Benefits of v2

Key benefits

  • Faster trading thanks to L2 (Arbitrum)
  • Better liquidity and lower spreads thanks to Uniswap v3 under the hood
  • Leverage for market makers - LP with leverage
  • All-new UI for easier trading on mobile and desktop
  • Dynamic liquidity for improved system stability and market responsiveness
  • Better composability, allowing integration with a host of other DeFi protocols and tools
  • Cross-margining for flexible position management

Key issues with v1

Long/short skew

Perpetual Protocol v1 encountered a design bottleneck around the dynamic adjustment of the AMM k value. Simply put, the amount of liquidity in the exchange could not be changed without introducing further hard problems to solve.

This led to a key 'weirdness' with v1 vAMMs. Typical perpetual futures exchanges rely on a relatively balanced number of long and short holders to exchange funding payments. On a vAMM with a static k like we have in v1, actually the long/short ratio has almost zero bearing on the funding rate. The funding rate is determined by

  1. how far the price is from initial AMM launch price
  2. how many longs (or shorts) are needed to move the price from initial to current

So let's say $PERP was $5 when the AMM launched. Now it's $17. That means we need a huge amount of longs to open and remain open in order to maintain the current price. No shorts are needed whatsoever. If funding goes positive for some reason, longs exit, sending the funding negative again.

This situation results in a lack of traders to pay funding payments—virtually all positions earn payments. This shortfall is covered by the exchange insurance fund, resulting in systemic instability. This is the fundamental flaw in v1 of the protocol.

Slow execution

Although it is several times faster than Ethereum, the xDai sidechain still takes up to 5 seconds to confirm transactions. In addition, node performance can be unreliable, sometimes taking several minutes to confirm transactions for unknown reasons. While xDai has performed very well overall and is also hard at work on further scaling solutions such as "Arbitrum on xDai", current Ethereum L2 solutions such as Optimism have much better performance today.