You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Protocols like Mina and Aztec have replaced Sha with Poseidon for the time being, as it is faster to optimize in their respective zk proving languages. We should standardize this option, as well as expose consistent and correct JavaScript libraries for it, as well as include an amendment to the EIP. I'm open to renaming V1 and V2 till like V1-Poseidon or something like that, or releasing V3 for Poseidon where the proving always happens on the user/client side because you can include all the hashing there.
They also be worth noting that v1v2v3 may not be the best ways to define the standard, because we're not incrementing the version such that later versions are better, they just all encapsulate different trade-offs.
It would be ideal to standardize the mina version so nothing has to change!
The text was updated successfully, but these errors were encountered:
Divide-By-0
changed the title
Add posiedon option to both v1 and v2
Add posiedon option to v1
Nov 29, 2024
So, my plan is the following. Along the way I will be adding abstraction/generality, though as much as possible while avoiding getting tumbled in weeds.
describe Mina implementation
add and reference Mina Nullifier RFC
look into the Noir lib pull request if there's anything to add to the result of previous steps
start to add this to the reference implementation here
in parallel review the patch to the ERC and commit the contribution
Protocols like Mina and Aztec have replaced Sha with Poseidon for the time being, as it is faster to optimize in their respective zk proving languages. We should standardize this option, as well as expose consistent and correct JavaScript libraries for it, as well as include an amendment to the EIP. I'm open to renaming V1 and V2 till like V1-Poseidon or something like that, or releasing V3 for Poseidon where the proving always happens on the user/client side because you can include all the hashing there.
They also be worth noting that v1v2v3 may not be the best ways to define the standard, because we're not incrementing the version such that later versions are better, they just all encapsulate different trade-offs.
It would be ideal to standardize the mina version so nothing has to change!
The text was updated successfully, but these errors were encountered: