Group Admins

  • Profile picture of Norbert Schnell

Group Mods

  • Profile picture of Frederic Bevilacqua
  • Profile picture of schwarz
  • Profile picture of borghesi

MuBu for Max

Public Group active 11 hours, 30 minutes ago

MuBu for Max user group.

pipo.slice help

Author 2 Subscribed Users |
Profile photo of danijoss

Hello everyone,

I’m really excited to have discovered the mubu package! I’m still trying to work through the documentation and examples.

My question is about [pipo.slice]. In the help file, it appears that the data track has three columns, i.e. 3 separate signals, that are output by [] as a list of 3 elements.

Then these lists go to [pipo.slice], that is supposed to do windowing for fft’s etc.
How are these three signals handled? Are they processed separately, i.e. we get 3 separate FFTs and spectra? Are they joined or summed somehow?
It looks like just the first is processed. If so, assuming we wanted to analyse all 3 signals, would we need to e.g. split the list and store to different tracks?

Also, the helpfile mentions “real-time fft using pipo.slice”, but I can see no fft process in the pipo object (like slice:fft).

Thanks a lot for any help!

November 21, 2018 at 00:31 #28392
Profile photo of danijoss

Could you guys please at least point me in the right direction?

November 23, 2018 at 08:53 #28419
Profile photo of schwarz

Hi Dani,

you’re right, the slice only uses the first element of the input frame, and the fft has been removed from the example (but not from the comment) for clarity.
To get 3 ffts you’d have to use 3 pipo instances for real-time, or 3 mubu.process with slice:select:fft for offline for now.

HOWEVER: If column-wise fft analysis is and important use-case we might consider adding a column-wise slice/fft option.


November 23, 2018 at 10:40 #28421
Profile photo of danijoss

Hi Diemo,

Thans for your answer.
It was more important for me to be clear on what input the various models need. I’m starting work on a project involving IMU sensors on dancers, so there’ll be plenty of data needing analysis. Separate instances is perfectly fine.

Thanks again,


November 23, 2018 at 11:58 #28422

You must be logged in to reply to this topic.

Log in now