Subject: MAX Digest - 21 Nov 1999 to 22 Nov 1999 (#1999-334)
Date: Tue, 23 Nov 1999 00:00:21 -0500
From:
Automatic digest processor <LISTSERV@LISTS.MCGILL.CA>
Reply-To: MAX - Interactive Music/Multimedia Standard Environments <MAX@LISTS.MCGILL.CA>
To: Recipients of MAX digests <MAX@LISTS.MCGILL.CA>


There are 8 messages totalling 255 lines in this issue.

Topics of the day:

  1. MSP/bw-G3/MacOS 8.6(/Audiowerk8) crashing problem
  2. sucking down audio streams
  3. Koblo synths/Max
  4. Additive synthesis control (2)
  5. 4 channels out of a PB
  6. What time is it?
  7. Had a look at the Tactex controller pad.

----------------------------------------------------------------------

Date:Mon, 22 Nov 1999 12:13:40 +0100
From:Trond Lossius <Trond.Lossius@BERGEN.IT-AKADEMIET.NO>
Subject: Re: MSP/bw-G3/MacOS 8.6(/Audiowerk8) crashing problem

I'm sorry I can't be of any help to you, but maybe you could help me? I've
tried to use a Audiowerk8 with Max/MSP without success. What driver(s) are
you using and where do you store them? Do you install the soundcard driver
in the extention folder or the Max folder? Do you use the Max Audio Library
or some other audio driver?

Thank you

Trond L.

------------------------------

Date:Mon, 22 Nov 1999 14:06:26 +0000
From:Roald Baudoux <roald.baudoux@BRUTELE.BE>
Subject: Re: sucking down audio streams

>there is the obvious thing of plugging a stereo mini cable patch from the
>out to the in... but is this asking for trouble?

I did it once with the MetaSynth demo and Sound Edit 16 recording in the
background. I didn't establish a direct connection between the input and
the output but I made the signal going through a simple preamplifier using
the "tape" output to feed the Mac input. It worked fine except for the poor
S/N ratio of the Mac's input and output.

Roald Baudoux

------------------------------

Date:Mon, 22 Nov 1999 16:45:35 +0000
From:Roger Carruthers <thelone.roger@VIRGIN.NET>
Subject: Koblo synths/Max

[Chris; sorry about the previous <HTML> version - new email prog c/w
Microsoft feature bloat]

I've created a patch to drive Koblo's Vibra1000 softsynth, which it does
without any problems, but when I try to use the same app (or any Max app)
with it's big sisters, Vibra 6000 or 9000, I get all kinds of clicks &
glitches from the audio,the Midi timing goes haywire and the screen becomes
very slow; when you record the synth's output to disc however, the audio is


clean. Any ideas what's happening here? I've given both Max & Tokyo (the
Vibra synth engine) plenty of memory and I'm using it with a 300MHz G3,
running OS8.6.
Incidentally, does the presence of MSP (albeit only in demo mode), put much
more load on the host machine's CPU or could it be in conflict with the
Koblo audio engine ? If so, how does one uninstall MSP?

------------------------------

Date:Mon, 22 Nov 1999 18:35:09 +0100
From:Georg Hajdu <hajdu@UNI-MUENSTER.DE>
Subject: Re: Additive synthesis control

Hi Sukandar,

Thank you for your reply.

> I might be wrong, but I think coll performs a symbol match even if you us=
e
> numbers, so tables would be better but I suppose you want to use floats..=
.

That's right. But I doubt that this alone will explain the almost
exponential degradation of coll's performance with large amounts of data.


> my suggestion would be to do your own external
> let me know your specs and I see if I can do a quick coding run

Since I'm not a C programmer I may even take you up on your offer.
Ideally, there should be a one-dimensional "array" object that can read
files from disk; but one should also be able to store data (list of defined
size) into it.

The data type should be defined by an argument (in analogy to "pack" or
"trigger"). A second argument should indicate the number of items per line,
i.e. with argument "1. 64", the message "1" would cause "array" to send out
its first 64 floats as a list etc.

> at what rate do you need to read data ?

Ideally, at 3 msecs per line. 5 msecs will do too, though.

> how big exactly can your files be ?

What about, say, 10 MBytes?

> can buffer sizes be fixed at instantiation time ?

Maybe one could use a third argument for buffer size.

> Pyrite

What does it do? How can I get it?

Thanks,

Georg

***************************************************
Dr. Georg Hajdu

Hochschule f=FCr Musik Detmold,
Abteilung M=FCnster
Ludgeriplatz 1
D-48151 Muenster


e-mail: hajdu@uni-muenster.de
http://www.mhs-muenster.de/Hajdu.html
****************************************************

------------------------------

Date:Mon, 22 Nov 1999 12:49:28 -0500
From:David Gamper <dgamper@EARTHLINK.NET>
Subject: 4 channels out of a PB

At AES '99, I talked to someone at Digigram (maker of the VXPocket
2in, 2 out) and he indicated that they were working on a 4 channel
i/o (at least) PCMCIA card. "Sometime next year" was the expected
release date I recall.

- David G.

------------------------------

Date:Mon, 22 Nov 1999 10:37:44 -0800
From:Peter Elsea <elsea@CATS.UCSC.EDU>
Subject: What time is it?

>The following patch demonstrates that even QT is about 1/4 per
>cent off ....

A thread not visited in a few years, but still a candidate for the FAQ.
Look at the clock in the menubar; I'll bet it's a few minutes fast (maybe
slow) unless you've set it recently. The Mac is a notoriously poor
timekeeper, bad enough that system 9 includes an online time utility that
synchronizes the clock with your server. Commercial grade crystals
typically have a spec of 100 ppm which translates to 0.01% or one second
out of ten thousand. (Every three hours it's a second off.) This is
compounded by software problems such as rounding errors and interrupt
precedences. Finally, the mac OS dosen't allow very intimate access to the
timing hardware, you can only make time manager calls, which is akin to
asking the hotel desk clerk to wake you up at 7:00 AM precisely. Software
timing is a mess, no matter what software and what platform.

But it wouldn't matter, even if the accuracy were 1 part per million,
because what you really need is syncronicity with the rest of the world,
which may be running at some other rate. If you need to do video, your
video card locks onto a video timebase to make your images stable at
whatever variant of 29.97 fps is in your neighborhood. If you want to do
digital audio, you lock to a master wordclock. If you want accurate MIDI
tempos, lock onto either timecode or beat clock.

(In Colorado, the US National Institute of Standards and Technology
http://www.boulder.nist.gov/timefreq/keeps a clock that is accurate to 10
^ -15 seconds. Every couple of years they have to reset the clock, because
the rotation of the earth isn't that stable.)

Peter Elsea
Director, Electronic Music Studios
University of California, Santa Cruz
http://arts.ucsc.edu/EMS/Music/index.html
elsea@cats.ucsc.edu

------------------------------

Date:Mon, 22 Nov 1999 15:52:04 EST
From:BHUOTT@AOL.COM
Subject: Had a look at the Tactex controller pad.

I recently saw a demo of the MultiTouch Controller made by Tactex. If you
haven't heard of it before it's a fiberoptic based sensitive pad which can
resolve multiple contact points in fine resolution (check tactex.com). The


demo unit was attached to a PC laptop with a graphic display of the pad input.

My first reaction was giddy delight at seeing gestures from multiple fingers
(even all 10) faithfully mapped using a pad that, on the surface, resembles a
simple mouse pad, but with a more supple feel. It was easy to see how this
data could be used for many things in MAX. I understand that they've written
an object to provide both the direct data stream from the device's sensors as
well as processed data with x-y position and pressure for individual points
in contact. Not needing to wear gloves or attach anything to the body is a
nice thought considering how many hours I've spent tethered to an iCube
bumbling over the trackpad, swimming in wires and gaffer's tape.

On the almost-ready side of things, the beta models out there scan too slowly
to trigger events in musically precise time. I'm told this will be much
faster in future models of the pad. Speed, along with other changes in the
works, will make this controller worth keeping an eye on. It has a lot of
potential and the company is interested in input from anyone with application
ideas.

I'd like to hear from any MAX'ers who are already using the MTC.

Bob Huott

------------------------------

Date:Tue, 23 Nov 1999 02:11:32 +0100
From:Sukandar Kartadinata <sk@ZKM.DE>
Subject: Re: Additive synthesis control

Hi Georg,

on second thought:

'Larry' by Peter Elsea should actually do the job

(part of the Lobjects available at the usual sources
e.g. http://www.employees.org/~max/mirror/IRCAM/
or (with some luck) IRCAM directly:
ftp://ftp.ircam.fr/pub/forumnet/max)

best,
Sukandar


>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Sukandar Kartadinata
Custom Music Technology
Hagenauerstr. 6, 10435 Berlin, 030-44051219
http://members.xoom.com/Sukandar/
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

------------------------------

End of MAX Digest - 21 Nov 1999 to 22 Nov 1999 (#1999-334)
**********************************************************