0
Please be positive and constructive with your questions and comments.

Re: n0nx0x2

by gelabs on Tue May 02, 2017 2:58 pm

Indeed, that works. :)
gelabs
 
Posts: 13
Joined: Sun Jun 19, 2011 4:13 pm

Re: n0nx0x2

by gelabs on Thu May 25, 2017 12:03 pm

The x0xb0x drifts (it's slowing down) slowly while being slaved to midiclock when using MIDI-to-DinSync prediction. Turning prediction off fixes the problem.

Also, is it possible to have another additional operating mode for the sequencer : think of a SH-101 for example, you have a list of notes and you can trigger them independently of the length of the sequence. Say I have a list of five notes (different or same) and 11 steps ("trigs") with slides and accents; if I press play, then the five notes will never play at the same "place" when the sequence loops, ie. they will sometimes be accented or slided depending on which step ("trig") they encounter. Dunno if that makes sense, English is not my native language :)
Anyway, that would make interesting and more evolving sequences possible. Just an idea, I couldn't do it myself outside of puredata ...
gelabs
 
Posts: 13
Joined: Sun Jun 19, 2011 4:13 pm

Re: n0nx0x2

by antto on Thu May 25, 2017 12:44 pm

it drifts and slows down!? that's weird
what are you using to generate the MIDIsync?

the only time i've had midi slowing down (or was it speeding up) was when i was generating MIDI from a sort-of DAW (SynthEdit actually) where (iirc) i wasn't using ASIO for the audio, so stuff desync'ed slowly in software.. stopping the DAW dsp process and then starting it again fixes that issue (but then it starts drifting again)

but based on what you said, this can't be the issue..

about the other question, it's not very clear to me how that should work (i'm not familiar with the SH-101 at all)
but the sequencer already is carying a heavy burden - the "scattor" (FreezeFrame), so adding yet another feature which mangles the pattern data while playing it would be complicated
this sort of thing could be added into a sepparate "mode" .. we got just one left (mode C)

antto
 
Posts: 1570
Joined: Thu Apr 15, 2010 3:21 pm
Location: 127.0.0.1

Re: n0nx0x2

by gelabs on Thu May 25, 2017 3:09 pm

Well, midiclock is generated by Roland TR-8 (with "precise" timing, not the 808-emulated one), so I guess it's not the best clock available :)

The "freeze frame" feature is indeed a nice thing I am beginning to learn.

As for how the 101 sequencer works, well, it's just a list of pitches and you trig them as notes by sending pulses in the 101 from an 808 for example.
The thing is, the melody evolves because the notes have not always the same value on a given step, they shift along the sequence.

Like this : ( "_" is a rest, "O" is a gate), the list of pitches is C,D,E,F,G, pattern is 16 steps long :

Code: Select all | TOGGLE FULL SIZE
O _ _ O _ O O _ O _ _ _ O _ O _ / O _ _ O _ O O _ O _ _ _ O _ O _ / O _ _ O _ O O _ O _ _ _ O _ O _ /
C _ _ D _ E F _ G _ _ _ C _ D _ / E _ _ F _ G C _ D _ _ _ E _ F _ / G _ _ C _ D E _ F _ _ _ G _ C _ /


Ah, there is actually a great depiction of this thing here : https://www.youtube.com/watch?v=sfKzHpsQkxo

Edit: I just realized that you can do that on the Roland TB-03 ...
gelabs
 
Posts: 13
Joined: Sun Jun 19, 2011 4:13 pm

Re: n0nx0x2

by antto on Sat May 27, 2017 10:50 am

so then it might be missing MIDI clock ticks
you can test that:
- run the master, check that both machines run "together" at first
- push the tempo on the master high (this should hopefully speed things up)
- wait for some time, and try to notice if the x0x "slows down" or glitches or anything
- bring the tempo down - see if the x0x is really running "behind"
- bring the tempo down a lot - see if that scales the lateness of the x0x - if it does - it means that the slave (the x0x) has missed one or more clock ticks
- press stop on the master, and then press play - does the x0x start and run "together" with the master as in the beginning? it probably would

now, it might be a good idea to also use another slave device, some other drum machine or something, preferably something from a supposedly "good brand"
because if both slave devices "fall behind" - the issue might be in the master (but i doubt, since it's a Roland, it should be decent)

and btw, which version of n0nx0x2 are you running?

as for the other thing, i watched the video.. this looks cool, but it will need some thinking in order to get it to "fit" and "make sense" with the rest of the sequencer
basically, you want to use the patterns which you already have, only as a list of notes, but trigger them externally
this will not work with the normal pattern playing stuff (the modes on the right side of the selector) but it can be (maybe) added to pattern-trigger mode, or (worst case) put on a dedicated mode

essentially, you need to have the pattern/chain selection mechanism available, so you can select a pattern or a chain of patterns like usual
but instead of playing those patterns like usual - the sequencer will only use their pitch data, and wait for "trigger"
now, the only thing i can think of as "trigger" is MIDI
so when a midi note comes in - the sequencer will play a note from the pattern/chain, and move to the next one, and loop when it reaches the end
this is very doable, but there are a few philosophical questions about the details:
- should the length of the played note equal that of the trigger note, or should it be fixed? should it even be "quantized" to the sync clock?
- what happens with pitches from the pattern which have the "slide" attribute? does it get discarded?
- what happens when you overlap trigger notes? slide? slide only if the pitch had the slide attribute?
etc..

antto
 
Posts: 1570
Joined: Thu Apr 15, 2010 3:21 pm
Location: 127.0.0.1

Please be positive and constructive with your questions and comments.