"AMM" VS "CLOB": Which of the two trading models is better?

Author: 0xNatalie Source: ChainFeeds

Recently, Flashbots strategy director Hasu pointed out: On the Solana chain, most of the trading volume is actually completed through automated market makers (AMMs) rather than through central limit order books (CLOBs, or simply order book models). This conclusion is surprising because many people once believed that one of the important reasons for Solana's success in the market was its ability to support CLOBs. As Feng Liu said: "One of the core selling points of Solana at the beginning was that it could finally do order book dex on it, and that 'order book trading is the future of dex'."

It is worth noting that the debate between AMM and CLOB did not start recently, and this contest has been going on for a long time. Since the DeFi Summer, AMM has quickly become the core of decentralized trading with its algorithm-driven asset pricing method, while CLOB is considered a more mature market mechanism due to its dominance in traditional finance and centralized exchanges. This competition has promoted their continuous innovation on various blockchain platforms. Especially on Solana, which focuses on speed and low cost, Phoenix successfully made CLOB the focus for a time.

Is the reason why AMM dominates the market only because of long-tail assets?

Hasu's discovery quickly sparked widespread discussion within the community. In this regard, Kyle Samani, partner of Multicoin Capital, explained that in the market of long-tail assets, there is a lack of real market makers (MM) to provide liquidity, and the emergence of AMM has made up for this deficiency, thus forming the current situation dominated by AMM. Solana's success does not only rely on CLOB, but because it can provide a consistently fast and low-cost trading experience and can provide support for various types of assets. In addition, Solana's no bridging mechanism is also an important factor in its success, because users generally have a negative attitude towards cross-chain bridging.

Taproot Wizards founder Udi Wertheimer also believes that AMM has unique advantages in supporting long-tail assets, which can help small communities quickly start liquidity for long-tail assets. There are a large number of memecoins on Solana, and AMM is a very suitable choice for these assets.

Krane further divides the market into three types: memecoin, major assets (such as SOL/USDC), and stablecoins. He pointed out that AMMs stand out in the memecoin market because these assets require good passive liquidity, while CLOBs perform poorly in this regard. For major assets, although CLOBs have a certain position in some cases, AMMs are still competitive. In the stablecoin market, the application of CLOBs has not yet been widely popularized.

However, Doug Colkitt, founder of Ambient, put forward a different view and refuted it with data. He pointed out that many people mistakenly believe that the AMM trading volume on Solana comes mainly from some inactive long-tail assets. However, the data he provided shows that even in major trading pairs (such as SOL/USDC), AMM's trading volume far exceeds CLOB. For example, Orca traded $250M in volume in a 24-hour period, while Phoenix did only $14M. Even using the most favorable assumptions for CLOB (using Phoenix’s 7-day average daily volume instead of the day’s lower volume, and including as much of CLOB’s volume as possible), AMM’s volume on major trading pairs is 50% higher than CLOB’s, and without these assumptions, the gap widens to 10x.

Community View: The development of CLOB is limited by the performance of the blockchain

AMM in Solana The reason why CLOB dominates the blockchain is not only because of long-tail assets, but also because of the limitations of blockchain performance. Many community members believe that the development of CLOB is limited by the performance bottleneck of the blockchain. Sam feels that the inherent challenges faced by blockchain (high latency, high gas fees, poor privacy protection, etc.) make CLOB unsuitable for effective operation in the current blockchain environment. In contrast, AMM is more adaptable to the characteristics of blockchain, especially in price discovery and liquidity provision. Enzo holds a similar view. He believes that CLOB faces the limitations of high latency, expensive gas fees and low throughput on Layer 1, but these limitations can be overcome in Layer 2 solutions, making CLOB more competitive in these environments. On the current Layer 1 chain, AMM is still a more practical choice. In fact, a similar view was also mentioned in the article "Death, Taxes, and EVM Parallelization" published by Reforge Research in April. The article points out that when implementing CLOB on blockchain platforms such as Ethereum, due to the limitations of platform processing power and speed, it often leads to high latency and high transaction costs. However, with the introduction of parallel EVM, the processing power and efficiency of the network have been greatly improved, the feasibility of CLOB has also increased, and DeFi activities are expected to increase significantly.