Call and response timing


I was in a 32 BPI jam yesterday and one guy threw in a famous riff that I happened to know how to play. Based on the one-interval delay of ninjam, as soon as I recognized the first few notes of the riff (his "call"), then within the same interval, I already started to play my "response" by playing my own variant of the riff.

Later I listened to the recording on ninbot.com and it sounded OK -- his "call" was followed within some seconds by my response.

In general we need to minimise the delay between "call" and "response" to have some semblance of musical cohesion, so I think the above is one valid strategy (immediately start to respond, within the same interval, when you hear a call, and continue listening to the call while you are playing your response, to adjust your response in real-time), especially for longer BPI jams. (The other strategy would be to keep the BPI somewhat lower so that there is less lag between call and response.)

It's a little unnatural to immediately begin to play your response on top of an ongoing call, but then again ninjam is a little unnatural so we have to adapt.

The above is written from the perspective of someone responding to a call.

On the other hand, from the perspective of someone making a call, I suppose the call should be more than just a one-time few-note call, but instead should be repeated (with embellishment/modification to keep it interesting) or extended to cover at least one complete interval, with the caller then waiting at least one, preferably two, complete intervals to listen for a response, to allow enough time for the responder and to accommodate the inherent interval delay in ninjam.

How do you guys handle call and response?

Krue wrote: "Problems arise

Krue wrote: "Problems arise when there are 2 or more instruments on the same job (playing chords or doing the bass or soloing)."

One solution to this would be a conductor who asks different persons to play at different times. Also see for reference pljones's old post here https://forum.cockos.com/showpost.php?p=2042668&postcount=2, where he says: "If each soloist waits for the backing band to start comping for one interval before they go off and then, after the solo, the comp carries on for an interval, you can keep things stable just about."

Going further, the conductor functionality could even be automated and assigned to a bot. I have some ideas along these lines. Does this idea seem appealing to ninjammers?

"Why play if no one listens"

"Why play if no one listens" is in the context of playing while there is already something to listen to, just so what we play can be heard in the next "reserved" segment like suggested InfiniteDrumLoop.

For example : John plays on seg. 1, during this time nobody hears anything until it starts playing at beginning of segment 2. George receives John's part and listens to it, then plays something (we're on seg. 2) that john will hear at beginning of segment 3. John heard nothing during most of segment 2 because George was listening. He can start playing at that point, but that will be played with George's part on segment 3, and so on.

If we're talking about different instruments with different roles, there's no problem to play at anytime, since everyone has a defined role. Problems arise when there are 2 or more instruments on the same job (playing chords or doing the bass or soloing). And unless everyone agrees beforehand on who will do what at what time, most of the time it will not work because of the example I gave with George and John. At best, they will play but not listen to what the others are doing.

> Why play if no one listens

> Why play if no one listens ?
I think that's really the key to it all. You're listening. If what you're playing makes sense and it's not breaking things, keep doing it. If what you find you're playing isn't fitting in, move on. But keep listening - maybe you find what you were playing got picked up and you can go back again.

@InfiniteDrumLoop:
ninjam is basically audio data transfer. OK, so Jamtaba can do other stuff over the same packet format and it could encode MIDI, but you'd only see it at the same time the audio packet that the MIDI triggered arrived. You'd get fun stuff happening if it didn't delay display until the start of the audio playback... It might display someone's MIDI fractionally after "1" of your interval and you'd wait until "1" of the following interval to hear their audio, whilst you'd just seen the display for someone who you were hearing, because there's arrived just before your "1". Worst case, the displays might not even be in the "right order", if one was very complex and the other very simple...

---
...But how do you play with people, for people. Playing fast around the drums is one thing. But to play with people for others, to listen to, that's something else. That's a whole other world. -- Tony Williams

Good observations Krue about

Good observations Krue about how starting with less can lead to more opportunities for creation, especially taking into account the interval delays.

I'm going to try to slowly wean myself off the drum machine and learn to play drums on the keyboard. But then I would have to change my username also. :-)

About your comment "musicians will often step on each other because they don't know or see what the others are doing" -- one idea I had was if ninjam could somehow display visually a musical score showing the notes of the about-to-be-received interval from other players. (This might be feasible, for example, if all participants used midi keyboards.) That could allow you to visually see what is happening musically with the other players, before you hear it played in the next interval. Sort of a "preview" of the upcoming interval.

Some of the best jams I had

Some of the best jams I had started with almost nothing. A simple, smooth synth chord held for the whole interval, then someone adds a few notes, leaving all the space for the others to add something discrete, one by one, until some kind of structure begins to shape without imposing itself. At that point one player may change his idea and it will not distort the structure.

When starting a jam with a definite beat/ chord sequence/ bass line, there is not much creation : it is mostly "replication" of know patterns, and the participating musicians fall back to their memory and experience to fill in the remaining space.

With the long delays ninjam doesn't leave much room for collaborative creation and musicians will often step on each other because they don't know or see what the others are doing. If we stop and wait for an empty segment to start playing, it means the others will hear 2 empty segments before hearing someone playing. And if we play while we hear someone playing, even knowing that the next segment was already set to be our segment, then it means we are not listening to what is playing at the same time. Why play if no one listens ?

pljones wrote: "Longer

pljones wrote: "Longer intervals just mean more patience. The rule tends to be the same, wait for ideas to permeate - four to six intervals at least before you decide no one's taking notice."

Question to you all: Do you consciously find yourself waiting four to six intervals to see if your ideas are permeating? During those four to six intervals while you are "waiting", what are you doing? Playing variants of the same idea? Playing generic accompaniment? Playing totally new musical ideas unrelated to your original idea? Or just playing nothing and listening?

My musical skills/understanding/vocabulary are still too weak to be able to intuitively identify and recognize musical ideas as they develop. Once in a blue moon I will toss something out and others will follow, but on the other hand whenever I try to follow the flow or structure of some musical statements, I often can't identify the effect of my contribution on the overall progression of the piece. That's also why I like shorter-BPI jams, because they require less short-term musical memory in order to gauge how quickly musical ideas (whether mine, or someone else's) are being propagated. With longer-BPI jams, each interval itself has so much structure, and then if you have to wait 4 or 5 intervals (each with rich structure) to see if some idea from 4 or 5 intervals ago has caught on, that starts to overburden my musical memory.

I think atmospheric/ambient jams (not so melodic and not rigidly adhering to the metronome) might be a better way of learning how to work with ninjam and how to propagate musical ideas within the restriction of the interval. However, it seems there aren't so many ambient jams lately, at least not when I'm on.

OK, to sum it all up, maybe this is the way to go:

1. Blissfully ignore the interval and just pretend you are playing in real-time with the others, even though you know that what you hear and what others hear is different.

2. Develop a musical vocabulary rich enough to identify specific musical ideas, statements, or structures with in the jam. Those statements might originate from oneself, or they might originate from others. The important thing is to identify, track, and where possible, build upon those ideas. Develop the ability (i.e. the short-term musical memory) to remember and track musical ideas over several ninjam intervals.

3. Assume and hope that others are doing the same thing as you (i.e. #2).

4. Ambient jams could be simpler than melodic jams for practicing idea propagation within ninjam, because "structure" in ambient jams could be simpler to identify and create.

I'm not a professional musician. Maybe the skills I mention in #2 above are just the same skills that would be needed in a face-to-face jam. Even with a face-to-face jam you need good short-term musical memory to identify what has happened, what is happening, and what could happen with the musical flow. It's just that with ninjam the propagation is different.

The idea of "predicting" is

The idea of "predicting" is fraught with problems - mostly because there really can be different multiples of the interval delay (well, one or two - I've never heard of three). So either some end up feeling they're behind (and are prone to trying to pull ahead to "match") or some feel ahead (and try to drop back to "match") and then the cycle repeats as everything randomly changed a bit... It generally just gets messy.

(You end up with my end of phrase drum roll and crash coming just as you'd finished thinking you were in the middle of something, so subconsciously recalculate, then I recalculate again four intervals later because I lost track and tried to get back, and then you do... etc... And anyone else in the room just starts to feel chaos is ensuing...)

But, if the chords don't change too much between intervals and you're lucky that everyone's one interval apart, it can work like you say.

Longer intervals just mean more patience. The rule tends to be the same, wait for ideas to permeate - four to six intervals at least before you decide no one's taking notice. But give up straight away if you hear something you want to follow and start following it - again, waiting to hear if everyone else catches onto the building idea or there's something else everyone else seems to favour.

(I do try to do that myself, honest...)

Thanks pljones. After doing

Thanks pljones. After doing some more reading about "call and response" in music, I think that I was misusing the term "call and response". I see that your comment is mostly referring to a call-and-response as a single structure that, if I understand correctly, generally requires a call to be immediately followed (in the same or the next bar) by a response. Something like:

Call: "Hey man!" Response: "Hoe man!"
Call: "Yay man!" Response: "Yo man!"
Call: "Say, man!" Response: "So, man?"

In all of the above kinds of call-and-response, it sounds natural only if the response comes immediately following the call, in the next bar or even in the same bar. It could sound unnatural for these patterns to be split up into something like, "Hey man! (one or more intervening 4-beat-bars with other musical ideas) Hoe man! (one or more intervening 4-beat bars with other musical ideas)". The intervening bars could undesirably weaken the relationship between call and response. I think this is what you are referring to when you say you treat call-and-response as part of the musical structure within one ninjam interval -- the call and response are tightly bound, close together, and hence form a single structure inside of one ninjam interval.

On the other hand, what I (incorrectly) referred to as "call and response" was more along the lines of simply "responding" to some new or interesting musical statement/idea that someone else is playing. And my observation was that to ensure that your response is timely, one way is to start playing your response while you are still listening to the original musical statement/idea, because with ninjam what you are hearing is always at least one interval old, so the original player who made the first musical statement has already finished that statement and has been "waiting" for one interval to hear any response.

So generally, if you want to respond to what others are playing, it seems to me that you have to respond more quickly in ninjam than you would in a live face-to-face session, because the instant you hear something worth responding to, it is already at least one interval old at that instant and the jam could be flowing in another direction, so you have to inject your input as quickly as possible (especially in long BPI jams) to ensure it gets heard in a timely fashion.

Probably you ninjam veterans have already internalized how to work with the interval, but I'm still learning as I go!

Thanks again for your feedback pljones.

If you've got 32 beats,

If you've got 32 beats, you've got 8 x 4 beats, which is 2 x 4 x 4 beats - i.e. two lots of 4 bars in 4/4. I tend to try to treat it as call/response within that - so you're all playing the first four bars "call" sequence and the second four bars "response" sequence. There's no clash that way.

Of course, if you want a longer form - say eight bars of 4/4 for the call, eight bars of 4/4 for the response - you either have to lump it and hope you don't lose track or extend the interval to fit the form. I generally go with the hope that the chords don't clash too badly (and, more often than not, they don't).



    © 2013 ninbot.com All rights reserved