"I hope more people doing ZK (zero-knowledge proofs) and FHE (fully homomorphic encryption) can express performance in terms of cost ratios (such as" time required for encryption computation/raw computation time "), rather than just saying" we can do N operations per second ", Vitalik wrote on the X platform.
This approach is even less hardware-dependent and provides a very useful metric: how much efficiency do I sacrifice when I change my application from "trust-dependent" to "cryptography-dependent"?
It is also usually more suitable for making performance estimates, because as a developer, I already know how long the original computation will take, and I can estimate the performance by simply multiplying the cost ratio.
(Yes, I know it's not easy because of the different types of operations between execution and proof, especially in terms of SIMD/parallelization and memory access, so even the overhead ratio is partially hardware-dependent. But even so, I still think "overhead magnification" is a very valuable metric, although it's not perfect.) "
Vitalik: I hope that more researchers working on ZK and FHE can express performance in terms of cost ratio, replacing the number of operations per second metric
2025-10-18 05:39:04
Disclaimer:
1. The information provided does not constitute investment advice. Investors should make independent decisions and bear all risks themselves.
2. The copyright of this content belongs to the original author. The views expressed herein are solely those of the author and do not represent the stance or position of this website.
Previous article：
Vitalik：希望更多从事 ZK和 FHE的研究人员能用开销比来表达性能，取代每秒运算量指标Next article：
两地址持续增持并质押FF代币，合计质押超4400万枚
