I've been thinking (harder) about your comment that "triggering movement"
and triggering as one moves from position cell to position cell with your
system are "not quite the same thing". I made the point that you can
mimic this quite well with a VELOCITY THRESHOLD TRIGGER, and that with
sufficently accurate velocity signal integration you can even preserve the
cell-to-cell feeling for the performer.
I think however there is something to your comment, especially
as perceived by the performer, and that in my planned CMJ article
on Motion-To-Sense algorithms (from now on I'll just write MTS,
and newcomers can ask me), I should include these discrete cases,
not only because the technology is sometimes discrete in nature,
but because discrete systems have some different properties.
We can consider a continuous system, such as extremely accurate
position measurement (POLHEMUS EM etc.). Firstly think of continuously
mapping the fine position measurement (say height) to "continuous"
pitch (16384 modulation possibilites under MIDI). A performer won't
perceive boundaries with this system, and will have a strong
relationship to both position and changes in position (movement)
without any sense of boundaries, because the system is finer.
It is however, exactly the same as your system (from what I can tell),
it's just that the discretisation scales are different.
Now using the same very accurate position measurement we can implement a
coarser discrete system. Common music scales are discrete, so one could
divide the mapping IN SOFTWARE into piecewise bits (notes). This is then
like your cell structure, and boundaries are now obvious to the performer:
Each position is mapped to a note (or light colour or whatever).
In this sense your algorithm is a simple position mapping.
Every time one crosses a boundary ("motion") a new note is sounded.
In this sense your algorithm is a motion trigger.
We see then that MTS algorithms are best represented by considering the
mapping and the triggering separately. Each possible combination of a
mapping and a trigger gives rise to a different algorithm - requiring
different performance techniques. The combination - together with the
input signal type and output signal type - define the MTS algorithm and
suitable performance motions.
We see also that there is an infinite range of feelings between
very fine (no strong boundary feeling), and coarse (strong boundary
feeling), and one could control that coarseness with a know on a
control computer without ever changing the system.
When I first considered ways to achieve what I want (drumming by
dancing etc.) I thought hard about video methods and then discarded
the idea, since I didn't want to fight with zones. I now see, however,
that discrete systems (whether sofware or hardware) have some advantages
for performers, depending on the application.
There is a hole* in my VELOCITY THRESHOLD argument, namely that if you
move your hand around in a circle without ever leaving the cell (of your
system) there will be no trigger. The hole can be patched a bit by
considering the integrated velocity (including vector properties). The
point that you can use perfectly accurate velocity measurements (movement
measurements) without ever measuring position directly to achieve exactly
your triggering algorithm (including boundaries) remains valid.
regards
Darren
* Those of you who jumped on my correct observations about particles
always moving, as do standing dancers, would have spent their
time better noticing this obvious blooper !
*****************************************************************
Darren Kelly | | |
| \ 0 / |
DESY -MPY- (Deutsches Elektronen-Synchrotron) | * o |
Notkestr.85 | - D E S Y - |
22603 Hamburg | o * |
Germany | / O \ |
| | |
phone: +49-40-8998-4569 | |
fax: +49-40-8998-4305 <--Note change | |
e-mail: kelly@x4u2.desy.de | |
****************************************************************
Amandastr.40a, | HOME |
20357 Hamburg | zu Hause |
phone: +49-40-4322602 | The PLAYhouse |
*****************************************************************