Author: ZmnSCPxj 2021-05-18 11:36:49
Published on: 2021-05-18T11:36:49+00:00
In this correspondence, ZmnSCPxj responds to Michael's comment on patents and assures that oPoW only uses SHA3 and low precision linear matrix multiplication, which are widely used in the industry for accelerating 8-bit linear matrix multiplication for AI. However, adding more to SHA3 still risks layer violations that might introduce hidden optimizations. Even if the combination of linear matrix multiplication and SHA3 may have hidden optimizations, there are no current incentives to find such optimizations until Bitcoin moves to this, at which point it would be highly infeasible to revert to SHA256d. Therefore, changes to PoW are highly discouraged. Additionally, ASICBOOST was not an optimization of SHA256 or SHA256d but an optimization of SHA256d-on-a-Bitcoin-block-header, meaning it cannot speed up general SHA256 or even general SHA256d.
Updated on: 2023-06-14T22:00:19.412479+00:00