1/ This thesis around Oracle usage in AMMs has mostly played out with the initial designs failing to capture high TVLs since the design was economically unfavorable for LPs
Kudos to @Bancor and @BreederDodo teams for evolving their models to find better designs
Kudos to @Bancor and @BreederDodo teams for evolving their models to find better designs
https://twitter.com/Rewkang/status/1310719318405951488
2/ The oracle design is expensive for both the protocol (gas fees to push updates) and LPs (excessive permanent loss)
Bancor V2.1 eliminates oracle usage all together and Dodo V2 seems to focus more on private MMs & customizable bonding curves
Bancor V2.1 eliminates oracle usage all together and Dodo V2 seems to focus more on private MMs & customizable bonding curves
3/ We still see many projects pitching oracle based AMM designs, but even if they kind of work on L2 I don't find them that exciting because relying on external CEX liquidity is fundamentally not aspirational and doesn't scale in a world where liquidity migrates to DEXs
4/ Uniswap and Sushiswap have been able to build massive liquidity partially because they serve as the liquidity launchpad for new projects, and liquidity elsewhere is not necessary
They've become the "source of truth" ie referential market price for many assets
Now that's bomb
They've become the "source of truth" ie referential market price for many assets
Now that's bomb
• • •
Missing some Tweet in this thread? You can try to
force a refresh
