Rachel | SynFutures

1/ I am excited to announce Lantern Capital @lanterncap_ , in partnership with @nethermind, with @tkstanczak joining as General Partner on behalf of Nethermind. Lantern Capital is an early stage venture capital firm, investing in founders accelerating machine-to-ledger communication in the near-AGI era.

对于最近@HyperliquidX ETH和Jelly两个case的问题，如果用一个共同的词来总结, 叫做 liquidity mismatch (1) ETH的case是建仓的时候liquidity太好，不应该这么好，而爆仓的时候liquidity不行，所以@chameleon_jeff 说如果更多更好market makers可以解决这个问题是对的，但只是说难点是随时在爆仓时liquidity都能match建仓的时候 (2) Jelly的case是Perp 比现货 liquidity更好，所以容易操纵现货来攻击perp，典型的oracle attack，所以最后解决方式，HL强制设一个无关的价格来settle，给人感觉就是HL 自己做了oracle attack 怎么解决这些问题呢？ 知道了核心是liquidity不应该太好时候，硬不计后果给了太好的liquidity，方式肯定是限制这个liquidity，但他带来的问题就是用户体验会相对较差。 这中间就是一个体验和安全之间平衡的哲学 （1） ETH这个case，最直接的方法，就是对开出来的仓位还未实现的利润的提取做出限制，HL后来的改进是部分限制去除 (2) Jelly这个case，可以防止的方法包括 a. 对流动性不足的pair的margin做隔离 b. 对insurance fund根据不同的pair做隔离 c. 在OI太大时候自动降杠杆（auto-deleveraging) /做限制 @SynFuturesDefi 是市场唯一的真正去中心化order- book的perp DEX，撮合和结算都在链上，以上的安全风控措施我们都实现了，所以对用户有更好的保护，但这个建立在三个前提 （1） 我们的观点是安全是体验的一部分，或者是最重要的部分 （2） 流动性越差，安全限制会更多。我们是真正的去中心化orderbook，在今天在公链区块链性能下（目前2秒一个区块），对比起HL流动性更差是客观事实，但我看到@base 缩短区块时间、还有市场上其他类似@monad_xyz @pharos_network的性能提升，对用真正去中心化方式支持orderbook抱有前所未有的信心，提升体验指日可待 （3）我们在市场上更久，所以看到过更多的攻击，不断改善我们的模型 虽然是同行以及被抢占市场份额，我对HL也是有极高的尊重，任何项目都会经历挫折后反思改进的过程，这个过程因为社区的参与而发生得更快，这也是开放透明的伟大之处。

对于最近@HyperliquidX ETH和Jelly两个case的问题，如果用一个共同的词来总结, 叫做 liquidity mismatch (1) ETH的case是建仓的时候liquidity太好，不应该这么好，而爆仓的时候liquidity不行，所以@chameleon_jeff 说如果更多更好market makers可以解决这个问题是对的，但只是说难点是随时在爆仓时liquidity都能match建仓的时候 (2) Jelly的case是Perp 比现货 liquidity更好，所以容易操纵现货来攻击perp，典型的oracle attack，所以最后解决方式，HL强制设一个无关的价格来settle，给人感觉就是HL 自己做了oracle attack 怎么解决这些问题呢？ 知道了核心是liquidity不应该太好时候，硬不计后果给了太好的liquidity，方式肯定是限制这个liquidity，但他带来的问题就是用户体验会相对较差。 这中间就是一个体验和安全之间平衡的哲学 （1） ETH这个case，最直接的方法，就是对开出来的仓位还未实现的利润的提取做出限制，HL后来的改进是部分限制去除 (2) Jelly这个case，可以防止的方法包括 a. 对流动性不足的pair的margin做隔离 b. 对insurance fund根据不同的pair做隔离 c. 在OI太大时候自动降杠杆（auto-deleveraging) /做限制 @SynFuturesDefi 是市场唯一的真正去中心化order- book的perp DEX，撮合和结算都在链上，以上的安全风控措施我们都实现了，所以对用户有更好的保护，但这个建立在三个前提 （1） 我们的观点是安全是体验的一部分，或者是最重要的部分 （2） 流动性越差，安全限制会更多。我们是真正的去中心化orderbook，在今天在公链区块链性能下（目前2秒一个区块），对比起HL流动性更差是客观事实，但我看到@base 缩短区块时间、还有市场上其他类似@monad_xyz @pharos_network的性能提升，对用真正去中心化方式支持orderbook抱有前所未有的信心，提升体验指日可待 （3）我们在市场上更久，所以看到过更多的攻击，不断改善我们的模型 虽然是同行以及被抢占市场份额，我对HL也是有极高的尊重，任何项目都会经历挫折后反思改进的过程，这个过程因为社区的参与而发生得更快，这也是开放透明的伟大之处。

对于最近@HyperliquidX ETH和Jelly两个case的问题，如果用一个共同的词来总结, 叫做 liquidity mismatch (1) ETH的case是建仓的时候liquidity太好，不应该这么好，而爆仓的时候liquidity不行，所以@chameleon_jeff 说如果更多更好market makers可以解决这个问题是对的，但只是说难点是随时在爆仓时liquidity都能match建仓的时候 (2) Jelly的case是Perp 比现货 liquidity更好，所以容易操纵现货来攻击perp，典型的oracle attack，所以最后解决方式，HL强制设一个无关的价格来settle，给人感觉就是HL 自己做了oracle attack 怎么解决这些问题呢？ 知道了核心是liquidity不应该太好时候，硬不计后果给了太好的liquidity，方式肯定是限制这个liquidity，但他带来的问题就是用户体验会相对较差。 这中间就是一个体验和安全之间平衡的哲学 （1） ETH这个case，最直接的方法，就是对开出来的仓位还未实现的利润的提取做出限制，HL后来的改进是部分限制去除 (2) Jelly这个case，可以防止的方法包括 a. 对流动性不足的pair的margin做隔离 b. 对insurance fund根据不同的pair做隔离 c. 在OI太大时候自动降杠杆（auto-deleveraging) /做限制 @SynFuturesDefi 是市场唯一的真正去中心化order- book的perp DEX，撮合和结算都在链上，以上的安全风控措施我们都实现了，所以对用户有更好的保护，但这个建立在三个前提 （1） 我们的观点是安全是体验的一部分，或者是最重要的部分 （2） 流动性越差，安全限制会更多。我们是真正的去中心化orderbook，在今天在公链区块链性能下（目前2秒一个区块），对比起HL流动性更差是客观事实，但我看到@base 缩短区块时间、还有市场上其他类似@monad_xyz @pharos_network的性能提升，对用真正去中心化方式支持orderbook抱有前所未有的信心，提升体验指日可待 （3）我们在市场上更久，所以看到过更多的攻击，不断改善我们的模型 虽然是同行以及被抢占市场份额，我对HL也是有极高的尊重，任何项目都会经历挫折后反思改进的过程，这个过程因为社区的参与而发生得更快，这也是开放透明的伟大之处。

Re the recent issues with @HyperliquidX ETH and Jelly cases, if we were to summarize the two with a single term, it would be "liquidity mismatch." (1) For the ETH case, the issue is that liquidity was too good when opening positions—it shouldn’t have been that good—while during liquidation, the liquidity was insufficient. So, @chameleon_jeff was right in saying that as market makers continue scaling up the problem could be solved. However, the challenge lies in ensuring that liquidity during liquidation can always match the level available when positions are opened. (2) For JELLY, HL's Perp had better liquidity than the spot, making it easy to manipulate the spot market to attack the perp —an example of a typical oracle attack. In the end, Hyperliquid’s solution was to forcibly set an irrelevant price for settlement, which gave the impression that HL itself conducted an oracle attack. So how do we address these problems? Once we understand that the core issue is Liquidity: providing excessively good liquidity when it shouldn’t be, without regard for the consequences, the solution must involve restricting this liquidity. However, this approach comes with the downside of a relatively worse user experience. This creates a philosophical balance between “user experience” and “security". (1) For the ETH case, the most straightforward method would be to impose restrictions on withdrawing unrealized profits. Hyperliquid later improved this by partially restricting the withdrawl of such profits. (2) For the Jelly case, possible preventive measures include: 1. Isolated margin for pairs with insufficient liquidity. 2. Isolated insurance fund. 3. Auto-deleveraging or imposing Total OI limits of a pair it becomes too large. @SynFuturesDefi is the only fully onchain orderbook DEX in the market , order-matching onchain , settlement on chain. We’ve implemented all the safety and risk control measures mentioned above, offering better protection for users. However, this is based on three premises: (1) Our view is that security is a part of the user experience—or perhaps the most important part. (2) The worse the liquidity, the more safety restrictions we apply. We operate a truly decentralized order book, and under today’s public blockchain performance (currently 2 seconds per block), it’s an objective fact that our liquidity is worse compared to Hyperliquid. However, I see developments like @base shortening block times, as well as performance improvements from projects like @monad_xyz and @pharos_network.... , which give me unprecedented confidence in supporting order books in a truly decentralized way. Enhancing the user experience in truly decentralzied Perp DEX is just around the corner! (3) Having been in the market longer, we’ve witnessed more attacks and continuously refined our model. Although we’re competitors and have had market share taken from us, I still have immense respect for Hyperliquid. Every project goes through setbacks, followed by reflection and improvement. This process is accelerated by community participation, which is exactly the beauty of openness and transparency.

Re the recent issues with @HyperliquidX ETH and Jelly cases, if we were to summarize the two with a single term, it would be "liquidity mismatch." (1) For the ETH case, the issue is that liquidity was too good when opening positions—it shouldn’t have been that good—while during liquidation, the liquidity was insufficient. So, @chameleon_jeff was right in saying that as market makers continue scaling up the problem could be solved. However, the challenge lies in ensuring that liquidity during liquidation can always match the level available when positions are opened. (2) For JELLY, HL's Perp had better liquidity than the spot, making it easy to manipulate the spot market to attack the perp —an example of a typical oracle attack. In the end, Hyperliquid’s solution was to forcibly set an irrelevant price for settlement, which gave the impression that HL itself conducted an oracle attack. So how do we address these problems? Once we understand that the core issue is Liquidity: providing excessively good liquidity when it shouldn’t be, without regard for the consequences, the solution must involve restricting this liquidity. However, this approach comes with the downside of a relatively worse user experience. This creates a philosophical balance between “user experience” and “security". (1) For the ETH case, the most straightforward method would be to impose restrictions on withdrawing unrealized profits. Hyperliquid later improved this by partially restricting the withdrawl of such profits. (2) For the Jelly case, possible preventive measures include: 1. Isolated margin for pairs with insufficient liquidity. 2. Isolated insurance fund. 3. Auto-deleveraging or imposing Total OI limits of a pair it becomes too large. @SynFuturesDefi is the only fully onchain orderbook DEX in the market , order-matching onchain , settlement on chain. We’ve implemented all the safety and risk control measures mentioned above, offering better protection for users. However, this is based on three premises: (1) Our view is that security is a part of the user experience—or perhaps the most important part. (2) The worse the liquidity, the more safety restrictions we apply. We operate a truly decentralized order book, and under today’s public blockchain performance (currently 2 seconds per block), it’s an objective fact that our liquidity is worse compared to Hyperliquid. However, I see developments like @base shortening block times, as well as performance improvements from projects like @monad_xyz and @pharos_network.... , which give me unprecedented confidence in supporting order books in a truly decentralized way. Enhancing the user experience in truly decentralzied Perp DEX is just around the corner! (3) Having been in the market longer, we’ve witnessed more attacks and continuously refined our model. Although we’re competitors and have had market share taken from us, I still have immense respect for Hyperliquid. Every project goes through setbacks, followed by reflection and improvement. This process is accelerated by community participation, which is exactly the beauty of openness and transparency.