Content
@
0 reply
4 recasts
4 reactions
Sergey Potekhin
@fastfourier.eth
Farcaster-solidity 0.1.2 is released 🎉 - Added Frame Actions support, so you can attest to Frames actions onchain In layman's terms - each Frame click comes with a signature, which proves that the user indeed clicked the button. Now you can verify this click onchain, creating trustless onchain frames!
7 replies
11 recasts
82 reactions
Wilson Cusack
@wilsoncusack
How much gas is this? Was worried it might be over the block limit :D
3 replies
0 recast
2 reactions
Sergey Potekhin
@fastfourier.eth
Not even close, but still expensive :) Blake3 hashing is 500k-1m gas, and Ed25519 signature verification is also 500k-1m gas, so most of the time it's close to 1.5m gas per message.
3 replies
1 recast
2 reactions
josh crites
@crites
It might be cheaper to generate a zk proof of the signature and verify that. iirc, verifying Noir proofs is ~500k gas and it looks like there are all of the required methods for verifying signature in the circuit
1 reply
0 recast
1 reaction
Wilson Cusack
@wilsoncusack
Definitely would be. But hard to do zk with blake3 hash and indeterminate message length, I think
1 reply
0 recast
0 reaction
josh crites
@crites
can you just use the message hash instead?
1 reply
0 recast
0 reaction
Wilson Cusack
@wilsoncusack
Well I think depends if you want to verify certain things about the message. Eg it may not be enough to simply know they signed it, you may to to assert something about the data
1 reply
0 recast
2 reactions
josh crites
@crites
ah yes, you're right. what about having a large, fixed length string as input to the circuit. then you can slice it to the appropriate length given your string length for input into blake3
1 reply
0 recast
2 reactions
