Content
@
0 reply
0 recast
0 reaction
franco
@francos.eth
Learning about Circle STARKs and fell into a deep rabbit hole of Riemann-Roch spaces. I love crypto. https://eprint.iacr.org/2024/278
6 replies
31 recasts
106 reactions
EulerLagrangodamus - bank/acc
@eulerlagrange.eth
What’s the improvement metrics?
1 reply
1 recast
8 reactions
franco
@francos.eth
I haven't looked into that yet. I'm focused on their use, along with covenants, to enable general computation on Bitcoin. If the soft fork to reactivate OP_CAT is done, that is.
3 replies
0 recast
0 reaction
franco
@francos.eth
From the paper: "As a preview of what to expect in terms of performance, we provide preliminary benchmarks of the circle FFT over the Mersenne prime field M31, compared to the regular FFT over the equally sized Babybear field, see Table 1. The advantage of vectorized M31 arithmetics (AVX2 multiplications are about 40% faster than with BabyBear, see also [HLN23]) directly translates to the single-threaded FFT cost. With high thread count, our current implementation becomes memory-bound, and thus, M31 CFFT and BabyBear performance is similar. Take note that the current implementation does not apply memory optimizations (such as Radix 2k FFT or 4-step). With such optimizations, memory bandwidth is expected to be much less of a bottleneck on personal hardware."
1 reply
0 recast
1 reaction
franco
@francos.eth
There's also some benchmarks in the stwo repo: https://starkware-libs.github.io/stwo/dev/bench/index.html Which I don't know how to interpret yet
0 reply
0 recast
0 reaction
franco
@francos.eth
Oh, I think this blog post explains it: https://starkware.co/blog/starkware-new-proving-record/
0 reply
0 recast
0 reaction
