Polyseq ... really?

I’m very interested in MIDIpal, mainly because of the polyseq. After reading the manual, I’m not really sure how the polyseq app works. Does it mean that I have to distribute the single notes of a chord over multiple monophonic tracks (or lanes)? In this case, the name “polyseq” is a little bit misleading. Or is it possible to enter complete chords step by step? If no, how are the chances to implement such a feature?

The dispatcher app is used for splitting chords to multiple tracks…

As far as I understand the polyseq app it is essentially a normal sequencer with overdub functionality - You record the first sequence, then “tack”/dub the next sequence on top of the first one, and so on…

Thanks for clarification. Unfortunately, that’s not what I’m looking for. I want to use MIDIpal to sequence chord progressions. Doing this with overdubbing single notes is a little bit too painful :wink: However, perhaps there is a chance for a improved firmware anytime.

I’m pretty sure that you can record chords in each sequence as well, but I’ll wait for pichenettes (or a midipal owner) to confirm it :slight_smile:

> Does it mean that I have to distribute the single notes of a chord over multiple monophonic tracks (or lanes)?

The MIDIpal polyseq is a recreation of the JX3P sequencer. You have to record each mono voice one by one. It is more a polyphonic extension of the simple step by step sequencer of the SH-101 than a MC50… I think what you would like would be something able to record a MIDI stream in realtime. The MIDIpal does not do that.

Two notes:

  • I think it would be a mistake to implement chord recording in such a simple step by step recorder. This would require the firmware to be able to draw the line between a chord (all the MIDI notes clustered to the same step) and a rapidly played arpeggio (each note goes into a step). I like recording rapid-fire arpeggio in a step sequencer - don’t want them to be mistaken as chords.
  • The best paradigm for what you’d like would be a traditional realtime sequencer. Then it needs a quantization menu. Then people will want tracks, edit functions, memories… The MIDIpal has only 1kb of memory anyway so no matter how good I make it, it would still be very limited. It’s a good platform for pattern-like / utilitarian features - but not to build a complete sequencer.

Thanks for your feedback. No, I don’t mean realtime recording. For this purpose I’ve enough stuff :wink: I’m looking for a simple and compact solution: Hit a chord, switch to the next step, hit the next chord and so on … Do you know the step recording mode in the Roland Fantom? Hit some keys, after releasing the last key of the chord, the step has been entered and the sequencer is ready for the next step (respective chord). My purpose is to use a poly synth for simple chord progressions as addition to my Electribe’s sequencer (which is monophonic only). From the technical view, is it difficult to store/send three or four notes per step instead of one single note?

@klangsulfat: i think midibox seq v4. (or the light version of it) does what you are looking for.

Yes, especially the seq v4 lite seems to be really great. But it’s DIY-only, not as compact as MIDIpal and featurewise oversized (at least for my special purpose) :wink: I also looked in the arduino world for a solution. There are some step/analog sequencers, but nothing is polyphonic.

The
MBseq v4 lite is cheap, so it won’t hurt your wallet that bad :smiley: It only takes an evening to make one. So, don’t be afraid - I’ve seen worse DIY projects.

mbseq4lite does NOT record note(s) after note(s) as you enter them. you either have to record them live, while the sequencer runs or step by step. and by step by step is meant that you first have to press a general purpose button referencing to the step position you want to enter notes to and then press the notes you want to enter. it has no overdub function. it always quantizes the notes entered to 16th (in length as as well as start position). so - it has its limitations but is a nice tool really. and its developed further and the codebase is well documented. you could implement the needed functions by your own.

I’ll chime in here, as I have just been playing my jx3p. I don’t know exactly how it works on the jx, but you can enter a chord per step without overdubbing.
I guess there is some sort of timer perhaps 50ms that determines if what you played was a chord or fast sequence of notes. If I want to enter a whole bunch of (monophonic) notes then I have to slow down otherwise it thinks that I played a chord. It takes soem getting used to.
I don’t know if that sort of a system would be hard to do???

Now that’s puzzling! Is that the original firmware or the kiwitechnics upgrade?

I guess I’ll have to modify it to make it similar to the original then…

Attached is the original manual of the JX3P (which I found here: http://fa.utfs.org/diy/jx3p/jx3p_man.htm). According to page 8 it’s possible to enter chords with up to six notes per step.

no, my one is not upgraded. perhaps there were different roland software versions though?
otherwise it works pretty much like the 101 except you can do chords and overdub.

I gave it a shot. This was trickier than I thought and I am not sure I got it right. If anybody wants to try it and let me know how it works for them…

I use a 100ms delay to switch to the next step after a note has been entered - so you have to make sure the notes are played less than 50ms apart when playing a chord. The JX3P manual mentions something about playing legato - I suspect they use the note on/off information to detect step transitions (switch to next step once all notes have been released) - I could not do it here because all the memory is occupied by the pattern, no room for a note stack.

I’ll check it out tomorrow! Thanks!

sounds cool. a truly jx3p-style sequencer would be a most welcome addition!

however: does this fw build also contain the sh101-style sequencer? or does the jx sequencer replace the sh sequencer? in that case, it’s not for me. there’s no way i’m letting go the sh sequencer, not even for a jx seq.

This firmware contains only the JX3P-style sequencer - because it requires the whole eeprom memory and cannot thus live with other apps.

Are you using the accent/slide feature? Otherwise this sequencer does the same as the SH seq.

accent and slide are what makes the sh seq so much fun!

oh well, if there is not enough room for both sequencers in one rom, i guess i’ll have to get a second midipal eventually…

(or keep waiting for the ‘midipal’ ios app suite. :wink: )

oh btw, shouldn’t it be possible to incorporate the sh sequencer’s ‘accent’ feature into the jx sequencer? or would that be another memory restriction?