bug with pad channel ?

[funktree]

Latest official firmware, tried this with 3 synths J6 (Roland), XFM (sonicware) JDXI (Roland) . At First I though the voices are "eaten" that is why I hooked up JDXI that has unlimited voices..

1. Turn on unit 
2. Make sure that the Pad channel is not modulated (turn all mods to 0%) and no song mode is engaged. 
3. Set the MIDI OUT to all or 1,2 whatever so you hear the synth. Use only PAD turn of Motifs and drone. 
4. Set the range to 5 voices 
5. Set the strum to slower so you can HEAR how many notes are played back
6. Set the Spread to 0 

All is nice in this mode, when you click any button to play a chord 5 notes plays back
7. Change the Spread to 1 
8. Play a few chords by clicking the clicky NDLR buttons   

Bug:
9. Not all 5 notes are played back. I would say that mode #3 root only + closed voicing would produce less than 5 notes played if you go out of range with the position knob. But mode #2,#5,#6 should play 5 notes back
10. If you have problem to reproduce, please turn on 7th chord type it'l get to this bug faster.

If this issue does not exists in lower version of firmware please point to me to one. I spent a few hours on this, reading forums, fresh manual and didn't know if I was doing something wrong but JDXI proved to me that this is most likely a bug 
My NDLR was collecting dust for 1-2 years , I hope it will not go back to that

This could be related to NDLR not sending Note OFF message on the Pad channel right ?


Also I noticed that even though I set the voices to 5 or 6 sometimes when you change the Spread the number of voices on the display changes up or down by one . 


Thanks!

[Jesse Johannesen]

Hey Funktree, I set it up in front of me had a look at the behavior you're describing and I was seeing the same thing over here, however I think it's just an artifact of the way the spread works. In any Spread setting other than "1" I think you will have some notes taken out out the pool in different octave ranges. This is mostly to avoid having the Pad part take up too much range in the mix, so in some settings the lower registers will only play the root and some octaves only play 3rds and 5ths or something, I had Steve go through the code with me a while back to see exactly what each setting does but it's been over a year ago and it's not on the top of my mind, I will see what I can do to chase the info down and pass it along to you, but all I had to do to get it back to playing 5 voices was to set pad Spread to 1. It will show all the strum/spread bars as being equally spaced and all will have the same number of segments when strum is turned up.

[funktree]

Hey Funktree, I set it up in front of me had a look at the behavior you're describing and I was seeing the same thing over here, however I think it's just an artifact of the way the spread works. In any Spread setting other than "1" I think you will have some notes taken out out the pool in different octave ranges. This is mostly to avoid having the Pad part take up too much range in the mix, so in some settings the lower registers will only play the root and some octaves only play 3rds and 5ths or something, I had Steve go through the code with me a while back to see exactly what each setting does but it's been over a year ago and it's not on the top of my mind, I will see what I can do to chase the info down and pass it along to you, but all I had to do to get it back to playing 5 voices was to set pad Spread to 1. It will show all the strum/spread bars as being equally spaced and all will have the same number of segments when strum is turned up.

Jesse,

Thanks for quick response. If I go with your explanation then how would I get the Pad channel playing open voicing with 5 voices ? If I understand your explanation NDLR decides to play 3 4 or 5 notes based on spread (Spread choice 3 actually works this way which is ok).
 
Some synths have only 4 or 6 (XFM sonicware) voices polyphony what If I want to use all the voices but spread to 2 octaves. Eg: Move the 3rd one octave higher which I think spread settings 5,6 would actually do.

Second issue. From the  manual (page 36 bottom) the spread feature is "ignored" for 4 voice pads thus 4 voices are always close voiced (Just checked and confirmed with JDXI). That is bummer too why I would need four voices to be always close voiced ? Wouldn't you spread even if you have 4 voice synth ? (most of the boutiques) Or utilizing poly chain with Elektron gear (Syntakt,,etc)  where I want to utilize only 4 Channels for chord(s) ? I suppose with syntakt I just move the 3rd/5th an octave up or the root octave down. But If I turn on inversions on NDLR then simulating "spread" with four voices it's nightmare. (the 3rd is not always the second voice).  Yes lazy me I can edit the sequenced notes if I hit record.  for four chords so be it.

So basically Ppl with boutiques (Roland) or utilizing poly chain due to monophonic gear (Volca, most of the elektron gear) cannot spread voices.

[Jesse Johannesen]

If I understand your question correctly you should be able to turn on the Polychaining feature in the settings menu and choose how many channels you want (1-4) however the channels will be sequential starting with the Pad channel number, so it's a good idea to make sure that the Pad channel is set to a number where there are A) 3 more numerically above it, and B) that these are not already being used as Motif or Drone part channels.
This feature just hockets notes to each channel in the order they come in though, so it's not going to get overly fancy as far as sending 3rd off to specific voices, but it may be worth a look. It's been long enough since I've used the feature though that I don't remember if there are any gotchas that would prevent you from using more than 4 voice pad settings. I will mess around with it and see what I can discover.

---

Here is a link to my previous post about the spread settings:
https://conductivelabs.com/forum/showthr...d+settings

[funktree]

If I understand your question correctly you should be able to turn on the Polychaining feature in the settings menu and choose how many channels you want (1-4) however the channels will be sequential starting with the Pad channel number, so it's a good idea to make sure that the Pad channel is set to a number where there are A) 3 more numerically above it, and B) that these are not already being used as Motif or Drone part channels.
This feature just hockets notes to each channel in the order they come in though, so it's not going to get overly fancy as far as sending 3rd off to specific voices, but it may be worth a look. It's been long enough since I've used the feature though that I don't remember if there are any gotchas that would prevent you from using more than 4 voice pad settings. I will mess around with it and see what I can discover.

---

Here is a link to my previous post about the spread settings:
https://conductivelabs.com/forum/showthr...d+settings

Holly macaroni!

Your link helped a lot. I've never looked at Appendix "D" in manual. Thanks for that Jesse.

OK now I get the "random" number of notes played and how it works.
BTW your explanation (following your link) is better than Appendix "D" you'd go deeper for each setting...

So if I understand correctly the Spread parameter basically "forces" certain intervals to be played ONLY in "certain" octaves besides the Spread=1. 

Here are a few "strange" behaviors I observed (C-major key:
1. Spread 2 / Range 8
Some triads play only 3 notes Yes you can "fix" it with moving the position knob

2. Spread 5 / Range 10
Most of the 7th chords don't play more than 3 notes (This behavior is "obvious" from the definition but then if you use 7th in your progression the PAD becomes weak for this particular chord type even with 10 voices) So I suppose warning do not use Spread=5 with progression containing 7th chord(s) could help.  

3. Spread 6 / Range 10
Most of the sus2 sus4 chords play only(!) 3 notes, moving the position knob does not help

I think I could squeeze 1-2 extra examples of this "strange" behavior even with Range=12+. Bottom line, even with voices set to 10 (range=10) the NDLR might not play 4 voices, thus making chord progressions uneven or weak. Wouldn't you agree ?

Thanks Again

[Jesse Johannesen]

I was definitely noticing some of these cases myself but I still think it likely comes down to how the range works against the spread and how it's kind of a labelling issue with any other setting than 1. I sort of have to think about the range value in terms of spread 1, and then the dropped notes are subtracted from that, so it makes sense that moving the position fixes the issue, (I think). I'll have another look, but I think that's what's going on here. It would be cool if the Pad position had a sort of "keytracking" so that it would always have the same pool of notes around each chord root.

Anyway, glad my link made sense and thanks for reporting back.