Created on 2020-10-12 15:08:14 by Didier Vibert. % Done: 0

in order to be able to let 2 candidates be close enough (for instance 2 candidates coming from the OII doublet ?) it may be worth finding 2 local maxima in a second pass window, instead of reducing the size of the window to handle each of the close candidates in separate non-overlapping 2nd pass windows.

note: after studying the typical displacement of a redshift candidate between 1st pass and 2nd pass (should be answered in #5625) we may conclude that a small 2nd pass window is sufficient and thus this issue would become useless.