smoothY options damage synthetic data (no noise)

[jobo322]

smoothY as true

In the case of the first algorithm, it implies a possible peak because it crosses zero. Should we deprecate this option? Data smoothing should be treated as an additional feature separate from peak picking in the GSD function.

[lpatiny]

This is the one with smoothing ?

Seems like a bug in the smoothing algorithm. What is the approach to do smoothing ?

[jobo322]

it uses the derivative zero from savitzky-golay filter