6/14/97 11:00 PM
Subject: MAX Digest - 13 Jun 1997 to 14 Jun
1997To: Recipients of MAX digests 

There are 7 messages totalling 208 lines in this issue.

Topics of the day:

  1. Scrypt object
  2. Max/SCII NuBus
  3. Recorded Outputs
  5. truncation
  6. Seduced and Abandonded by Opcode
  7. non SCSI CD-ROM drives


Date:    Sat, 14 Jun 1997 08:38:42 +0000
From:    lionel hubert 
Subject: Scrypt object

Hi !
I'm a newcomer in this list.
I got problems with the scrypt object by James Mc Cartney.
is there someone who know this object, in the list?
                                                        Thank, and excuse my
Bad English


Date:    Sat, 14 Jun 1997 13:07:10 +0300
From:    Gil Wasserman 
Subject: Max/SCII NuBus


I'm thinking of getting a SampleCell II card for my 7100/80. Can anyone
tell me how Max runs it, via OMS, MIDI manager or directly (is there a
driver around for the SCII) and what is the best solution?




"Frogs are not forever"


Date:    Sat, 14 Jun 1997 01:36:23 -1000
From:    Darin Au 
Subject: Re: Recorded Outputs

I've had no problem recording samples out of my Mac using the little
mini-jack.  I take it out (stereo mini to dual 1/4") into two channels of
a mixer (I've used several different mixers) and take it to DAT.  Works
fine, sounds fine.  I've done entire tape pieces this way.



Date:    Sat, 14 Jun 1997 08:24:28 -0800
From:    Christopher Dobrian 

Regarding the question by Ciaran Maher at c.maher@DARTINGTON.AC.UK:

>I'm running a loop which resets the 'metro' speed on each bang to the
>result of an 'expr'. My problem is that the 'expr' result is a 'f ' which
>is truncated by 'metro' to miliseconds even when the decimal is 5 or

If I understand correctly, Ciaran is concerned about accurately producing
beat speeds that are not an integer number of milliseconds between beats.
For example, at MM=90 a quarter note lasts 666.666667 milliseconds, so
setting your metro speed at 667 results in an error of 1/3 millisecond,
which adds up to a substantial error after a few measures of music.

However, one question this raises in my mind is "What's a 'substantial
error', given that Max's metro is the timekeeper?" I mean, it's not like
any of US is going to recognize that Max is actually playing a tempo of
89.955 instead of 90. And given the notorious innacuracy of the Macintosh
clock with respect to "real" time, such a concern would be pretty ludicrous
anyway. Max's metro is correct with respect to itself, and any error with
regard to idealized "real time" usually goes unnoticed.

However, the error MAY be significant if it conflicts with some other
timekeeper in the musical context.

For example, this would be a problem if you're trying to do a polyrhythm
such as 7:4 (e.g., septuplet sixteenths against straight sixteenths) with
two different metros. At MM=60 the 4 will have a pulse speed of 250 ms and
the 7 will have a pulse speed of 142.857143 ms. So whether you round or
truncate the floating point pulse speed, your 7 and your 4 will eventually
get out of synch.

The best solution to the problem of synchronzing such polyrhythms is to use
the tempo object instead of metro, which will accurately keep you on the
specified tempo (beat speed). Obviously, tempo still only outputs with
millisecond accuracy, and (apparently) does its own internal rounding of
non-integer-millisecond output rates. (This despite the fact that it allows
you to specify floating point tempi with the tempo message.)

Another solution is just to periodically restart (with a single trigger)
each of the timing objects that must be in synch.

Another (clumsier) solution is to calculate the passage of time as a float
and keep track of the running time. At the same time, keep track of the
running time of the metro (by adding its ms rate to an accumulatated total
on each beat). On each beat, compare your floating point running time to
the integer running time produced by the metro, and whenever the two times
differ by 1ms or more, add in an extra millisecond to the tempo of metro
for a single beat.

In any case, Max will never be more precise than (the Mac's imprecise
concept of) a millisecond, so I advise to just synchronize what needs
synchronizing and don't sweat any microseconds you can't hear.


             Christopher Dobrian / School of the Arts - Music
             University of California / Irvine, CA 92697-2775
                Phone: (714) 824-7288 / Fax: (714) 824-4914


Date:    Sat, 14 Jun 1997 12:47:20 EDT
From:    Nick Longo <71477.2332@COMPUSERVE.COM>
Subject: Re: truncation

Subject: Seduced and Abandonded by Opcode

I've just received this message, but haven't worked with Timeline at all,
nor with the MTP.  Can anyone offer any help?  If so, please send a
response directly to Trip Khalaf, since he's on the road and evidently in
dire need of help.  Millions of Michael Jackson fans will thank you.

>Date: Thu, 12 Jun 1997 18:35:25 -0400 (EDT)
>Subject: Seduced and Abandonded by Opcode
>My name is Trip Khalaf. I'm the senior engineer at Clair Brothers Audio
>mixing for Michael Jackson on tour in Europe. Some time ago I started using
>Max to display show notes and send MIDI program changes around the place.
>this tour I decided to take things one step futher and pick off SMPTE from
>the central computer on stage. I've spent a lot of time on the project and
>have run into what I hope is a minor snag. When I call Opcode and mention
>Max, strange noises happen in the background and people run out of the
>They then refer me to a University in Canada that either doesn't respond,
>denies any knowledge of anything called Max. After much searching, your
>came up.
>     I hate to bother you, but I'm desparate. Opcode has proven useless.
>Here's my problem:
>I'm trying to use the timeline object to trigger a bunch of events.
>Everything works perfectly using the internal clock, but I can't get the
>timeline object to synch to the MIDI Timpiece II, which I'm using as the
>SMPTE interface.  I created a setclock object which the timeline recognizes
>when I click on the clock in the patcher. The first argument of the
>is the name that the timeline sees, the second argument is SMPTE (I also
>tried EXT). I'm receiving 29.97 fps and have the time units in the pop up
>menu set to 30 fps. OMS recognizes the MIDI Timepiece II. The MTP console
>my computer sees midi time code. Still no time code in the timeline. I'm
>running Max 3.5.1 and OMS 2.3.1. I'm using a Powerbook 540c. Any ideas?
>Please help.

Richard Zvonar, PhD                    
(818) 760-8055 voice/fax               



Date:    Sun, 15 Jun 1997 11:43:04 +1000
From:    Garth Paine 
Subject: Re: non SCSI CD-ROM drives

I regularaly do projects using a number of CD drives to output audio
controled by MAX.  I use FWB CD-ROM toolkit to provide the drivers to the
CD-ROM drives.  This works fine with the MAX CD object.  I just thought I'd
mention this because it may also provide an interface to the PowerBook 1400
CD drive.

Regards,  Garth

Garth Paine
Composer, Sound Designer,
Museums, Dance, Film, Theatre
Interactive Immersive Environments
61 3 95259844 (W)



End of MAX Digest - 13 Jun 1997 to 14 Jun 1997