Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Slides and Accents
#1
For Accents, assume just use value of 127 for velocity, but for SLIDES, am I missing something or does there need to be a way to overlap note lengths such that midi will send slide information to my synths?  In particular, looking to get an acid vibe going and can't do so without slides/accents being in their happy place.
Reply
#2
(09-06-2020, 08:48 AM)shagghie Wrote: For Accents, assume just use value of 127 for velocity, but for SLIDES, am I missing something or does there need to be a way to overlap note lengths such that midi will send slide information to my synths?  In particular, looking to get an acid vibe going and can't do so without slides/accents being in their happy place.
If you know what the CC number is that your synth receives slide data on you can make a mod matrix routing that uses any motif pattern as an LFO shape and has that MIDI CC as a destination. Not a perfect solution but a little creativity with this method will provide tons of rewarding experimental results. I don't know for sure if setting the pattern length in the pattern affects it's length as an LFO or not, but I would mess with that and see if I were you. you can also edit the pattern while it's playing for even more interesting results.
Jesse
Reply
#3
(09-06-2020, 11:49 AM)Jesse Johannesen Wrote:
(09-06-2020, 08:48 AM)shagghie Wrote: For Accents, assume just use value of 127 for velocity, but for SLIDES, am I missing something or does there need to be a way to overlap note lengths such that midi will send slide information to my synths?  In particular, looking to get an acid vibe going and can't do so without slides/accents being in their happy place.
If you know what the CC number is that your synth receives slide data on you can make a mod matrix routing that uses any motif pattern as an LFO shape and has that MIDI CC as a destination. Not a perfect solution but a little creativity with this method will provide tons of rewarding experimental results. I don't know for sure if setting the pattern length in the pattern affects it's length as an LFO or not, but I would mess with that and see if I were you. you can also edit the pattern while it's playing for even more interesting results.
Jesse

Interesting, good stuff thank you. Doesn’t just overlapping notes result in a slide normally?
The note “off” of the first just needs to wait until the note “on” of the second note so there is an overlap.

I could see this being a simple feature request, just add a “Slide” setting for each step in the rhythm editor, just like the REST and TIE’s?
Reply
#4
(09-06-2020, 03:21 PM)shagghie Wrote:
(09-06-2020, 11:49 AM)Jesse Johannesen Wrote:
(09-06-2020, 08:48 AM)shagghie Wrote: For Accents, assume just use value of 127 for velocity, but for SLIDES, am I missing something or does there need to be a way to overlap note lengths such that midi will send slide information to my synths?  In particular, looking to get an acid vibe going and can't do so without slides/accents being in their happy place.
If you know what the CC number is that your synth receives slide data on you can make a mod matrix routing that uses any motif pattern as an LFO shape and has that MIDI CC as a destination. Not a perfect solution but a little creativity with this method will provide tons of rewarding experimental results. I don't know for sure if setting the pattern length in the pattern affects it's length as an LFO or not, but I would mess with that and see if I were you. you can also edit the pattern while it's playing for even more interesting results.
Jesse

Interesting, good stuff thank you.  Doesn’t just overlapping notes result in a slide normally?
The note “off” of the first just needs to wait until the note “on” of the second note so there is an overlap.

I could see this being a simple feature request, just add a “Slide” setting for each step in the rhythm editor, just like the REST and TIE’s?
There's no such thing as a slide in MIDI. Slide is a control voltage function internal to the sound module. Theoretically, one could simulate slide with MIDI by carefully inserting a pitch bend between notes, but it would require some kind of calibration feature to match the synths pitch bend and tuning. Its too complicated to set up, so as far as I know nobody has ever implemented it in a MIDI device. I would love to hear if it has been!

It might be possible to get a synth's Glide to sound like a slide, but The NDLR motifs are strictly monophonic. Its not something a typical mono synth MIDI implementation would handle as far as I know. And a polysynth would just play both notes unless its configured for a unison mode. Assuming the synth will glide on overlapping notes, it would still take some futsing with the glide time to get it to work with the selected BPM and clock div.
Reply
#5
(09-11-2020, 11:45 AM)Darryl Wrote: There's no such thing as a slide in MIDI. Slide is a control voltage function internal to the sound module. Theoretically, one could simulate slide with MIDI by carefully inserting a pitch bend between notes, but it would require some kind of calibration feature to match the synths pitch bend and tuning. Its too complicated to set up, so as far as I know nobody has ever implemented it in a MIDI device. I would love to hear if it has been!

It might be possible to get a synth's Glide to sound like a slide, but The NDLR motifs are strictly monophonic. Its not something a typical mono synth MIDI implementation would handle as far as I know. And a polysynth would just play both notes unless its configured for a unison mode. Assuming the synth will glide on overlapping notes, it would still take some futsing with the glide time to get it to work with the selected BPM and clock div.

True, nothing in the MIDI spec specifically supports, or mentions, slides. But, many monosynths do implement an auto-glide based on how data is received. For example, my MAM FB-383 will make a slide if the second note's Note On comes before the first note's Note Off is issued. Since the NDLR's Note On follows a Note Off with only about a millisecond between them, the suggestion of swapping Note On/Note Off in time would make many monosynths do an auto-slide.

One possible implementation would be to have in the Rhythm Editor a scenario like this:
Note 1 Vel 120
Note 2 Vel 110
Note 3 Slide
Note 4 Vel 121

This would be played as:
Note 1 On Vel 120
Appropriate wait
Note 1 Off
Note 2 On Vel 110
Appropriate wait
Note 3 On Vel 110 *
Note 2 Off
Appropriate wait
Note 3 Off
Note 4 On Vel 121
Appropriate wait
Note 4 Off

*Note 3 On velocity is unimportant as the EG is still triggered from Note 2 at this time and is sounded in the decay of Note 2.
Reply
#6
Thank you for expressing that much better than I was trying to! Indeed nearly all the synths in my studio simply respond with a slide behavior with this overlap. If you look at, for example, the F-R Mobius sequencer, it even has an adjustible slide time knob, which varies both the CV rate of pitch bending as well as the midi note overlap timing. For an classic TB-303 style acid slide, a slide duration (aka Glide Duration) of 50% is perfect. I've even seen samplers respond to the same midi note overlap situation with a glide behavior added to the pitch of the samples, etc. For majority of creative baselines and indeed acid, having specific notes be able to slide and not others is "mission critical" to get things to bob and weave properly. A synth's portamento amount is too universal in this regard (across all the notes in a sequence/arp).

There has to be a way... because just coming at this from an artist perspective, the NDLR has a chance to be the single most important badass sequencer in the acid universe, but exactly only to the extent that slides and accents can be "assigned per step" on, say, the RYTHM wheel of the NDLR. This is the acid way... yes, but it is also the key to some of the best basslines and melodies in the world, too. IMHO, respectfully, and extremely lustfully, can we have this as a FR on the roadmap? I'm willing to shell out of pocket to help dedicate cycles towards this. The NDLR is that important to me and the acid/techno/trance community!

I'm going to paste a post from awhile back.. this rang a bell for me from the old MachineDrum days of working with some of the same limitations, but this poster might have an elegant way to achieve note specific legato on the NDLR too... leveraging two midi tracks targeting the same channel, in order to achieve step-specific legato. In a separate post he demonstrates this by getting his x0xb0x to do a proper acid line with the MachineDrum's midisequencer:


----8<----8<----8<----Sure. When using MD midi machines, if you use one track to one synth and set the note length of one step to be long enough to overlap the next note triggered, you would expect legato behavior. But, you do not get it. A note off message is always sent before the next note on. The MD sequencer is optimized for drum sequencing this, I guess, was not seen as a priority. I'd like a 2nd type of midi machine added with different features. I doubt that much development will come of the MD's OS though.

Legato is especially important to me, for synths with a glide/slide on legato setting or a retrigger LFO on legato. How a monosynth reacts also depends on if the note priority is high, low, or last.

So, to get legato you need two tracks. Set the midi machines of both tracks to the same channel your synth is receiving. Set the note length of the 1st track long enough that it overlaps a note trigger on the 2nd track. You now have the ability to program legato in your sequences. Cumbersome at 1st, but it does lead to things I would not have come up with on a different sequencer. If you save a different note number or velocity with each midi machine, rather than P-locks with one track, you eat up more tracks, but some very interesting things can emerge very fast.

Tips:
1. Use many tracks with midi machines on the same channel and use mutes to introduce slides and other variation. The tracks in addition to the main track can be very simple, but introduce a lot of variation.
2. You can add really interesting instant slides on the x0xb0x. The tracks on the MD with midi machines are triggered in order. If two notes are places on the same step on different tracks, the lower numbered track will send a message just before the higher numbered track. So you can have a slide come from anywhere, not necessarily from the same direction or distance as the previous note that was sounded.
3. Hold the original note long enough that its length surpasses the 2nd note and you will get legato back to the 1st note
----8<----8<----8<

via: http://www.elektron-users.com/index.php?...551#191551

post with sound demo of x0x: https://www.elektronauts.com/t/how-to-ma...nes/1076/8


Could this same technique work on the NDLR's current architecture?
Reply
#7
I can't speak to the feasibility of implementing that in the NDLR firmware, but my suggestion is that if the target synth has a portamento function that is MIDI CC controllable, you would be able to use a motif pattern as an lfo shape in the mod matrix to turn it up and down for different steps. If this would work it would be a much simpler solution. I'll see if I can try it out and get back to you as to the results.
Jesse

(09-13-2020, 11:22 AM)Jesse Johannesen Wrote: I can't speak to the feasibility of implementing that in the NDLR firmware, but my suggestion is that if the target synth has a portamento function that is MIDI CC controllable, you would be able to use a motif pattern as an lfo shape in the mod matrix to turn it up and down for different steps. If this would work it would be a much simpler solution. I'll see if I can try it out and get back to you as to the results.
Jesse
I actually ran into some problems testing my idea, (mod matrix mod range is buggy when targeting MIDI CCs it looks like) and will have to add a bug sighting to the list, and as such, I wasn't able to confirm my theory, but I did try using the idea of sequencing 1 synth with 2 sequences, and sure enough overlapping notes created slides, so that does appear to be one possible solution.
Jesse
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)