You may have noticed in the User Manual there is an extra “Unassigned” knob in the looper mode. I haven’t been able to decide what to do with this yet. I have a couple ideas, but I thought I’d open it up to the group.
Remember that it’s a four track asynchronous looper. It’s meant to be abstract, fluid, playful, and not sync’d.
Another important note is that I write the fades when the loop is recorded. This is convenient and easy on the processor because I only have to manage 1 envelope instead of 4. But sadly it makes random-access and loop trimming a little challenging.
Panning: Per-track panning. Pro: easy to implement. Con: Might be useless for people using the pedal in mono.
Retrigger probability: “Every time the loop hits its start point you get a weighted coin flip on whether it plays through that rotation or goes silent for that rotation.” Pro: easy to implement. Con: Might feel awkward if you overdub new material through the loop start/stop point.
Track Speed Scale: Changes the speed intervals: no scale/smooth, chromatic, octaves and fifths, fourths, etc. Pro: Easy to implement. Con: None that I can think of
Completely Rewrite Looper to be Four Track Sampler Instead of Looper: Scrap overdubbing completely. This way you only get 1 punch per loop. Controls would be: speed/direction, volume, loop length, randomness. Pro: Introduces granular aspect to looper and more synchronized/sequency options. Con: No more overdubbing. Need to switch/record to new track to add layer. You only get 4 layers instead of infinite layers.
It occurred to me that it might be fun to be able to sync the loops by ratio - i.e. make loops 2-4 a specific length with respect to the first loop. You could maybe have the unassigned knob select between “free/async,” 1:1, 2:1, 3:1, 1:2, 1:3, etc…
That way you could set it so that loop 2 plays three times for every one time through the first loop, and/or loop 3 plays through one time for every two times through the first loop, etc…
I am not a programmer though, and that might be too taxing/difficult.
I actually noticed that when I just read the manual! :).
I am with the @PeterBregman line of thought. As this seems to be a pretty experimental pedal, having the last knob activate some sort of “interaction” between the 4 loops would be awesome-sauce (and, I believe, pretty one-of-a-kind).
Otherwise, yeah, panning would be neat. I don’t think I would find a “retrigger probability” knob that helpful.
PS. This pedal looks absolutely amaze-balls! Thanks so much for designing it.
Loop length trim – turn knob all the way to the right, and then rotate it back to the left to decrease the length of the loop. Hit the touch button to lock in the new loop length time (and delete audio after the new loop end marker). I was thinking this would apply to all 4 tracks, but I like the @PeterBregman suggestion of having different loop lengths per channel.
I’m not sure how fine the trimming would need to be – I don’t think that it would have to be the entire 40 seconds as you could just trim off 5 seconds, save it, and trim off another 5 seconds for more exact trimming.
Turn the knob to identify the start marker of a given track. Turn it all the way to the left to initiate, and then be able to play scrub through the sample until find the desired starting marker. Hit Touch immediately after turning the knob to lock in the new starting marker of the track. Alternately, be able to play the track manually by turning the knob, like manually turning a turntable. I don’t know if it’s possible to do both 1 & 2 so that you can edit the start and end of the loop with the available control inputs, but that would be sick to essentially have a sample editor within the looper functionality.
Peter’s idea is great and panning seems like the simplest idea to implement.
I haven’t forgotten about this thread! These are all great ideas. I just need some time to consider the pros and cons of each and see if they’re possible with the limited resources left on Ribbons
Full clockwise would be off. Everything in total free time slip. You start moving it left and it starts resetting all of the loops at a shorter time span. Full CCW would be everything frozen at a single waveform.
I know this would break from the UI being every page is something is different, but it’s just a thought.
I think the unassigned knob could be great for global attack/decay when engaging/bypassing in looper mode so that with loops already recorded they could slowly fade in or out, but I love these other ideas I’m seeing!
here for the loop length trim i can already think of creative uses for it
or you could have a thing kind of like touch mode 1 that you can turn on/off or up /down to have the loop be more/less spliced randomly/granularly
The only trouble with this is that I write the envelopes/fades to memory when they’re recorded.
This makes it so I only need to manage 1 envelope. If I do a loop length trim I need to have 4 more envelopes, one for each track, which requires more processing power.
Adding a length trim also removes the fun of overdubbing through the loop start/end point, which results in amorphous loops where the end/stop point isn’t clear. A loop trim would require me to force a fade out/in at the start/end to avoid a loud thump at the wrap and that would put an end to overdubbing through the wrap point.
Granular stuff is hard with the looper because, like i said above, it requires 4 more envelopes. It also complicates overdubbing on the same loop. If we’re jumping around and playing back/reading random parts of the loop, where the heck is the write head?
One thing that COULD be done is:
Scrap overdubbing entirely. This way you only get 1 punch per loop. The controls could then be:
Speed and direction on the same knob
Volume of the loop
Slice length
Randomness
This would just be a straight rip off of Montreal Assembly’s CT5 mode II though, except longer loop times and four loops instead of one.
what about an offset knob? cuz then u could also play with the knob to do scrubbing effects…
also i like having overdubbing… the 4th knob could just be like a degrade knob? like over time the loop gets more and more effected by the tape side of things. or it could be pre/post fx record? so when its on pre and u overdub it keeps feeding back and getting more effected
This is an aspect of the looper that is definitely very appealing to me, on paper at least. I really love letting asynchronous loops run alongside one another, so I’d love to see this functionality remain.
One thought I’ve got on if this know were to be used for Pan is that it would be really great if the knob were assigning a modulation source to the Pan of the loop. Specifically I’m thinking about the “attenurandomizers” on Mutable Instruments’ Beads module, and I’m thinking about how Acrostic for Norns (which itself was inspired in this regard by oooooo) feature similar functionality or conceptualization in how the panning of the loops is handled.
One thing that I wish was possible, would be a way to “bounce” all of the tracks to one loop, and free up the other loops to go back and add more. They would all bounce in at their current playback volumes, speeds, directions.
Maybe I missed it, but I wish it was possible to “undo” a layer from a looper layer. Sometimes I have built a beautiful texture underneath and then hit one wrong note or time a note wrong. Having the ability to undo would prevent me from having to clear that track. I could also accomplish this with a workaround if you implement the bounce feature as described above.
Maybe this one is not possible but the unassigned knob - could this be turned into a DJ style hi/low pass filter? That way each loop could be individually tweaked. Exiting looper mode would allow you to do global effects.
Would be awesome if it was possible to add a fifth page that would be a DJ type page, putting low and high pass on the same page. Having them on two separate pages currently means I have to navigate if I want to tweak both of them. Maybe the other two knobs are Reverb Mix and Tape Mix?
@PeterBregman the redesign you mention I think already accomplishes this… If the goal was to not change the current implementation, a 5th through 8th page could be added as an alternate interaction method.
Clearing the loop isn’t super intuitive today. Maybe holding down both footswitches for 3 seconds could be a way to clear it? Otherwise I am almost forced to use my hands if I make a mistake in a live situation.
Moved your post to this topic, since it is more related to this subject than the UI framework changes in the works.
I think the current framework already has some “undo” capabilities. Instead of overdubbing on the same loop, you could jump to one of the other four tracks and record another loop. They might not be synchronized but if you goof something, you can just clear the track.
Keep in mind, Ribbons is not a looper. It’s a real-time effects processor that has a looper as an easter egg mode. In addition the looper was designed to be simple, amorphous, playful, and asynchronous. I wasn’t intending for it to be a workhorse looper with lots of bells and whistles – I wanted to keep it simple but have it still be fun and immediate.
Totally possible and a very cool idea! …but a little processing intensive. I have a couple of very nice virtual analog filters in my DSP toolkit but I would have to run 8 of them (four stereo tracks) to do what you’re asking, which might be too much. I’ll keep it in mind in-case I come up with some major optimizations that free up the processing power to do this.
The new UI framework with more “thematic” pages will be available on Friday when the v0.2 firmware is released. This will have the low pass and high pass on the same/front page along with mix and compression. I want to stick with having four pages for the tape effects and four tracks for the looper. I think any more would complicate the UI too much for some users.
I’m working on a new “vote for feature requests” forum topic. I’ll add this to that.
The current implementation allows you to latch recording with a quick tap or record by holding down the footswitch. I would have to get rid of the “hold to record” feature to implement the “hold to clear” feature. It would be a trade off but if more people are into it I could make it happen.
Definitely possible but I can’t think of any intuitive button combos to implement it. From the UI perspective, can you think of an intuitive way to go about bouncing between tracks?
My thought initially was:
from any of the looper pages that have a loop running do the following 2 step procedure:
Press and hold the Push button
While holding the Push button, press and hold both footswitches for 1 second
Letting go could then show the page LEDs cascading back to the first page and those loop slots would be cleared and freed up again.
This is the opposite/shift functionality for my proposed clear functionality - holding both footswitches in looper mode for 3 seconds to clear the loop.
I have no idea if this is possible and I defer to you on the complexity, as again, I know this is supposed to be more of an easter egg.
Others, please feel free to weigh in on the utility of this feature, whether you have an alternate implementation for the bounce concept, or otherwise!
I think some random-ness element would be cool. Perhaps it’s a probability that grabs a little sample from a different loop even? Almost like a glitch, or randomly jumps to a different point in the loop. Might be hard to assign all that with one knob, but infusion of “random” is always fun in my book.
I too think the pan function would be a great option. It’s not as sexy as some other options but I think makes it even more compelling in the studio context to grab different layers and nuances in stereo.
It could work where you hold the three buttons from the procedure I previously outlined for as long as you want (until maximum buffer limit is reached), allowing you to resample it as a smaller layer. I have no idea if this is technically possible, but this seemed like a cool creative way to make it happen or impose other artifacts on the loops altogether.
It could work where you search the audio for which one is longest, and then multiply the other shorter loops to match the exact length of the longest loop.
a) For example, I have a set of four loops. Loop A is 10 seconds, Loop B is 2.5 seconds, Loop C is 5 seconds, and Loop D is 1 second. Loop A is our longest loop, so Loop B is multiplied 4 times, Loop C is multiplied twice, and Loop D is multiplied 10 times. They would be printed based on their current alignment on the overall clock, so even if the pattern from Loop D was repeated 10 times, it wouldn’t necessarily have to align to the 1 second mark. Does that make sense? I don’t know if I am explaining this concept that well. I don’t know if it is technically possible.
