Renaud Dubois
Renaud Dubois
The obtained gain is around 40%: (results measured using --print-info with a single call to public_key_ ECDSA test standard implementation over sec256k1 Number of steps: 190047 (originally, 190047) Used memory...
Using IR==true in toml crushes FCL performances, leading to incorrect bench. Provide benchmarks in line with on chain performances: https://goerli.basescan.org/address/0xe9399d1183a5cf9e14b120875a616b6e2bcb840a
PR #52 fixed a problem in Inner inlined edge case. Deploy the latest version of FCL on chain.
test r and s value not zeroes
Current benchmarks are not consistent with on chain results. FCL is displayed as providing 370K tx, while on chain measurements are maximum 227K. Also, what is state as average case...
Toml is not set to provide the optimizer for the number of steps, explaining (partly) the difference between real measured onchain performances and this repo. Set toml (see PR feat/onchain),...
### Component Forge ### Describe the feature you would like Add P256 verification keyword. ### Additional context Currently there is no way to test P256 in simulation environnement, lacking a...
This proposal creates a precompiled contract that performs NTT and Inverse NTT transformations. This provides a way to have efficient and fast polynomial multiplication for Post Quantum signatures and Starks...
Edwards curves shall follow the RFC8032 encoding rather than the BIP62 low S to comply better with standards. For instance implementing a RFC9591 will not succeed with zkkit when point...
A description of the round constants generation is lacking. Launching the script provided by the Poseidon authors seems to be non compliant: https://extgit.isec.tugraz.at/krypto/hadeshash/-/blob/master/code/generate_parameters_grain.sage with starkcurve parameters: FIELD = 1 #Prime...