first author of SIGMOD Asia, mainly focusing on the direction of infra.

Posted by:CoinYuppie,Reprinted with attribution to:https://coinyuppie.com/4d-analysis-of-the-essence-of-the-cross-chain-protocol-web3s-tcp-ip-will-change-the-competitive-landscape-of-public-chains/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (0)
Donate Buy me a coffee Buy me a coffee
Previous 2022-09-21 23:08
Next 2022-09-22 11:48

Related articles