Subject: MAX Digest - 21 Oct 1998 to 22 Oct 1998 (#1998-46)
Date: Fri, 23 Oct 1998 00:00:02 -0400
From: Automatic digest processor 
Reply-To: MAX - Interactive Music/Multimedia Standard Environments
To: Recipients of MAX digests 

There are 15 messages totalling 585 lines in this issue.

Topics of the day:

  1. MSP and Digi Sound Drivers? Mac AV cards?(was :G3 boards for 8100)
  2. Lin
  3. Floating numberbox
  4. copy protection, part 8215
  5. symbol madness
  6. Programming question
  7. common MIDI drivers (2)
  8. Re Lin, symbols and strings
  9. Could you help me please?
 10. notation
 11. lists
 12. Audiomedia III
 13. port closing
 14. MAX Digest - 20 Oct 1998 (#1998-44)

McGill is running a new version of LISTSERV (1.8d on Windows NT).
Information is available on the WEB at


Date:    Thu, 22 Oct 1998 01:38:43 -0400
From:    Jon Witte 
Subject: Re: MSP and Digi Sound Drivers? Mac AV cards?(was :G3 boards for

> Neville John Cross Wrote:
> I'm using a 7100/80(OS 7.6.1) with the 'old' Protools Project card ,an 882
> with Digi Sound Driver 1.4.1 and Digi Init 3.1.1. I have been  waiting for
> Newertech G3/240mhz board for seven months now!! - I should
> receive it next week.
> Anybody have any experience with Newertech G3\240 or 210 cards,
> the old Protools 'Project' _and_  MSP ?
> One of the only reasons I need the upgrade card is so I can run MSP.
> ( and IRCAM's 'Spat')
> Anyone think I'm going to have problems with the card and MSP?

MSP Runs good on my 8100 w/G3-240mhz (Newer Technology).  The example
patches like Cross Dog AB and Forbidden Planet run using just a little
over 50 percent of the CPU Power.  I would guess that this is pretty
good since the patches both have a little note in them that reads "CPU
Intensive"  Any one else have stats of this nature for comparison's
sake?  I'm running the patches via Sound Manager using the 1/8" stereo
outputs.  When I try to use the  Digi Sound Manager Driver to to use my
Pro Tools 442 system (I don't know if this is older or newer than the
Project Card) , MSP becomes very unstable.  I don't think that this is
caused by the G3 card, but by the crappie programming of sound manager
and the Digi sound manager driver.  With the 442 system, I can't run a
digi system init that is newer than 2.9.6 (I actually use 2.9.5 - 2.9.6
is for Motu Products.) I also can't run any Pro Tools software greater
than version 3.2.  So by the fact that you are running version 3.1.1 of
the Digi System Init, you might have better results.

> Also,could anyone explain the Mac AV card to me? Is it still
> possible to buy them? Where from and how much :) ?

Back in the old days........there was an AV option on the 7100's and the
8100's.  A computer with the AV (7100AV or 8100AV) on the end of it's
name had an Audio/Video card in it that came stock straight from Apple.
These cards gave you a monitor hookup with 2 megs of VRAM on the card.
Also you got  a stereo set of inputs and outputs on RCA jacks as well as
a VCR style video input and output on RCA jacks.  The AV card resides in
your PDS (Processor Direct Slot).  If you have one of these AV cards (or
as in my case an Apple Video Card) and you upgrade to the G3, the AV
card will plug into the flexible expansion plug that hangs off of the
Newer Tech G3 card.  The newer Tech card in turn will plug into the
PDS.  Physically (Not plugged in) the AV card will reside in the center
nubus slot.  It is possible to regain the center nubus slot by doing
some remodeling of the computer case with a saws all.  I did this to my
8100 with excellent results.  When the modification is finished your AV
card will sit horizontally across the tops of the Nubus cards.  I don't
know if a similar modification is possible on the 7100.  If you do
decide to hack up the computer case with some kind of electric Saws All
or something, be sure to remove all the components from the case as to
not ZAp anything with static electricity.  This AV card is not a pro
sound quality card, but it does do OK audio at 16 bit/44.1khz.  I think
it would be fine for getting audio in and out of MSP, especially when
there really is no other alternative, as MSP doesn't support Nubus
stuff.  This is what David Z. wrote me about nubus cards from Digi:
Hope you don't mind me posting this David.  :  )

David Z. Wrote:

> If I could support the Nubus Protools cards, I would. But
> Digi won't release information for "direct" support of their
> cards and they aren't actively supporting Nubus cards with
> their Direct I/O driver, which is what I use with MSP. I
> looked into using DAE with MSP (which would allow support
> for all Protools systems) but it only plays sound coming
> from the hard disk, so it doesn't make any sense to use
> with something like MSP, which is more like a synthesizer.
> You may be able to use the Protools card with the Digidesign Sound
> Manager driver, but that driver is completely lame and will
> only do input or output but not both at the same time. And
> there isn't any indication from Digidesign that this will ever
> be fixed (although I know they feel bad about it).

I don't know how well the AV card will work with MSP, but my gut feeling
is good as a lot of people have these AV cards and are use them for
sound manager recording without problems.  I have not heard of any
problems with the Apple AV Card and the Newer Tech G3 cards as of yet.
I think I'm going to get one.
They are available from Galaxy Hardware Publishing (a great Mac
Resource) for $195.00.

Galaxy Hardware Publishing contact info:

Hope this helps

Jon Witte - Music Machine Studio
*         Dancing in the moonlight, to a spiritual waltz-        *
*            Dancing to the music that's in my soul.             *


Date:    Thu, 22 Oct 1998 09:32:09 +0200
From:    "Dr. Karlheinz Essl" 
Subject: Re: Lin

Dear MAXers!

Roby Steinmetzer wrote:

>  and  are part of Peter Elsea's Lobjects.
> They used to be at
> I tried to verify the location but I just couldn't get a connection with
> the server, even with Fetch.

The correct URL is:

Hope that helps.

Cheers, and happy MAXin'

   Dr. Karlheinz Essl - Composer
   Vienna / Austria
   Studio for Advanced Music & Media Technology


Date:    Thu, 22 Oct 1998 00:07:52 -0700
From:    David Zicarelli 
Subject: Re: Floating numberbox

ryohros  asks:

>How can I make the preset object recognize floating point numberboxes
>it's this a problem of having 3.5 instead of latest upgrade?

I believe this was fixed in 3.5.1; it's certainly fixed now.

David Z.


Date:    Thu, 22 Oct 1998 00:24:14 -0700
From:    David Zicarelli 
Subject: copy protection, part 8215

The other day I arrogantly claimed that it was impossible to
lose authorizations when downloading the Max update from the
Opcode web site. Well, I think that's wrong--it is possible
to lose authorizations if you download the OS8 version (the
one that works with the 3.5.8 disk) but you have a 3.5 or 3.5.1 key
disk. In this scenario, you could corrupt your key disk if
you run your newly downloaded Max, then it asks for a key disk
(because your drive isn't authorized for the 3.5.8 key), then
you innocently insert your 3.5 or 3.5.1 key disk and some error
message appears. Later you might find that your precious key disk is
completely trashed.

Those of you with old 3.5 key disks, please get them turned into
new ones. It's supposed to be free since the old disks will not
authorize under OS8 or later. I know it's difficult for
people outside the US to do so, but I would like to be able to post
updates for one key disk only to the web site. As it stands now,
Opcode has to maintain an old version of the copy protection just
to prepare new Max versions for people with old key disks,
and I have to make two versions of everything. If I were running
Opcode, I would have just sent new key disks to all Max users
rather than waiting for people to call and ask for them. This
nightmare has dragged on long enough.

David Z.


Date:    Thu, 22 Oct 1998 00:48:30 -0700
From:    David Zicarelli 
Subject: Re: symbol madness

Stephen Kay  writes:

>For example, if you connect a metro, counter, and numbox to a
>sprintf %s, then every single number creates a new entry in Max's
>Symbol Table, and eventually the memory will spiral down to zero.
>Correct?  Whether or not you get tnewptr errors or crashes or not.
>This doesn't seem like an optimal solution, especially for cases
>where Max installations are expected to run for days or weeks.

Well, yes, that's true, but show me an installation based on
this technique, unless it's one designed to make some point
about the limitations of Max.

>At the risk of asking a stupid question, why isn't there some
>way to do string manipulation *without* allocating symbols in
>Max's Symbol Table?  That's what this whole discussion was about,
>and what my attempted object was trying to do. Apparently, there
>*is* no way to take a string in, manipulate it, and then send
>it out without creating an entry in Max's Symbol Table, and
>allocating memory.

No there isn't. We haven't really discussed why symbols are
cool here. A symbol allows any two words to be compared by
looking at a simple reference rather than the entire
character string. This is much faster for the computer to
perform. For instance, when you send the "start" message to the
dac~ object, the proper function to run (the one that starts
the audio) is determined by comparing the symbol "start" with
the symbols in the table of messages that the dac~ object
understands. This comparison can work quickly by looking at
single numbers rather than checking if the text of a particular
entry in the message table matches the text of the word start,
which has to be done character-by-character. In this way, you
could generate a symbol on the fly with sprintf or Stephen's new
object and then pass it to a Max object to execute a message,
without (in the best case) doing any string comparisons
whatsoever. (You might have to do some string comparisons
to store a new symbol in a table, this is known in the literature
on hash tables as "crashing" but I said yesterday I wasn't
going to get into that.)

To create a separate type called "string" where stuff is not
stored in the symbol table would mean that one would
have to symbolize it in order to do anything useful with the
string. My sense is that this issue has only been a problem
due to the bug I just fixed having to do with Overdrive and
symbol creation, and that no one is creating symbol tables
so large as to overwhelm their computer's memory.

Miller Puckette's design of this aspect of Max (which I had
nothing to do with) is in my opinion a brilliant compromise
between dynamism and efficiency, and I don't feel like
breaking it right now, or any time soon. I'm not saying that
it couldn't use a little tweaking now and then, however.

David Z.


Date:    Thu, 22 Oct 1998 15:59:10 +0100
From:    Todor Todoroff 
Subject: Programming question

I've got two questions concerning programming external audio modules in MSP.
The second one is specially puzzling. I hope someone might help me explain
the situation (David?).

1) What would be a good practice when allocating memory in *module_new()?
I used (on the ISPW-16) to try to allocate memory first and, if successfull,
I created the object.
In MSP, I allocate memory with newhandle (is this the right method? it works
rather well for two complex modules I wrote, but I seldomly experienced
crashes when load a patch: the computer doesn't crash, but the modules don't
function correctly and I have to reload the patch) and, if succesful I
create the object with
   x = (t_module *)newobject(module_class);

As there is no such example in the manual, I'm wondering what would happen
if I create the object first and then am not able to allocate all necessary
memory. How should I then destroy the object?
Are there any guidelines?

2) I experienced immediate crashes when setting audio on when initialising a
rather short vector of 2048 floats with
   set_zero(float *dest, long int);
in module_dsp(t_module *x, t_signal **sp, short *count)
I don't experience chrashes when this is located either in *module_new() or
in the perform method.
Is there any reason why it shouldn't work?
I tried while (--n) *p=0.0; and it also crashed.
It doesn't crash for n=100 though, so I would like to know whether there is
anay specified limit on the amount of computation allowable in module_dsp()
Thanks for any help,
Todor Todoroff


Date:    Fri, 23 Oct 1998 01:04:17 +0930
From:    Ross Bencina 
Subject: common MIDI drivers

Hi MAX people,

I will begin by apologising for a slightly off topic question, but this list
has the highest concentration of expert Macintosh MIDI users that I know of.
So here goes...

I've been asked to write some 'generic' (really compatible) cross-platform
interface/glue code for Macromedia Director. The code should run with all
'common' MIDI interfaces /
driver setups, etc... I'm trying to write code that can be installed on any
client machine and co-exist happily with whatever other midi software they
have installed.

For the Macintosh I am planning to support OMS and native (direct serial
port, no manager) midi. As Apple is not supporting MidiManager any longer
(as far as I know),
I'm not planning on supporting it.

For the PC I am just planning to support the standard midi driver

I have heard of MidiShare, but I'm not sure how common it is.

Does anyone know of any other midi management tools / drivers that are
commonly used?

Any suggestions regarding MIDI management/compatibility issues on either
platform would also be welcome.

ThankYou for your time and bandwidth.

Ross Bencina. 

BTW. If you have access to a PC you may want to check out my shareware
real-time dsp environment AudioMulch:


Date:    Thu, 22 Oct 1998 10:48:51 -0700
From:    Peter Elsea 
Subject: Re Lin, symbols and strings

The Lobjects are at The server was
down most of yesterday. The computing crew just moved the server to a new
machine, and are having some growing pains.

About symbol madness-
The symbol table is an elegant part of Max, with tremdous advantages for
traditional musical problems. But now we are using Max in ways that get
farther and farther afield, and direct string manipulation is required. Lin
addresses part of this issue, providing an input mechanism to get user
input as acsii lists. However, if you connected the right output of Lin to
sprint, you would be abusing the symbol table, creating a new symbol on
each keystroke. That's why the help file shows how to display input with an
lcd object. lcd works, but it is less than ideal for this purpose.

What we need is a message box that users can type into, and that displays
text from lists of ascii. It's on my to do list, but at the rate I'm
getting to things, you'd be lucky to see it by next August. Steve? David?

Peter Elsea
Electronic Music Studios
University of California, Santa Cruz


Date:    Thu, 22 Oct 1998 11:06:42 -0800
From:    James Livingston 
Subject: Could you help me please?

Neil Ziesing is the in house person responsible for the International

If you are having prblems obtaining anything you need from your local
distributor, please let Neil know all of the details of your situation by
sending him an e-mail:

For info on who the international distributor is for any country outside of
the US and Canada visit:

Roby, I have forwarded your message to Neil.  In the mean time, you may
also want to try  contacting your distributor again:

                                 Company Name:
                                                  Dr. Persoonslaan 46
                                                  323 860 9090
                                                  323 860 9091

If you are continuing to have problems, please do not hesitate to let Neil
know the nature of the situation you are encountering.



Date:    Thu, 22 Oct 1998 14:09:14 -0400
From:    Chris Murtagh Hrdc-drhc 
Subject: Re: common MIDI drivers

From: "Ross Bencina" , on 10/22/98 11:34 AM:
>Does anyone know of any other midi management
>tools / drivers that are commonly used?

 There is also FreeMidi which is done by Opcode's rival MOTU. It
theoretically has OMS emulation, but it really stinks, and is not all that
stable (I think OMS stinks too BTW). Maybe Apple would be so nice as to get
of their butts and support/develop a standard (new) MidiManager that ALL
developers could live with. Imagine, a stable Midi manager that would work
with all software that needs/uses Midi... what a concept!!

Chris (always opinionated)
To say that Windows 95 is just like Mac OS is like finding a potato in the
shape of Jesus and thinking you have witnessed the second coming.


Date:    Thu, 22 Oct 1998 14:54:03 -0400
From:    Don Malone 
Subject: notation

>musical notation display within the Max environment... something which has
>been overlooked for some incomprehensibe reason.
while notation is important to some
many others of us bypass this interface
to deal more directly with sound

happy tunes



Date:    Thu, 22 Oct 1998 15:05:57 -0400
From:    Don Malone 
Subject: lists

someone mentioned another ea list
is there a list of ea lists with URLs?

happy tunes



Date:    Thu, 22 Oct 1998 15:46:33 -0400
From:    Chris Murtagh Hrdc-drhc 
Subject: Audiomedia III

Does anyone here have any idea what an Audiomedia III card is worth? I want
to sell mine, I don't remember what I paid for it, and I don't have a clue
what it goes for.  Any info appreciated (please email me directly), even
better would be a reliable link (URL) to updated marked prices to audio
gear.  Thanks.



Date:    Thu, 22 Oct 1998 23:17:38 +0000
From:    marc ravelomanantsoa 
Subject: port closing

Is there other way to close the serial port used by Max (modem i.e.) without
using menus : Midi set Up/ OMS mdi set up.... I mean with a max object ?



Date:    Thu, 22 Oct 1998 23:34:43 +0000
From:    marc ravelomanantsoa 
Subject: Re: MAX Digest - 20 Oct 1998 (#1998-44)

Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854";
Content-Transfer-Encoding: 7bit

 Thanks David for the new cd object.
It seems that the file was not updated so I don't know how to tell
cd object
witch of my two cd drives (asci ID # n  and ATAPI bus m ).
The cd object had recognize the two drives regarding the message window
 "  CD interface type scsi
    cd : SCSI ID 4, LUN 0
    CD interface type atapi
     cd : ATAPI bus 1
     cd : found 2 drives


marc .


   -- if you are wondering why mac

Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

 Thanks David for the new cd object.

It seems that the file was not updated so I don't know how
to tell  the cd object

witch of my two cd drives (asci ID # n  and ATAPI bus m ).

The cd object had recognize the two drives regarding the message window

 "  CD interface type scsi

    cd : SCSI ID 4, LUN 0

    CD interface type atapi

     cd : ATAPI bus 1

     cd : found 2 drives




marc .
   -- if you are wondering why mac

--------------E0DA3032DD8C8DCE1EAC05AB-- ------------------------------ End of MAX Digest - 21 Oct 1998 to 22 Oct 1998 (#1998-46) *********************************************************