Subject: MAX Digest - 19 Apr 1999 to 20 Apr 1999 (#1999-121)
Date: Wed, 21 Apr 1999 00:00:23 -0400
From: Automatic digest processor 
Reply-To: MAX - Interactive Music/Multimedia Standard Environments
To: Recipients of MAX digests 

There are 8 messages totalling 330 lines in this issue.

Topics of the day:

  1. midi and blue G3/ breakfast
  2. Sensor question (3)
  3. floats in msp
  4. Max Performance
  5. where max messages index & glossary?
  6. where am I?


Date:    Tue, 20 Apr 1999 00:48:58 -0800
From:    Peter Nyboer 
Subject: midi and blue G3/ breakfast


makes a "doo-dad" for about US$40 that lets you use the blue g3's internal
modem port as a serial bus.  This allows you to use your trusty old Studio
4 :) with yer spanking blu g3, and not have to buy a USB Midi device,
assuming you can even get one.   I just hooked one up to a Studio 5 running
oms 2.4.6(?) and there was no problem.  OK, I relent, I only had one
device, a Peavey 1600 sending control changes, so I didn't exactly put it
through rigorous tests, but I can say the "doo-dad" actually pushes MIDI
through it.

>Also, anyone know where Berkley is? :) Seeing as I have never been to
>California, my understanding of the geography is somewhat lacking.

The most important thing to learn about "East Bay" geography is that
Berkley is near Oakland, the Land of Breakfast.  The royal title of Ruler
of This Land is Lois the Pie Queen.  There, one can order the Reggie
Jackson special, named after a sports hero of yesterday--2 pork chops,
potatoes, eggs, toast, and perhaps even a fruit cup.
Lois the Pie Queen.
851 60th St, Oakland
(510) 658-5616
If you are bound to public transit, you can take BART to the Rockridge
station and eat breakfast at the Rockridge Cafe.  I recommend the pancakes.
And eggs.  You can't go wrong with pancakes and eggs.


Peter "Is BBQ off-topic?" Nyboer

Peter Nyboer
"Now, I not some guru or Dog or anything"


Date:    Tue, 20 Apr 1999 09:46:41 +0200
From:    Michael Wieser 
Subject: Re: Sensor question


I built an analog videodecoder (PAL) 2 years ago, which put down the
monochrome PAL  resolution to 20collumns/15 rows. Form my experience with
this unit it should be possible to do it with video, but your uC should be
able to handle a huge and fast datastream in a short time (a Pic wont do
it...) (My Unit sent this 300*8bit and 9bit Adressinfo within 20mS to the
screen ~1Mbit/s)
I would suggest some of this 16/32uC which can be found by Hitachi, Siemens
or others.
I don`t think, that you will need much hardware (some RAM and Flash and a
A/D Converter)

This low resolution is well enough to make some calculations about the
mouths size... (with low Tech) and send this Mouth-Size via MIDI...
The only thing is- your "target" should stay in the area of the camera.....
and its low tech, no 18,5bit resolution and Megasomething....

If you have some high speed PCs I think you can do this with high tech too,

>I am working on a piece involving a Basic Stamp and various sensors with
Max. I am
>seeking ideas about ways to capture vocal data by means other than
microphone. Does
>anyone know if there exists any type of sensor (sonor?) that can read the
size of a
>mouth cavity, for instance (e.g. as it changes by articulating different
>sounds)? Any other ideas?
>Bob Gluck

Michael Wieser

Service and Audiodesign


Date:    Tue, 20 Apr 1999 13:31:27 +0200
From:    Peter Castine 
Subject: Re: floats in msp

On around 19-4-99 23:11, Michael McNicholas said something like:

>I am wondering how long a float is in MSP (been assuming it is  32 bits)
>and (excuse my inexperience) how many bits are for the exponent (asumming
>4). If people have any info on the floats in MSP or any suggestions, I
>would appreciate any tips.

David Z. is the authoratitive source for an answer, but presumably he's
using whatever Metrowerks Code Warrior uses for the C++ double keyword on

Making assumptions about how data are stored is generally bad programming
Kharma, even if it can speed up the implementation of an algorithm.

If I may, I would suggest starting without making any assumptions about
bit layout; do your conversions with multiplication, round by adding 0.5
and then assign to a long int. The C++ compiler will then instigate any
necessary bit mangling (typecasting a double to a long is a slightly
different animal from typecasting struct pointers... unless you're
typecasting a pointer to a double to a pointer... that is

   double x = 3.14159265;
   long   l = (long) x;            // l == 3, explicit typecast not
                                   // actually necessary
   long   m = *((long *) &x);      // dunno what m is after assignment,
                                   // but it ain't likely to be 3!


Only if it then turns out that your externals are suffering from serious
performance problems would I then try to tweak the manual mangling of

----------------- -----------------
Dr. Peter Castine          | I am very pleased to announce that the
4-15 Music & Technology    | 26th International Computer Music Conference
                           | will take place in Berlin in the year 2000.
                           | We look forward to seeing you here!


Date:    Tue, 20 Apr 1999 09:00:51 -0400
From:    Gordon Adams 
Subject: Re: Sensor question

>I am working on a piece involving a Basic Stamp and various sensors with
>Max. I am
>seeking ideas about ways to capture vocal data by means other than
>microphone. Does
>anyone know if there exists any type of sensor (sonor?) that can read the
>size of a
>mouth cavity, for instance (e.g. as it changes by articulating different
>sounds)? Any other ideas?

Hi -- I missed the beginning of this thread, but thought I would chime in:
I have used the SOUNDBEAM (which is a sonar-based system) with physically
handicapped kids.  It can indeed be set at a high enough sensitivity that
it can  respond to the mouth cavity -- if I remember correctly, the low end
of its range is about 15cm (adjustable to up to 8m), at which it can send
7-bit continuous controller data.  Be forewarned, though, that this is only
gives you a reflection of distance, not volume, and won't distinguish
between teeth, tongue, etc.

If you're interested in evaluating specifically vocal (as opposed to
general oral) data, the easiest and most flexible way to do this might be
by building a sort of MIDI vocoder in MSP -- that is, a multi-band envelope
follower.  (I believe this is possible in MSP, yes?) Add pitch following,
and you've got a fairly comprehensive data set.  With the return of the
commercially available (now usually digital) vocoder, there may already be
some sort of standalone hardware.

Or -- here ya go -- if you did want specifically oral data, how about this:
combine your MIDI vocoder with some variation of the old Heil VoxBox
shooting white noise into the victi- er, subject's mouth?  That could yield
very interesting shape and position data...

Of course, vocoding would involve a microphone in the chain, even if the
end product yielded no audio signal as such.  Is there a reason to avoid
mics specifically?

One other thing which comes to mind are the breath controllers once sold by
Yamaha and such -- I'm sure that these are still reasonably available, and
could provide a degree of gas volume information (think breathalyzer).

Hope this helps,
Gordon Adams.


Date:    Tue, 20 Apr 1999 09:54:28 -0400
From:    David Bianciardi 
Subject: Re: Sensor question

On 4/20/99 12:00 AM, Automatic digest processor wrote:

>Re: Sensor question

this is a tough one!  let me preface this by saying that i know very
little about the way sounds are formed by humans (?!).  i sense it has to
do with the volume and shape of the mouth cavity, the sinuses, the chest,
and vibration from the vocal cords, and god only knows what else.
however, the suggestions below are based on assumptions.

my first reaction was to think of a way to measure the volume of the
mouth cavity.  this is made difficult by the fact that it's an open
system.  were it "closed" we could start with a known gas volume and
easily measure the changing pressure inside, and get mouth cavity size
from that.  next idea was to measure the size of the mouth opening
(lips).  some sort of resistive sensor, perhaps an elastic filament that
could be glued outside the lips, or measuring the angle of the jaw?.  but
that might be uncomfortable, or impractical.  the VNS idea wouldn't be
bad, and the head movements could be compensated for by attaching the
camera on a head mounted lightweight boom.  again, uncomfortable,

anyhow, before i give up, why not a microphone?  you could do some
analysis of the signal and derive info off the harmonic components, and
extrapolate from there, no?

perhaps you can tell us more about the application.  otherwise i fear
i'll just keep coming up with silly ideas ;)

David Bianciardi

212.353.3947 fax
IDRC || 415 Lafayette St || NYC, NY 10003-7000


Date:    Tue, 20 Apr 1999 09:53:40 -0400
From:    Don Malone 
Subject: Max Performance

please forward to anyone who might be interested

Ishtar Records is releasing *Musing 4.0*
with a release performance
April 20th - 9pm
at the Nervous Center
4612 N Lincoln
Chicago, IL

*Musing 4.0* is a collection of unedited,
unoverdubbed real time exerpts from free improv
Don Malone performs with his own software *ALL MAX ALL THE TIME*
joining him on the album and this release performance are:
Chicago Artists
Rob Parton, jazz trumpeter & leader of the Ropa Big Band
Michael Zerang, new music percussionist & composer
Fred Lonberg-Holm, new music cellist & composer
Robbie Hunsingwe, new music double reedist & composer
Anna Brown, performance artist & poet
Emerging Chicago Artists
Adam Sonderberg, new music guitarist & composer
Sam Dellaria, electronicist & composer
and long time musical partner
Elise Kermani, New York performance artist & composer

Don Malone is
"...the Grand Ol' Man of Chicago free improv." Kyle Gann - Village Voice
"Genuinely Inspiring" Laurie Spiegle - Computer Music Pioneer
his music will
"...add a sense of displacement to your day" Jeff Olson - Screaming Popeye

happy tunes

downloads & events

it takes all of us


Date:    Tue, 20 Apr 1999 22:52:42 +0200
From:    Roland Cahen & Ruth Sefton-Green 
Subject: where max messages index & glossary?

; dsp driver playthrough 1  *****

Dear David Zicarelli,

Where can we find all these messages together ?
I can't remember where I put few of these !
Like a mostly capital one :
what is the message for setting the sound manager options from the dac ?


Date:    Tue, 20 Apr 1999 16:51:19 -0600
From:    Kevin Walker 
Subject: where am I?

I'm trying to write an external with local scoping rules (sort of like
_send and _recieve, but with names shared only with a patcher).  How can an
object, at creation time, tell where it's at (i.e. which patcher or
subpatcher it's contained in)?  Is there a pointer to a window or patcher
structure accessible through the symbol table (gensym("#P"), or something
like that)?  Is there another strategy for acheiving local scoping?



End of MAX Digest - 19 Apr 1999 to 20 Apr 1999 (#1999-121)