I have read several Event (and Audya and Lionstracs, so I’m not just piling on the Event!) reviews that mention a bit glitchy chord transitions, and I think that I have some thoughts about the matter…

Conventional MIDI based arrangers have some very clever tricks to deal with slightly late chord input. If you look at the MIDI output of a capture in a DAW, you’ll see the occasional weird codes in between a note and the next note that is only a few ticks later. What is happening here is, you either played the next chord slightly late (or even bang on time, because it takes the arranger’s chord recognition some milliseconds to figure out what it is) and the arranger is forced to change the notes from the previous chord to the new one BEFORE the next note of the pattern comes in.

How it does this is usually a portamento of zero time to the new note. This changes the note’s pitch, but most importantly, doesn’t trigger a new note. So you don’t hear much of a glitch, far less than if it retriggered the note (which is what early arrangers sometimes did).

But audio loops can’t do this. They would have to crosscut to the new chord, which may be very different in voicing, timing, timbre, whatever. And this is going to lead to more noticeable glitches.

And here’s the rub. Players that primarily play arrangers sort of get used to ‘leading’ the changes very slightly (under a 16th or so depending on tempo) so the next chord is recognized BEFORE the loop starts. But if you play keyboards and pianos conventionally, you work your whole life on both hands being perfectly in time. Coming in ‘early’ is rushing, highly frowned on!

So it appears that to get the best out of an audio loop arranger, the ‘better’ you play, the more likely you are to get noticeable glitches, and this might be leading some reviewers to comment about hearing them more than they do with conventional arrangers. And it takes a long time to train yourself to rush slightly with the left hand without rushing with the right. I think this may be the reason I hear so many user demos that push pretty horribly, it’s just someone rushing the right to match the left…

The thing is, with great timing in both hands, it’s less likely to HEAR your arranger struggle to change chords fast enough with a MIDI arranger. If you go between playing conventionally (say with an SMF or audio backing) and arranger play, I find it quite difficult to adjust my left hand from being on time to rushing a bit. I guess if you use arranger mode for most of it, you get used to it though.

So, I guess my question is… if you have played an audio loop arranger NOT doing drums alone (Audya, Lionstracs, Event, any more out there?), did you find yourself having to adjust your timing, or did you hear more glitches than your MIDI arrangers? And if you’re one of the few with an Event, what sort of window does it take for the chord to be recognized and the correct loop to start streaming..?

The other side of this is also, what happens if there’s an anticipation chord in a spot where the style doesn’t actually have one? How well does the audio streaming change the chord in an area that the recording didn’t?

A MIDI style has the same exact rhythm for all same type chords, all that changes are the notes. There might be a different pattern for major and minor etc., but anticipate with a similar chord type (or even a different one if the style doesn’t change that much) and you won’t get rhythm glitches, or at least they’ll be mitigated by the same portamento codes that help at bar boundaries too.

An audio track is going to rely on the studio player being inhumanly accurate to duplicate the same pattern in all keys. To the point where jumping from one file to the next may introduce a glitch you won’t hear on a conventional arranger.

Audio loops SEEM like a great idea, but in the hands of less accurate players may exhibit problems you don’t come across so much with ‘old fashioned’ arrangers… 🎹