Implosion - onsets to offsets

[michalszostalo]

There is one thing that has always puzzled me in OM (ok, actually, not just one )

Let’s say I have a chord-seq with all of the individual notes as chords. And I would like to turn them into offsets, but in as many different ways as possible, as a tool for making variations.

This is possible manually of course, but there is the puzzling fact that the offsets go backwards from the last note selected, which is not very useful for what I’m trying to do. (I would love to know what the rationale is for that - there’s almost always a good reason for things like this in OM, that allows me to do something I never thought of before).

I’ve gotten better at processing lists over the years, so I suppose it might not be a big deal to reverse them. Alternately, one could group the lists of onsets accordingly. I’ve spent a considerable amount of time trying to find a practical and accurate way to do this, and it always came out wrong. There are some new tricks I’ve learned lately with lambda patches, that I’ll be trying out this afternoon, but if anyone has experience with this and I’m missing something simple and obvious, I’d be very grateful for a tip.

best,
Michal

[haddad]

Dear Michal,

Your description is rather puzzling in itself. Can you please describe what you want to do “musically” , (an example would be welcome).

• I don’t understand what you mean with " individual notes as chords" and their offsets. Usually offsets where designed in om for chords having multiple notes (at least two).
  Maybe if you reformulate your problem music wise and not starting with OM’s object, you can achieve it in more straightforward and efficient way.

best
K

[michalszostalo]

Hi Karim, I’ll try to be clearer. I’ve “exploded” the musical fragment into individual notes, so that now, I can, at liberty, group them as I like. (You can do this manually in the editor with the shift* command.) This way I can group notes at the level of, say, a beat, or a measure, or whatever. Once I’ve done that, I can insert new temporal values, or submit them to other processes, like permutation. Of course it’s possible to do similar things with onsets rather than offsets, but it seems to me that the offsets were designed specifically to be able to do this - so you can manipulate the temporal foreground without messing up the harmonic background. The object I want to work with wasn’t created in OM at all, I’m importing it specifically so I can make variations. Of course I could do this with a pencil and paper, but… it would take a lot more time and a lot of the processes I have in mind would be nearly impossible.

So - here is the problem: the shift* command ties the note to the onset of the LAST note in the group, rather than the first. This means that the offsets are negative values relative to the last note. I have tried numerous things to fix it, and have come pretty close, but something always gets messed up. Particularly if the groupings aren’t simple measure-based groupings. Additionally, there’s the problem that it makes it more logical to group the notes to the NEXT downbeat, rather than the one before them, but that’s not usually what I want to do.

Of course, I can also manually insert parentheses in the lists, but that takes so much time that I’ve forgotten what I wanted to do by the time I finish, and even then, weirdness usually slips in at the x->dx stage.

There is a tool in OMTristan that lets you grab fragments from a chord-seq, and I can put it in an omloop and implode them, but so far that hasn’t worked very well either.

What would probably be easiest (for me) would be if the shift* command tied the offsets to the first onset in the group, rather than the last one. But I suspect that might require a lot of programming. So I wonder if there’s a simpler way. Thanks for your time.

best,
Michal

[haddad]

Dear Michal,

Now i get you. The problem here is even if you group your chords, they will still sound as ungrouped. Therefore i suggest using a better strategy based on onsets, which preserves the rhythmic content of the note sequence. Using the offset shift* trick will only group the notes graphically which in my opinion will lead to in-coherency and make your object unreliable. But i could be wrong.

Now if you need to inverse group use: ctrl+shift* (linux - on mac it could be cmd+shift*) (i.e last note to first) , this worked for me.

Best
K

[michalszostalo]

Thanks for your quick reply. I tried ctrl-shift and it still tied the groups to the last note.

I’m note sure what you mean by incoherence. I frequently use this method if I’ve created chords in om, arrange them in a sequence, and then use one offset pattern for the whole thing. Or several. The spiral permutation is a particularly good one for making an evolving but coherent offset pattern. However, those are always positive values tied to the first note.

Here is an example I’ve tried to work with an couldn’t get it to work. To make matters worse, it’s in triplets, so you get downbeats at 997, 998, etc.