double precision?

[ayham-1]

Hello,

is it possible to use this project for f64 calculations?

if not, what would implementing such a system look like?

I would like to help if possible.

[tonitch]

so if i'm not mistaken, the project is mainly based on the F32xX type witch is a @vector(f32, X) (where X could be 4, 8 or 16)

So no the f64 would be limited by the operator on @vector(f32, X) of zig. But I suppose that changing all of the f32 by f64 could be a solution. I haven't tested but I don't think that would be a problem as the @vector type of zig should handle major issue.