# Research: zkEVM Comparison/Deribit,Coinglass API
- [[Enshrined Rollup]]
- Enshrined rollup v.s. Discrete rollup -> Public Sector v.s. Private Sector
- Enshrined rollup -> Original L1 based zkEVM -> The Splurge
- Feature -> No need Merkle paths or Verkle proofs
- **no reexecution**
Cause only need to rerun proofs, potentially increase gas limit.
- **simpler consensus**
- **no state witnesses**
Greatly increases the consensus bandwidth efficiency for validators and unlocks a higher EVM gas limit.
- **safer light clients**
Quickly filter invalid state roots, in contrast to the slow filtering of invalid state roots with fraud proofs. This allows for safer Ethereum-to-L1 bridges.
- Advantage
- **social consensus** Removing the need for governance tokens to perform rollup upgrades
- **subsidised proof verification** ~~廢話~~
- **settlement latency** ~~廢話~~
- **optimal liveness**
- **state root EVM equivalence**:
Tooling and light clients for an enshrined zkEVM work out of the box. Many smart contract rollups may elect to not have EVM state root equivalence, instead targetting a Solidity-compatible VM (e.g. zkSync) or a bytecode-equivalent EVM (e.g. Scroll).
- **network effects** ~~廢話~~
- Disadvantage
- **no public goods funding** -> No token, incentive, income
- **suboptimal compression** -> Tech Issue, Data
- **VM inflexibility** -> Tech Issue, VM
- **harder preconfirmations** -> MEV, tx order issue
- **last mover** -> Tech Issue, circuit
- 我要發瘋，要確耶
- STARK > SNARK
- https://twitter.com/apolynya/status/1507586019171835905
- https://twitter.com/VitalikButerin/status/1466411377107558402/photo/1
- https://www.reddit.com/r/ethereum/comments/vrx9xe/ama_we_are_ef_research_pt_8_07_july_2022/
- [[zkZVM Comparison]]
- https://vitalik.ca/general/2022/08/04/zkevm.html
- Competitors
- Polygon Hermez
- zkSync
- Scroll
- [PSE](https://github.com/privacy-scaling-explorations/zkevm-circuits)
- Startware
- [Alpha Compiler](https://medium.com/starkware/starknet-alpha-2-4aa116f0ecfc)
- Overview
- Full Function v.s. Faster Deploy v.s Faster Proof Gen Time
- Type1 - PSE
- Fully Compatible - Long Prove Gen Time (Hours)
- Type2 - Polygon, Scroll (Closer to Type3)
- zkEVM instead EVM - Long Prove Gen Time
- Unproven, still a long way
- Type2.5
- Additional gas limit
- Type3
- Easier, Shorter Prove Gen Time, Imcompatible
- Remove
- Precomplie - Not sure influence
- Contract Code, Memory, Stack
- Type4 - zkSync, Starkware
- Higher Lang to compile Solidity, Vyper
- Decentralize, Lower Cost and Entry
- Imcompatible
- [[Deribit API]]
- [[Coinglass API]]