Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Logic’s Drum sequencer probability (chance). Does it work as you’d expect?.
Coming from other sequencers like Zenbeats and Octachron, or the Digitakt, where I use probability extensively, I was surprised to see how bad the implementation is in logic. Makes it pretty much useless, unless I’m missing something…
Here’s a screenshot to try and explain it
- Chance is calculated/generated ONCE, when some chance value is changed. So the pattern will be exactly the same every time you hit play. This is not how probability should work. Useless for variation, randomness and ghost notes.
- Even worse, chance is only calculated and applied to the “main” sequence NOT to loops. See screenshot… top sequence is 2 bars long so it generates 2 chance variations, bottom sequence is 1 bar long so it generates just one variation (ehem) which stays throughout the whole loop.
This is terrible and disappointing. I’d rather they didn’t include this feature as it’s misleading. Probability or chance should be evaluated per instance, every time. That’s how it makes sense and works everywhere else.
Edit: changed the post title. I was looping a sequence, and that is not what you’d do for chance. You have to extend the drum pattern, not loop it. So it’s not “terrible and useless”
Comments
I do realise it's not helping much, mate, but your description reminded me of how Kia implemented the automatic stop/start function in an earlier generation of their cars. The function was there, it had a nice button and you could create theoretical conditions in which it actually worked. BUT in practice it just never did because it was crap. 🙂 However, this was enough for Kia to claim to be in compliance with European conditions of producing more environmentally friendly vehicles.
Apple seems to have implemented this probability function the same way 👍
Anyway, as you were.
Yeah, the ‘chance’ implementation is a bit confusing for me too.
But if you extend the region by just pulling the handle to the right (not copying or looping) it will work as expected.
Not 100% perfect imo because I believe it still plays the same within a cycle (like your 4 bars above).
Yes, actually l found the same here. Luckily, Octachron is also working quite well in Logic, I think. And we really spent quite a long time optimizing our randomize and mutation features.
But if anybody still has suggestions, what we could optimize, just tell us. We always try to improve things.
Best from Munich ✌️🙂
Tas
I don't understand. To me, Logic's implementation seems correct, while all others seem wrong.
How would it be in any way useful for a pattern to randomly change EVERY TIME IT IS PLAYED? That makes the composition completely unpredictable and thus probably useless in the majority of cases.
Only Logic's implementation makes sense -- you can experiment with the variations it generates, and when it spits out one that incidentally "sounds good", you keep it.
Or is there a fundamental misunderstanding on my part?
Maybe this is a documentation issue. I personally find the chance function confusing and have a feeling that it "works" sometimes and doesn't at others, though reading through this thread maybe I'm just not understanding how the behavior varies. I totally get where you want to use chance to create a fixed length phrase and then capture that in a determinative way. I also understand the use case where you really do want it to be re-calculated for every bar. Sounds like there are (maybe?) ways to achieve both in Logic, but it isn't really clear how from the pattern editor/timeline/clip launcher.
I do also think there's a difference between a "chance" function and a "mutation" function like you see in e.g. Bram Bos's sequencers. And as far as I'm aware, Logic doesn't really do "mutation" by itself.
I think Logic’s Convert Midi to Seq Pattern would be nice.. and more options for ‘melodic’ sequencing.. Thanks..
Edit.. I guess a simple ‘import midi’ can do the same..
The functionality is called “chance” applied to a note. That means “this note will have a chance of playing of xx%”. It’s not a sequence randomizer. It literally states that that note will play or note according to the chance factor, that’s every time the note plays. It’s how it works on other sequencers.
I do see your point, which if I understand correctly is “apply the chance and if I don’t like it I’ll do it again until I’m satisfied”. But that’s a different concept.
Of course it’s useful. Also fun and interesting!. It’s like playing with a real drummer. Do you know every single ghost note or little fill they’ll perform?. No. You live a little room for variety. You don’t apply chance to the kick on the beat or the off-beat snare. You use it for little ghost notes, decoration, fills, dynamic hihats…
BTW the implementation of different types of “when to play a note or not” on the Digitakt is the best and most complete I’ve seen.
Yes, that's how I understood it.
Yes, that's how I meant it, and yes, it's a different concept
I still don't think applying the "chance" anew every time a note is played makes any sense. It makes the composition completely unpredictable. What if you play a sequence and exactly this time, all "randomness" is exactly right and sounds good -- how do you "capture" this? I mean, this "randomness" appears to me to be a composition tool. But it's pretty useless (or hugely inefficient) if every variation can only be played ONCE because the next time it's played, it'll again have changed.
That's why the only way this makes sense to me is to have a button that applies randomness once, and you press that button until you're happy with how it sounds.
But maybe my brain is just wired differently from most "normal" people. Fair play!
To me this makes total sense. Logic even gives you a preview of how the randomness is applied in the MIDI region on the timeline, it will show which notes play when, as determined by the "chance" parameter.
And just to reiterate a point made earlier: you shouldn't use the looping function on a sequence, just extend the sequence to fit, that way you will get different randomness on each bar.
I see we have different point of views about what “note chance” means.
This is probability in zenbeats:
It’s a loop, and every time the sequence reaches those notes, probability is applied. Logic’s way, which you seem to prefer, is some kind of randomization applying the given parameters, much like the “dice” function in many apps to randomize some parameters. It’s not the same as chance of a note playing, which as the name implies is tied to the note, not a sequence.
Yes, I’ll do this. Extending a sequence is just like looping except it’s not, doesn’t make sense to me but it’s a good workaround.
But if ultimately I end up with a fixed, unvarying sequence, I might as well program the little chances myself.
It’s interesting how we find such different point of views. I never expected any discrepancy on this one!.
Cheers and proper chancing!
So what you’re all saying is that what’s wrong with chance is it only works sometimes? 🤔
I’ll see myself out
I have to agree with you on this. You don't want utter chaos and randomness every time you go back to listen to your mix. It should be random once, then repeatable after.
I don’t get the “chaos” argument. It’s intended for subtle variations, to add movement and a bit of diversity. When you’re playing with a human drummer do you fear utter chaos?.
Again, I think you’re confusing note probability with something else, it might be better, but it’s something else. I don’t think Elektron invented note probability, but they’re widely considered a modern standard in sequencing quality and it’s a huge part of their success, Drambo and many other apps are not hiding their inspiration. Something like “mutability “ in BramBros apps or Octachron can get more chaotic (although I personally enjoy it) but here it’s only applied to some notes and with a lot of control. Really makes a difference for a more human feel.
@SevenSystems, I hear you, were a pain in the ass. Although I’m 100% probability right on this one 😜
@tahiche : I don’t think it is a matter of a right way or wrong way to approach randomness/chance. One approach is real-time and the other more of a render time approach. Both approaches have their uses. Neither is inherently better than the other.
The advantage of the non-real-time randomization is fine-tuning. You use randomization to generate a drum score then can edit it and lock it in.
Logic’s approach makes sense to me for a DAW. It might not be your preferred way of working but it doesn’t strike me as terrible or useless .
I had the same confusion, like it wasn’t working. It turns out it only does the “probability calculation” once, when you change some value. BTW this makes the “try out different variations until you find you like” rather awkward and dodgy as you need to be changing the probability values for it to actually do something.
Mutation is different, it can also add or shift notes, the dice on Logic drum track is probably closest (although also static unlike Bram Bros’ which is actual mutation).
Probability just determines how likely the notes you choose are likely to trigger. For example you can fill a row of hihats, apply probability of 90% and only a few here and there will go quite, which is a very nice subtle effect. Or the oposite, a 20% chance on a couple of strategically placed off snares to mimic the ghost notes of the drumstick bouncing, which happens in real life.
As a side note, Zenbeats also has velocity randomization (yes, Drambo too), which I don’t think is common and I love. It really helps with hihats.
Yes, I accept that it might be preferable for some, I don’t think I was clear. I totally get it. The “useless” was maybe too much?. It just applied to my expectations.
I 100% see this for the dice functionality, not for note chance. Which IMO should always be real-time as it applies to the note, not a generated sequence. When you generate that sequence, there’s no probability any more.
I’d love to know if any other apps or hardware share the same note-probability implementation as Logic’s, I’ve only seen the kind I’m advocating for.
Internally, that is how Band in a Box, iReal and other similar apps tend to work. They generate some number of measures and apply the randomization when the measures are generated.
An an obscure composition/sequencing app a former client developed worked like that. You entered randomization parameters and told it how many measures to generate.
It is something that hardware wouldn’t tend to do.
We clearly need both, for different cases! In the timeline the randomise-once-on-creation approach makes sense… yet in Live Loops randomising each time would certainly be my wish if using for live backing.
I absolutely love how Logic handles chance. I hate every other implementation that is non deterministic. I want controlled randomness. I love that I can either drag out to extend to add more randomness or loop to "lock-in" my initial randomness. It's seriously amazing.
Some apps have a lock switch for when you want to lock the randomness into the pattern, do they not? I think replicant does for instance. Is this not the best of both worlds?
I am also in 100% agreement. Glad somebody finally said it!!!!
Some clarification on the initial post title…
I initially didn’t see the difference between looping a sequence (loop tool) or extending a sequence. Extending a sequence does produce chance, a rendered-static version, but indeed usable.
I was expecting the loop to render chance (like Zenbeats, for example), but it doesn’t, as @richardyot pointed out. So extending the sequence is the way.
I might prefer the other way (per note real-time chance) but it’s not broken and unusable. Changed the title accordingly.
@tahiche I answered that right away… https://forum.audiob.us/discussion/comment/1261028/#Comment_1261028
Guess I’m on ignore
It’s all cool. Many apps work the opposite to how I think they should too. I’ve been accustomed to the ‘Logic way’ as I’ve used it on the Mac for so long but I feel the same way (that everything is wrong
) whenever I try other DAWs.
I can definitely understand why people are frustrated with logic. All the major DAWs on desktop are equally opinionated and if you use and like one then having the ‘wrong’ one on iOS when it’s the only option can be infuriating.
Cubasis was always a non starter for me, partly because it is so cut down from the full fat version but mostly because it drives me mad.
Yes you did, sorry for that, credit extended to you. 🙌
Thanks for changing the title.
I've been thinking I was going mad trying to use the chance thing. At one point, I worked out that I could get the last kick on a 64 step sequence to not play if I set the chance to 75%, which seems sane - as in the first three passes it will play it, the fourth pass it will skip it. I was quite pleased with myself. I've never managed to get this to work again. Only last night I was trying for ages, every adjustment I made to the chance would randomly turn off one of the other kicks, never the one I was trying to target.