Back to the main page.

Bug 334 - specest_mtmconvol shifts time-bins 1 bin to the right, it shouldn't

Reported 2010-12-21 12:24:00 +0100
Modified 2011-01-14 09:54:43 +0100
Product: FieldTrip
Component: core
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P1 normal
Assigned to: Roemer van der Meij
Depends on:
See also:

Roemer van der Meij - 2010-12-21 12:24:56 +0100

This caused all data to be shifted by 1 sample. The impact is minimal, but it is there. How should I go about implementing the fix for this bug? Data analyzed in the pre- and post-fix will differ slightly. The bug could sit here for eternity, as it's only a shift of 1 sample....

Roemer van der Meij - 2010-12-21 15:35:39 +0100

On second thought, it should. If it doesn't it will try to select the first time-bin (even though in the end it won't because of the time-window) with sample number 0. And the last one will be selected with sample number N-1. This is incorrect, and it was so in the past in freqanalys_wltconvol.