can't replicate results

[nukadelic]

[arakis]

It seems it was introduce by the PR #1 by @Franzsardoma . Instead of reverting it, i pulled the threashold much higher. I'm not sure if this is the correct way, but for now it should work. Here's the change i made: 19a16a5

@Franzsardoma @wtaisto any idea?