Walter Smuts

Not sure about specific applications, but I can get some rough numbers for my specific machine: ```rust use std::mem::MaybeUninit; use std::time::Instant; fn main() { let start = Instant::now(); let mut...

I also initially got a couple of runs where the non-zeroing version is slower. Weird. Not really sure what's happening there. Could be noise. CPU heating up. Warming up some...

I tried out @ejmahler's suggestion and it doesn't seem to make much of a difference to consistency. But since I re-attempted the test I noticed I was running with the...

https://docs.rs/rustfft/latest/rustfft/#normalization

Woot woot!

I don't think the issue will be resolved anytime soon so I would recommend to merge for now. Unfortunately it means a major version change for `fftconvolve` so consumers would...

> Yeah unfortunately I can't pass in a planner from Python Not sure if this is completely true, but consider using [easyfft](https://crates.io/crates/easyfft) that manages the planner for you.

Have a look at the rendered markdown here: https://github.com/WalterSmuts/rofi-pass/tree/markdown-fix

https://www.markdownguide.org/basic-syntax/#reference-style-links

I think it's okay to use two different language features :shrug: But ultimately is a style/taste decision.