Loops. Ever since sampling workstations such as Akai’s MPC series have functioned as enablers, they’re omnipresent in contemporary dance, pop and hip-hop music. Taking a recording of a drum or rhythm section groove, preferably from a formerly unknown old R&B or fusion jazz record, and then repeating that ad infinitum.
This – loops with a duration of one, at max two bars, and a length of a few seconds – is not the kind of loops we’re talking about here.
This awkward hand drawing shows the logo for Erlanger Programm, and it implicitly also shows the loops contained here. With the starting point at center on the far left, we progress through all the parts along the infinity symbol’s line, taking detours for the circle-like paths, and end up where we started after completing the whole sequence.
So what can be looped here? First of all, the whole sequence, the album as a whole: titles a, A, b, B, c, C, d, D, a’ to start over with a again. This is, if we think CD album, the player set to “Repeat All”.
But there’s also the four circle-like thingies, each one representing one of the main parts, A, B, C and D. To take the CD player analogy again, looping them would be like “Repeat One”.
So what does that mean from a musical standpoint? And from an audio engineering one?
This is the full sequence. For this, it shall work both if you only play it once, and also if you play the full sequence over and over again.
Musically, it means that a has a proper beginning, and a’ has a proper end. They are both in the same key, and in the same tempo and meter.
There is a considerable pause between the last notes of a’ and this track’s end. This is, however, a space where some delays and a longish reverb decay. This means that those either must fade to -inf before the track’s end (slowly, or apruptly before that on a beat), or anything that trails on after that must then continue on to a’ – if that fits sonically.
The choice between those three options (fade to black, clean cut and short period of silence or wraparound) is something I am still considering. Only the third one would have an effect on a – but this part is arranged in a fashion that it would work with each of the choices for a’.
The four Microloops
Micro is relative here, of course – those loops are each about seven minutes in length.
The situation here is radically different than that of the macroloop: for the macroloop, the eternal golden braid can be seen as a finite structure, of which the endpoints get glued together to form the loop. In case of the microloops, the braid gets cut out of the larger structure, and then has its ends glued together for the loop.
That means that both ends (or rather the beginning and the end) of the loop don’t continue into silence when not looped, but rather into the surrounding minor parts.
Musically, and also sonically, it means that the beginning must both continue the preceding minor part and the end of the loop. And of course, the end must both move into the beginning of the loop and the following minor part.
That last sentence was set in bold letters, because it’s really the important statement here to which we’ll come back a lot.
Groove: Lots. None. Or a little.
There’s differences in the four microloops, of course. There’s B and D, which are both distinctively groove-based tracks with a kind of rhythm section. This means parts b and d need to end in a way that you can jump right into that very groove – make it a drum fill, some kind of intro, whatever. The same is true for the end of these major parts – obviously, not with an intro. Of course, the most simple solution would be to have the groove run straight through the part, and then have the following minor part somehow end the track. Of course, those long tracks won’t just have a silly beat running through, but in the case of D, there’ll be the same groove at the beginning and the end. Putting that into the context of d and a’, I’m currently seeing a drum roll coming from d, and a’ continuing the harmonic/melodic concept of D without beat.
The approach for B is similar, however, to be able to move into the sequencer figure for part c properly, it needs to be introduced at the end of B – which means it needs to bear relevance at the beginning as well. While I first saw that as an issue, it turned out to be an opportunity, as it works well as an additional theme to go with (or rather against) the main one of this track.
Flexibility is much greater in the case of A, as the beat is less pronounced here. What’s more, the big orchestral bass drum hit opening part b goes really well with a timpani hit (the drums for part A), so A can be ended nicely with some timpani fill that either leads into the theme starting A, or the single forte hit starting part b. And this goes well either way with the end of a, which has the timpani coming in with a slow crescendo.
Which only leaves C. Noise. Nerdville. As this is a very slow-moving one, with the surrounding parts c (sequencer madness) and d (granular rain and trombone) rather different, and so there’s only two easy ways to do this properly. One: have a silence at the beginning (easy) and at the end (might be musically challenging in all that slowly-fading context). The other: have part C leak into the surrounding minor parts and set the track markers accordingly. This in turn has a rather tricky impact on the sound design for C with its long reverb tails; essentially, the reverb tails for the instruments at C’s end need to be recorded and then faded in at the end of c, to fade out at the beginning of C. This decision (and the realization of the ensuing concept) are still work in progress.
A World Without Loops?
I’d like to come back to that statement I so proudly printed in bold above:
The beginning must both continue the preceding minor part and the end of the loop. And of course, the end must both move into the beginning of the loop and the following minor part.
So the preceding minor part moves into the beginning, the end moves into the beginning, and the end moves into the following minor part. Does that mean we can cut out all major parts and have a working sequence of a-b-c-d-a’?
Right from the standpoint of groove, harmony and reverb tails, yes. However not musically, at least not as a general rule. Why is that?
Remember the statement about drum fills at the end of minor parts, and short endings moved to the beginning of other minor parts? The result here would simply be some buildups, crescendi and drum fills, moving into nothing else than a groovy track’s last bar. For that to work, the start and end points would have to be set differently – which would jeopardize the loops.
I can live with that – after all, that “World Without Loops” concept wasn’t among the things I had planned for this album. But maybe…an idea for a future project?