10 comments

[ 3.0 ms ] story [ 38.0 ms ] thread
Thanks for sharing this - as a former jazz musician I find this problem space fascinating, and have thought a lot about it in free time. Here's [1] another paper I found a little more approachable, stating that ratios between 1:1 to 3.5:1 are found in recordings.

[1] https://www.ntnu.edu/documents/1001201110/1266017954/DAFx-15...

I’m just a listener, but also find this fascinating. I think the jazz is in some ways a revival of baroque performance traditions. Improvisation was hugely important in the baroque, and also they had their own version of swung rhythm; what was notated as a string of 8th notes was not played that way.
my sax teacher says that if you swing Bach you get jazz.
This was one I came across a while back, was very useful: Jazz drummers’ swing ratio in relation to tempo. It seems far easier to understand than a few I checked out today including yours and the OP. And it sounds like the authors know what they're talking about, which is not the norm in papers about aspects of jazz.

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.416...

It looks at the swing ratios of great drummers, i.e. people who actually swing! Tony, Jack, Tain and..well ok, Adam Nussbaum on an Aebersold playalong is a bizarre choice. It has some good graphs though, especially Figure 1. Mean swing ratio as a function of tempo (left) and absolute duration of second note as a function of tempo (right) for the four drummers and all excerpts and Figure 2. Swing ratio for the four drummers as a function of tempo.

Anybody ever seen 'curve' modeling of music ? representing scales as surfaces and progressions as twists ?
No, but now I'm curious! Do you have somewhere I can learn more about this? Couldn't find anything from Googling
Was just asking, as a guy into CGI, geometry and funk/jazz (syncopated stuff) .. I couldn't stop but to think about progressions or modulation as nurbs curves accelerations when you place control knots too close.