Subject: MAX Digest - 30 Mar 1998 to 31 Mar 1998
Date: Wed, 1 Apr 1998 00:00:25 -0500
From: Automatic digest processor 
Reply-To: MAX - interactive music/multimedia standard environments
To: Recipients of MAX digests 

There are 9 messages totalling 268 lines in this issue.

Topics of the day:

  1. Midi 50 feet
  2. CD-ROM drives for audio
  3. MAX Digest - 28 Mar 1998 to 29 Mar 1998
  4. QT3 Officially released
  5. Midi 50 Feet
  6. Overdrive, bend and more...
  7. Quicktime and bend
  8. follow that body
  9. follow the body


Date:    Tue, 31 Mar 1998 00:36:48 -0500
From:    Christopher Murtagh 
Subject: Re: Midi 50 feet

On Sun, 29 Mar 1998, Elliott Earls wrote:

> Hey,
> I need contoll my computer via a midi keyboard 50 feet away. can I send
> midi 50 feet on midid cables?

 Back in my rock 'n' roll days I used to send Midi from my onboard
sequencer on stage to the lighting console across the room without any
problems. I used a regular shielded twisted pair cable with a 5pin din on
one end. Sometimes I would run it down the snake with all the audio, but
in the clubs that had bad snakes (ie get really bad induced noise), I
would just run an XLR taped to the ceiling. These were quite often over
100 feet. This never really gave me any problems.

Christopher Murtagh         | snail mail: 2009 de Bullion       |             Montreal, Quebec
                            |             Canada
McGill University           |             H2X 2Z7
Faculty of Music            |
                            |      Phone: (514) 842-5872


Date:    Tue, 31 Mar 1998 08:08:49 +0200
From:    Jeffrey Burns 
Subject: CD-ROM drives for audio

Can anyone tell me if there is a SCSI CD-ROM drive with professional audio

Jeff Burns


Date:    Tue, 31 Mar 1998 12:09:35 +0200
From:    lu 
Subject: Re: MAX Digest - 28 Mar 1998 to 29 Mar 1998

>I need contoll my computer via a midi keyboard 50 feet away. can I send
>midi 50 feet on midid cables?
>DO I need a "repeater?" do they make a "repeater?"

no midi cables, no "repeater". some soldering will do: 2 canon plugs and a
thick cable, should carry at least 100 meters.



Date:    Tue, 31 Mar 1998 05:52:20 +0000
From:    Johnny DeKam 
Subject: QT3 Officially released

Quicktime 3.0 was quietly and officially released yesterday...

Get yours now! -- trash those beta copies...

The built in QT instruments is vastly improved from 2.5, and QT3 is fully
web streaming now
(both audio and video)

Johnny DeKam


Date:    Tue, 31 Mar 1998 14:06:27 +0200
From:    Jens Jacobsen 
Subject: Re: Midi 50 Feet

Elliott Earls wrote:

> I need contoll my computer via a midi keyboard 50 feet away. can I send
> midi 50 feet on midid cables?

The MIDI-specification says that midi-cables shall have a maximum length of
(or 15 meters). So you are right on the edge.
You have 3 possible solutions:
1) Use a 50 feet midi cable and cross your fingers. Allthough it is not
recommended, it is sometimes possible to use even longer cables. But in
surroundings (near lightdimmers and stuff like that) it's a problem.
2) You can put a "dummy" midi device in between your keyboard and your
just to "clean" and amplify the midisignal. It could be anything: a
whatever, just let the midi thru it.
3) I would recommend you to check out the Midi-Match device from Anatek. Its
small box, that lets you use balanced microphone cables instead of ordinary
midicables. In that way you can send midi at really long distances.

Best wishes

Jens Jacobsen

Danish Institute of Electroacoustic Music


Date:    Tue, 31 Mar 1998 10:49:49 -0800
From:    Peter Elsea 
Subject: Overdrive, bend and more...

About problems with G3-
I've found that the G3 can easily compute enough MIDI data fast enough to
crash OMS- apparently filling the output buffer, which causes a "too many
MIDI messges" error and the need to reboot. This is not the same as a stack
overflow, but perhaps there is a relationship? This will vary with the
interface- My fancy 64xtc is a lot touchier than a plain translator. But as
DZ said, stack overflows are almost always caused by loops, so try a trace.

About bends
>but now wondering...if my QT is handling 14bit-bend-date
> ear is not good enough to test this...ah..
>So sad if it's only handling it as the 7bit thing and ignoring the rest.

You can hear the effect of 7 bit pitch bend on most instruments by
extending the bend range as far as it will go. If the range is over an
octave, you will hear distinct steps in the bend. Unhappily, few
instruments use all 14 bits of the MIDI bend messge, and fewer generate it.
Some use 12 or 14 bits internally (from their own keyboard) but send 7.
I woud expect QT to use 14 bits, as it has nothing to gain by not- it's
living in a 32 bit world anyway. However,  Inside Mac-Sound, suggests that
bend is not supported at all- the note play routine accepts MIDI notes as
integers. This is an old book- anybody seen an update that contradicts this?

Long MIDI cables
>I need contoll my computer via a midi keyboard 50 feet away. can I send
>midi 50 feet on midid cables?

We do it here all the time (sending MIDI from room to room). The only trick
is to use a powered MIDI interface, and use only one of the outputs. (Some
of those 1 in 3 out jobs simply parallel the output ports!).

Tapping into data structures
>Is it possible to assign a value directly to a variable that is part of an
>external object's structure (from C, say). For example, if I want to assign
>a window pointer to the *b_wind field of the b_ob structure of the object
>bomb, is there some way I can do that?
Remember that abstractions like My_window->b_ob structure->(*b_wind) are
figments of the compiler's imagination. Once an object is complied, that
value is simply stored at some arbitrary address. There are ways to find
it, but it's really not a good idea. If you do find something like this,
you might get away with reading it, but trying to change it is an even
worse idea.

Peter Elsea
Electronic Music Studios
University of California, Santa Cruz


Date:    Tue, 31 Mar 1998 11:16:41 -0800
From:    Peter Elsea 
Subject: Quicktime and bend

Well, as soon as you send a message off, something turns up to make you
look silly....

I tried some bends with quicktime instruments, and it is pretty clear that
only 7 bits are supported.
First set bend range to 24 semitones in the OMS quicktime instruments
Using the Xbend help file, the slider moves quite some distance before the
pitch jumps.
The step sizes are the same as those produced by the bendout object.

If you want to do microtuning, msp is your best bet. (Actually there are
some really inexpensive instruments that do microtuning nicely- the Korg N5
for instance. You can just send it the scale you want played.)

Peter Elsea
Electronic Music Studios
University of California, Santa Cruz


Date:    Thu, 2 Apr 1998 07:43:35 +1000
From:    David Rodger 
Subject: Re: follow that body

Arjen van der Schoot  wrote:

>I have been involved in a few interactive music/animation projects and am
>about to embark on another one. For this piece the movements of 5 people
>across three 6meter*6meter rooms
>should be tracked simultaneously. Right now I am tossing about ideas to get
>this done.

Are you trying to differentiate bodies or simply sense their movement in

>So far I have hands on experience with Icube, sensorlab, and data coming
>from weight sensors, light breakers and simple potentiometers. Also I have
>been playing around with Big Eye a bit.

Ah, I see.  Sounds as if you're trying to detect where the motion happens,
not who causes it.  Of course, if you want to identify the mover, the
methods you cite are difficult to adapt.  Big Eye allows differentiation if
your movers are wearing blobs of different colours.

Garth Paine or Todd Winkler may care to comment on VNS's capabilities in
this regard.

Regards, David

David Rodger ------------ Audio Engineer, Pool Lifeguard, RLSS Trainer
Personal:     (not yet exciting)
Research:   (a bit more exciting)
"Officially the La Trobe Music Department is in existance for the next
2 years and will close on 31 December, 1999.  (This is really just our
way of avoiding the Y2K bug)."                          -- David Hirst


Date:    Tue, 31 Mar 1998 18:36:33 -0500
From:    todd_winkler@BROWN.EDU
Subject: follow the body

I have been using the VNS for more than a year for several dance and
installation works. I am also currently working with ICUBE and we are
developing custom sensors for installations.

The VNS is very dependable and accurate. However, it cannot differentiate
the movements of five different people since it detects motion only  by
measuring changes in light. So, anything moving within the view of one or
two video cameras will be detected (in my recent dance piece, a
computerized lighting board "plays" the music with patterns of light, along
with the dancers - we also have video projections triggering sound). I know
that David Rokeby, who designed and builds the VNS, has recently added
object tracking to the software, but I haven't tried it yet. The only
commercially available system I know of that might be able to do this is
Big Eye, since it can track specific colors.

If you are interested I have a paper about the VNS entitled, Creating
Interactive Dance with the Very Nervous System. You can find it at:



End of MAX Digest - 30 Mar 1998 to 31 Mar 1998