From herbstmondwind at web.de  Thu Jul  1 10:32:59 2004
From: herbstmondwind at web.de (daniel mayer)
Date: Thu, 01 Jul 2004 19:32:59 +0200
Subject: [CM] Gentlecoord-1.0
Message-ID: <531226665@web.de>

Gentlecoord is an interface for CM, written in CL. 
It may be used for the coordination of algorithms with voices 
(in an instrumental sense),  for which rhythm, pitch and  
dynamics are defined via patterns to generate MIDI files or
MIDI realtime output.
It includes a MIDI setup facility to define instruments and pitchbend,
a function to start an algorithm with several voices in parallel,
options to solve pitch conflicts between them,
a syntax for cr/dim on single events and new pattern classes for 
establishing logical relations between distinct pattern objects.
Maybe it is interesting for people who use CM to compose
for acoustic instruments.

At the moment it has only been tested on the following Mac combinations: 

OSX / OpenMCL / cm-2.4.2 and
OS9 / MCL 4.3.1 / cm-2.4.2, cm-2.3.4

Gentlecoord assumes familiarity with CM syntax, so see 
the CM documentation and Rick's book for reference.
Download (sources, examples and documentation) from 

http://on1.zkm.de/zkm/stories/storyReader$4101


If anyone uses it: feedback is welcome !

Many thanks to Rick Taube and Tobias Kunze, 
Hanspeter Kyburz and Ludger Bruemmer 

Daniel Mayer

____________________________________________________
Aufnehmen, abschicken, nah sein - So einfach ist 
WEB.DE Video-Mail: http://freemail.web.de/?mc=021200



From anthony.kozar at utoledo.edu  Fri Jul  2 12:13:34 2004
From: anthony.kozar at utoledo.edu (Anthony Kozar)
Date: Fri, 02 Jul 2004 15:13:34 -0400
Subject: [CM] Gentlecoord-1.0
In-Reply-To: <531226665@web.de>
Message-ID: <BD0B2C1D.B2D8%anthony.kozar@utoledo.edu>

Daniel, 

Thanks very much for this new and interesting way of working with CM!

I installed Gentlecoord and ran through some of the examples.  I think it
will take some real study for me to understand fully what is going on, but I
am impressed with the results that one can get from such "simple" code.

I am wondering though whether you think it would be relatively easy or not
for me to adapt Gentlecoord to produce Csound output instead of Midi?  I
don't really work with Midi much these days.

Thanks again.

Anthony Kozar
anthony.kozar at utoledo.edu
http://akozar.spymac.net/



From andersvi at notam02.no  Thu Jul  8 08:32:01 2004
From: andersvi at notam02.no (Anders Vinjar)
Date: Thu, 08 Jul 2004 17:32:01 +0200
Subject: [CM] configure in CLM
In-Reply-To: <40CECC73.2070500@ccrma> (Bill Schottstaedt's message of "Tue,
 15 Jun 2004 03:16:19 -0700")
References: <40CECC73.2070500@ccrma>
Message-ID: <m3briqtszy.fsf@notam02.no>

>>>>> " " == Bill Schottstaedt <bil at ccrma.Stanford.EDU> writes:

     > The current CLM uses a configure script to set up the C
     > side switches -- a long overdue change, but one that is
     > bug-prone.  Let me know of any problems!

The configure-script works fine here.  A minor adjustment would
be to make it callable regardless of which directory you're in
while loading all.lisp.  Heres a change which adjusts ./configure
to clm-directory/configure:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: all.lisp.diff
Type: text/x-patch
Size: 816 bytes
Desc: diff output
URL: <https://cm-mail.stanford.edu/pipermail/cmdist/attachments/20040708/19aba0c4/attachment.bin>
-------------- next part --------------

-anders

From guillaume at nulko.com  Fri Jul  9 12:22:40 2004
From: guillaume at nulko.com (Guillaume Germain)
Date: Fri, 09 Jul 2004 15:22:40 -0400
Subject: [CM] Making changes to snd's gui
Message-ID: <40EEF080.6050706@nulko.com>

Hi all,

I'm currently working on developing a voice synthesis extension for snd.

I would like to be able to select a region in the transform graph 
(specifically on a sonogram). In fact the best for me would be to mirror 
marks and selection between the time domain graph and the transform 
graph. Also another alternative could be to replace the time domain 
graph by a sonogram with the same selection / marking capabilities as 
the time domain graph.

I'd be grateful if someone could point me at the places in the sources 
where I could make modifications and/or give me hints on how I can 
accomplish what I want to do.

Many thanks,

Guillaume



From k.s.matheussen at notam02.no  Sun Jul 11 12:12:44 2004
From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen)
Date: Sun, 11 Jul 2004 21:12:44 +0200 (CEST)
Subject: [CM] Making changes to snd's gui
In-Reply-To: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu>
References: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu>
Message-ID: <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain>

Guillaume Germain:
> 
> Hi all,
> 
> I'm currently working on developing a voice synthesis extension for snd.
> 
> I would like to be able to select a region in the transform graph 
> (specifically on a sonogram). In fact the best for me would be to mirror 
> marks and selection between the time domain graph and the transform 
> graph. Also another alternative could be to replace the time domain 
> graph by a sonogram with the same selection / marking capabilities as 
> the time domain graph.
> 
> I'd be grateful if someone could point me at the places in the sources 
> where I could make modifications and/or give me hints on how I can 
> accomplish what I want to do.
> 

It should be quite easy to do this in scheme. Take a look at the
mark and selection handling code I have made in gui.scm and 
snd_conffile.scm.

Something like this could be a start:

(add-hook! after-graph-hook
	     (lambda (snd ch)
                 (let ((das-marks (get-selection-data snd ch))
                       (das-selection (get-selection-data snd ch))
                       (das-sonogram (get-sonogram-position snd ch)))
                    (for-each (lambda (mark-info)
                                 (put-mark-in-sonogram das-sonogram mark-info))
                              das-marks)
                    (put-selection-in-sonogram das-sonogram das-selection))))

-- 



From bil at ccrma.Stanford.EDU  Mon Jul 12 03:47:01 2004
From: bil at ccrma.Stanford.EDU (Bill Schottstaedt)
Date: Mon, 12 Jul 2004 03:47:01 -0700
Subject: [CM] Making changes to snd's gui
In-Reply-To: <40EEF080.6050706@nulko.com>
References: <40EEF080.6050706@nulko.com>
Message-ID: <40F26C25.1090202@ccrma>

Those ideas are on my "to-do" list -- I noticed that the
sound editor used by the folks on the Cassini project had
a tracking cursor in a sonogram.



From guillaume at nulko.com  Mon Jul 12 12:11:29 2004
From: guillaume at nulko.com (Guillaume Germain)
Date: Mon, 12 Jul 2004 15:11:29 -0400
Subject: [CM] Making changes to snd's gui
In-Reply-To: <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain>
References: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu> <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain>
Message-ID: <40F2E261.5050201@nulko.com>

Kjetil Svalastog Matheussen wrote:

>Guillaume Germain:
>  
>
>>[...]
>>    
>>
>
>It should be quite easy to do this in scheme. Take a look at the
>mark and selection handling code I have made in gui.scm and 
>snd_conffile.scm.
>
>Something like this could be a start:
>
>[...]
>

Thanks a lot for your suggestions and pointers. This is really helping 
me to get started.

I've hit a few small snags trying to make 'snd_conffile.scm' works. I 
have what I think is the latest version (July 12th), and compiled with 
"./configure --with-gtk". The first problem was that Guile reported 
'gtk_widget_get_name' to be undefined. I've traced that problem back to 
the 'xg' module loading code, present in a few files ('gui.scm', 
'gtk-popup.scm', 'snd-gtk.scm', ...), where xm.so is loaded instead of 
xg.so:

(if (not (provided? 'xg))
    (let ((hxm (dlopen "xm.so"))) ;; Should be xg.so ?
      (if (string? hxm)
          (snd-error (format #f "snd-gtk.scm needs the xg module: ~A" hxm))
          (dlinit hxm "init_xm"))))

The next small problem was in the "-notebook" part of the 
snd_conffile.scm around line 1500, (list-ref (main-widgets) 5) is #f for 
me so a type error is signaled by string=?, a simple check was enough to 
make it load the file succesfully:

(if (or (not use-gtk)
        (let ((w (list-ref (main-widgets) 5)))
          (and (string? w) (not (string=? "GtkNotebook" 
(gtk_widget_get_name w))))))
              ...)

I found loading "snd_conffile.scm" to be very illustrative of the power 
of snd, and I'd like to take that opportunity to thank all those who 
contributed to make that great piece of software.

Guillaume



From k.s.matheussen at notam02.no  Tue Jul 13 08:32:23 2004
From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen)
Date: Tue, 13 Jul 2004 17:32:23 +0200 (CEST)
Subject: [CM] Making changes to snd's gui
In-Reply-To: <20040712190001.19334.40029.Mailman@cm-mail.stanford.edu>
References: <20040712190001.19334.40029.Mailman@cm-mail.stanford.edu>
Message-ID: <Pine.LNX.4.58.0407131728100.19616@hensten.localdomain>

Bill Schottstaedt:
>
> Those ideas are on my "to-do" list -- I noticed that the
> sound editor used by the folks on the Cassini project had
> a tracking cursor in a sonogram.

Actually, when thinking about it, wouldn't the best and
quickest way just be to let the graph function create sonograms?

-- 



From k.s.matheussen at notam02.no  Tue Jul 13 12:24:15 2004
From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen)
Date: Tue, 13 Jul 2004 21:24:15 +0200 (CEST)
Subject: [CM] Making changes to snd's gui
In-Reply-To: <40F2E261.5050201@nulko.com>
References: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu>
 <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain> <40F2E261.5050201@nulko.com>
Message-ID: <Pine.LNX.4.58.0407132120440.19616@hensten.localdomain>

On Mon, 12 Jul 2004, Guillaume Germain wrote:

> Kjetil Svalastog Matheussen wrote:
> 
> >Guillaume Germain:
> >  
> >
> >>[...]
> >>    
> >>
> >
> >It should be quite easy to do this in scheme. Take a look at the
> >mark and selection handling code I have made in gui.scm and 
> >snd_conffile.scm.
> >
> >Something like this could be a start:
> >
> >[...]
> >
> 
> Thanks a lot for your suggestions and pointers. This is really helping 
> me to get started.
> 
> I've hit a few small snags trying to make 'snd_conffile.scm' works. I 
> have what I think is the latest version (July 12th), and compiled with 
> "./configure --with-gtk". The first problem was that Guile reported 
> 'gtk_widget_get_name' to be undefined. I've traced that problem back to 
> the 'xg' module loading code, present in a few files ('gui.scm', 
> 'gtk-popup.scm', 'snd-gtk.scm', ...), where xm.so is loaded instead of 
> xg.so:
> 

I think that is correct, actually. Its name is actually xm.so, even for 
gtk.(?) Anyway, configure snd with the --static-xm switch to fix this.


> 
> I found loading "snd_conffile.scm" to be very illustrative of the power 
> of snd, and I'd like to take that opportunity to thank all those who 
> contributed to make that great piece of software.
> 

Thank you. Yesterday, I added automation for ladspa plugins. Heres a 
screen-shot: http://folk.uio.no/ksvalast/snd_ladspa_automation.png
(Code has just been sent to Bill)

-- 



From fbar at footils.org  Tue Jul 13 14:04:05 2004
From: fbar at footils.org (Frank Barknecht)
Date: Tue, 13 Jul 2004 23:04:05 +0200
Subject: [CM] Making changes to snd's gui
In-Reply-To: <Pine.LNX.4.58.0407132120440.19616@hensten.localdomain>
References: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu> <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain> <40F2E261.5050201@nulko.com> <Pine.LNX.4.58.0407132120440.19616@hensten.localdomain>
Message-ID: <20040713210405.GA10556@fliwatut.scifi>

Hallo,
Kjetil Svalastog Matheussen hat gesagt: // Kjetil Svalastog Matheussen wrote:

> I think that is correct, actually. Its name is actually xm.so, even for 
> gtk.(?) 

Yes, AFAIK.

> Yesterday, I added automation for ladspa plugins. Heres a
> screen-shot: http://folk.uio.no/ksvalast/snd_ladspa_automation.png

Kjetil, you are creepy! Holy cow, this is cool.

Ciao
-- 
 Frank Barknecht                               _ ______footils.org__



From dlphilp at bright.net  Tue Jul 13 15:02:12 2004
From: dlphilp at bright.net (Dave Phillips)
Date: Tue, 13 Jul 2004 18:02:12 -0400
Subject: [CM] Making changes to snd's gui
In-Reply-To: <20040713210405.GA10556@fliwatut.scifi>
References: <20040711190003.20184.39471.Mailman@cm-mail.stanford.edu> <Pine.LNX.4.58.0407112105090.10823@hensten.localdomain> <40F2E261.5050201@nulko.com> <Pine.LNX.4.58.0407132120440.19616@hensten.localdomain> <20040713210405.GA10556@fliwatut.scifi>
Message-ID: <40F45BE4.2070908@bright.net>

Hi all:

  Very cool ! I want one...

Best,

dp


Frank Barknecht wrote:

>Hallo,
>Kjetil Svalastog Matheussen hat gesagt: // Kjetil Svalastog Matheussen wrote:
>
>  
>
>>I think that is correct, actually. Its name is actually xm.so, even for 
>>gtk.(?) 
>>    
>>
>
>Yes, AFAIK.
>
>  
>
>>Yesterday, I added automation for ladspa plugins. Heres a
>>screen-shot: http://folk.uio.no/ksvalast/snd_ladspa_automation.png
>>    
>>
>
>Kjetil, you are creepy! Holy cow, this is cool.
>
>Ciao
>  
>




From ricktaylor at speakeasy.net  Wed Jul 14 06:01:23 2004
From: ricktaylor at speakeasy.net (ricktaylor at speakeasy.net)
Date: Wed, 14 Jul 2004 13:01:23 +0000
Subject: [CM] Making changes to snd's gui
Message-ID: <W10065942867671089810083@webmail2>

> From: Frank Barknecht [mailto:fbar at footils.org]

> > Yesterday, I added automation for ladspa plugins. Heres a
> > screen-shot: http://folk.uio.no/ksvalast/snd_ladspa_automation.png
> 
> Kjetil, you are creepy! Holy cow, this is cool.

 :} Sure is...




From andersvi at extern.uio.no  Wed Jul 14 03:44:48 2004
From: andersvi at extern.uio.no (Anders Vinjar)
Date: Wed, 14 Jul 2004 12:44:48 +0200
Subject: [CM] configure in CLM
In-Reply-To: <m3briqtszy.fsf@notam02.no> (Anders Vinjar's message of "Thu,
 08 Jul 2004 17:32:01 +0200")
References: <40CECC73.2070500@ccrma> <m3briqtszy.fsf@notam02.no>
Message-ID: <copllhmsw9r.fsf@bruno.localdomain>

Why not have all.lisp leave config.h and friends in
*clm-binary-directory*, and modify 'definstrument to write
include config.h from this directory (along the model of building
and using 'sndplay)?

There might be flaws to this model, but it seems to work
allright, not having to put <config.h> explicitly (by superuser)
in any /usr/include/* (or having a copy of config.h in every
directory you might be working).

Here are 2 diff-files making the changes to all.lisp and
defins.lisp to test with:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: all.lisp-diffs
Type: application/octet-stream
Size: 1821 bytes
Desc: diff output for new all.lisp
URL: <https://cm-mail.stanford.edu/pipermail/cmdist/attachments/20040714/722a44ff/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: defins-diffs
Type: application/octet-stream
Size: 1672 bytes
Desc: diff output for new defins.lisp
URL: <https://cm-mail.stanford.edu/pipermail/cmdist/attachments/20040714/722a44ff/attachment-0001.obj>
-------------- next part --------------

-anders

From andersvi at extern.uio.no  Wed Jul 14 09:02:37 2004
From: andersvi at extern.uio.no (Anders Vinjar)
Date: Wed, 14 Jul 2004 18:02:37 +0200
Subject: [CM] local variables not shadowing in defprocess
Message-ID: <cophdsashk2.fsf@bruno.localdomain>

I beleive theres some bug somewhere in latest (CVS cm-2.5.0).
Local variables set inside #'defprocess are not looked up if
theres a global variable with the same name.

Heres an example:

CM> *cm-date*
; Evaluation aborted
CM> (cm-version)
"Common Music 2.5.0"
CM> (setf localvar 200)
200
CM> (defprocess tester ()
       (let ((localvar 100))
         (process repeat 4
                  do
                  (pprint localvar))))
TESTER
CM> (events (tester) "test.midi")

200
200
200
200
"test.midi"
CM> 



From taube at uiuc.edu  Thu Jul 15 07:00:35 2004
From: taube at uiuc.edu (Rick Taube)
Date: Thu, 15 Jul 2004 09:00:35 -0500
Subject: [CM] local variables not shadowing in defprocess
In-Reply-To: <cophdsashk2.fsf@bruno.localdomain>
References: <cophdsashk2.fsf@bruno.localdomain>
Message-ID: <5809A2A8-D667-11D8-8948-000A95674CE4@uiuc.edu>

> I beleive theres some bug somewhere in latest (CVS cm-2.5.0).
> Local variables set inside #'defprocess are not looked up if
> theres a global variable with the same name.

If I understand you correctly this is not a bug its just a nasty 
consequence of working with closures and "special" variables in common 
lisp:

? (defparameter param 1)
PARAM
? (defparameter fn (let ((param -99)) #'(lambda () param)))
FN
;; so now a funcall of fn  should be -99, right?? wrong...
? (funcall fn)
1

Now try scheme, which does not believe in special variables(nor needs a 
"funcall" kludge:

guile> (define param 1)
guile> (define fn (let ((param -99)) (lambda () param)))
guile> (fn)
-99


personally i would throw away common lisp but for the fact that, for 
whatever reason, scheme still has no reasonable answer for someone that 
wants a good introspective object system AND fast code. but maybe 
someone can enlighten me.




From bil at ccrma.Stanford.EDU  Fri Jul 16 03:48:01 2004
From: bil at ccrma.Stanford.EDU (Bill Schottstaedt)
Date: Fri, 16 Jul 2004 03:48:01 -0700
Subject: [CM] local variables not shadowing in defprocess
In-Reply-To: <5809A2A8-D667-11D8-8948-000A95674CE4@uiuc.edu>
References: <cophdsashk2.fsf@bruno.localdomain> <5809A2A8-D667-11D8-8948-000A95674CE4@uiuc.edu>
Message-ID: <40F7B261.3060905@ccrma>

 > personally i would throw away common lisp

I would too... But, I just spent 3 weeks hacking like a fiend
to rescue the CL version of CLM ("clm-3") and I hope to have it
ready by sometime next week.  I had stopped work on the previous
version (clm-2) because the code made me angry just looking at
it, so it was either a total rewrite, or abandon CL.



From juanig at ccrma.Stanford.EDU  Mon Jul 19 07:26:32 2004
From: juanig at ccrma.Stanford.EDU (Juan Reyes)
Date: Mon, 19 Jul 2004 10:26:32 -0400
Subject: [CM] Re: [PlanetCCRMA] Blender for animations
In-Reply-To: <40F98D25.3090409@bright.net>
References: <1090009652.21914.42.camel@strawberri.maginvent.org>
	 <40F98D25.3090409@bright.net>
Message-ID: <1090247191.2029.67.camel@strawberri.maginvent.org>

I can't resist but making this comment for our sake and for that of
other Unix users: Blender should contain more Emacs or SND shortcuts.
Perhaps this will make that "learning slope" not so steep. 

As Dave points out Blender results look very sexy indeed.

I remember the days attempting visualizing the sound-object under
acousmatique terms. Perhaps one should leave and effort time to see that
object with plain eyes and not ears.

  --* Juan

On Sat, 2004-07-17 at 16:33, Dave Phillips wrote:
> Hi Juan:
> 
>   I've only toyed with earlier versions of Blender, but I recently 
> attended a very good presentation about it at the libre software meeting 
> in Bordeaux. Some of the demonstration images and animations were 
> impressive, and Martin Poirier made its use seem attainable by mere 
> mortals. I'd suggest spending the time to get acquainted with its 
> basics, and try to use a recent release.
> 
> Best regards,
> 
> dp
> 
> 
> Juan Reyes wrote:
> 
> >Is anyone using Blender for animation stuff on Linux. It seems to me
> >that it has steep learning curve because of its rather complex
> >multi-button interface. 
> >
> >Results seems to be good though.
> >
> >Any alternatives, any suggestions ?
> >
> >  --* Juan Reyes
> >

> 
> _______________________________________________
> PlanetCCRMA mailing list
> PlanetCCRMA at ccrma.stanford.edu
> http://ccrma-mail.stanford.edu/mailman/listinfo/planetccrma



From anthony.kozar at utoledo.edu  Mon Jul 19 22:31:37 2004
From: anthony.kozar at utoledo.edu (Anthony Kozar)
Date: Tue, 20 Jul 2004 01:31:37 -0400
Subject: [CM] CLM-3
In-Reply-To: <40F7B261.3060905@ccrma>
Message-ID: <BD222679.B60C%anthony.kozar@utoledo.edu>

Just wondering ... is the new clm-3 file on FTP a "finished" version or not
ready to be used yet?

This is kind of exciting -- a new major version of CLM !  :)

Anthony Kozar
anthony.kozar at utoledo.edu
http://akozar.spymac.net/


On 7/16/04 6:48 AM, Bill Schottstaedt<bil at ccrma.Stanford.EDU> etched in
stone:

> I would too... But, I just spent 3 weeks hacking like a fiend
> to rescue the CL version of CLM ("clm-3") and I hope to have it
> ready by sometime next week.



From bil at ccrma.Stanford.EDU  Tue Jul 20 03:55:49 2004
From: bil at ccrma.Stanford.EDU (Bill Schottstaedt)
Date: Tue, 20 Jul 2004 03:55:49 -0700
Subject: [CM] CLM-3
In-Reply-To: <BD222679.B60C%anthony.kozar@utoledo.edu>
References: <BD222679.B60C%anthony.kozar@utoledo.edu>
Message-ID: <40FCFA35.4060203@ccrma>

CLM-3

CLM has moved to version 3.  This turned out to be a much bigger
change than I originally intended.  The initial plan was to
incorporate clm.c into the CL run macro, but one thing led to another,
and by the time I was done, much of the program was gone: the no-ffi
(clisp) support, real-time graphics and controls (better handled by
Snd), old Mac MCL/MPW code, the parallel instrument handlers, the help
system, etc.  The basic goal was to unify the various versions of CLM
-- all are now based on clm.c -- while minimizing instrument-level
changes.  A side effect that became a goal was to refocus CLM on its
original purpose (sound synthesis).

Here are the specific changes:

Removed clisp/no-ffi support:
     rc.c, rc.lisp
     no-ffi and clisp switches.

Removed real-time controls support:
     noffishm.c and all "control" functions
     rt.lisp
     bess*.cl
     preverb.ins, stochasticRT.ins, revon.ins

Removed old Mac OS/MPW support:
     ToolServer.Lisp, maclib.lisp, make-clm.cl, mac.help
     much of mac.lisp (except for ffi links)

Removed with-psound/defpinstrument support:
     vp.ins
     rcp.c

Removed obsolete files:
     jl.ins, load-clm.cl, make-clm.cl
     clm-changes.html, translate.cl (CLM-1 stuff)
     fltdes.lisp
     moog.ins (moog.lisp appears to be more recent)

Removed obsolete/redundant functions and variables:
     clm-simple-mix
     *input* (and all related function args)
     rec-any
     cnv-reverb in cnv.ins

Removed clm-help:
     the help subdirectory and indexing support for it
     clm-help.lisp, menu.lisp
     clm-inspect, clm-help, linux-help
     *clm-help-browser* *clm-linux-helper* *clm-helper-choice*

Removed the *explode* feature, and the notion of a lisp instrument
     (definstrument now assumes C -- use "defun" if you want lisp).
     also removed save-signal.

Removed mus-open-write, mus-reopen-write, mus-open-read, mus-create, mus-close.
     These are available under more perspicuous names in clm1.lisp.
     Also mus-file-open-descriptors, mus-write-zeros, mus-write-float.
     Also mus-file-to-array (use file->array), mus-array-to-file (use file->array)

Removed excessive speed: CLM-3 is a bit slower than CLM-2 --
     according to my timing tests, at least half of the slow down comes
     from using doubles in place of floats -- a few of the benchmarks
     are faster in clm-3.

Removed def-clm-struct support for generator fields (I still need to
     update dlocsig)

Removed instrument-let

open-input and close-input are almost no-ops -- they just save their
     arguments for some future reference (backwards compatibility).
     removed open-output, reopen-output, close-output (you can
     get the same functionality from sndlib's mus-sound-open-ouput).

Added run support for:
     mus-name, mus-file-name, mus-xcoeff, mus-ycoeff, mus-equal
     mus-describe, mus-inspect, polar->rectangular,
     sound-comment, mus-header-type-name, mus-data-format-name

Added "declare" types (run macro): string, mus-any, bignum,
     int*, double*. strings and bignums can be passed through run.
     Due to CL package confusion, these declare types should
     prefix a colon (i.e. use the keyword package).

Added off_t sample number support.

Added generators: ssb-am, phase-vocoder, also run-time function
     args for editing and so on.

Added classes: sample->file, frame->file

changed pv-* to phase-vocoder-* (snd7.scm has old names)

The tarball is named clm-3.tar.gz; I'll leave the old clm-2.tar.gz
at ccrma-ftp indefinitely.  There are still 30 or 40 things-to-do,
but I think the basic program is usable.

I was mulling over these changes while driving into work, and it
occurred to me that clm's precursor (Sambox) lasted about 15 years,
clm-1 lasted 10, clm-2 5 -- what does this sequence portend for clm-3?



From marco at trevisani.net  Mon Jul 19 16:09:52 2004
From: marco at trevisani.net (Marco Trevisani)
Date: Mon, 19 Jul 2004 19:09:52 -0400
Subject: [CM] Re: [PlanetCCRMA] Blender for animations
In-Reply-To: <1090247191.2029.67.camel@strawberri.maginvent.org>
References: <1090009652.21914.42.camel@strawberri.maginvent.org> <40F98D25.3090409@bright.net> <1090247191.2029.67.camel@strawberri.maginvent.org>
Message-ID: <20040719230952.GA26772@trevisani.net>

 19/07/2004 alle 10:26:32, -0400, Juan Reyes [AKA juanig] ha scritto/escribi?/wrote:
 
> As Dave points out Blender results look very sexy indeed.

I started to look at it last winter (and tried out different sofwtare
available in the animation field, povray in particular) and it is
quite good.  Just do not try to understand the logic behind certain
shortcuts....:-) The mixer/sequencer is quite good (sound needs to be
processed with other tools, clm and snd of course...) it really helps
to put together a final work with audio and graphics.

In a, short, moment of fanaticism, i used blender for an architectural
project....well not so crazy if you think that blender can export in dxf
- the equivalent of .rtf in the CAD family.

Povray is just a scripting language, very easy to understand and
probably faster to program than blender, if you prefer (like i do) the
non-gui oriented world...(povray is not free software in the "pure"
sense, it is just free, like in free beer...)

       short comment for an off topic theme...

       ciao,
       marco

-- 
************************************************************************
* marco trevisani                                                      *
* http://marco.trevisani.net       marco at trevisani.net                 *
* Neither MS-Word nor MS-PowerPoint attachments please:                *
* See http://www.fsf.org/philosophy/no-word-attachments.html           *
* Gpg Fingerprint = A9CD 12D6 B658 A545 9BD1  77E9 47AD 8753 8110 7070 *
************************************************************************



From fd106 at york.ac.uk  Tue Jul 20 10:03:29 2004
From: fd106 at york.ac.uk (Fergal Dowling)
Date: Tue, 20 Jul 2004 18:03:29 +0100
Subject: [CM] csound score, 2 instruments
Message-ID: <B95BB006-DA6E-11D8-9A7F-00039367667A@york.ac.uk>

Hi,
I wish to write a csound score for that uses two instruments, i1 and 
i2, and control their respective start times using independent time 
varying envelopes. To achieve this, I now output two separate .sco 
files, one for each instrument, and then copy and paste one into the 
other. How can I append the out the second files to the first?

Thanks in advance.
Fergal Dowling.

fd106 at york.ac.uk
http://www-users.york.ac.uk/~fd106/


;;; Define new class object for csound instrument 1.
(defobject i1 (i)
   (amp)
  (:parameters time dur amp))

;;; Define new class object for csound instrument 2.
(defobject i2 (i)
   (amp)
  (:parameters time dur amp))

;;; Define first start time envelope
(define start-env '(0 .9 1 .1))

;;; Define second start time envelope
(define start-env2 '(0 0.1 1 0.9))

;;; Define process
(define (my-first-process start-env end)
(let ((rhy 1))
     (process for beg from 0 by rhy
     until (> (now) end)
     for nstart = (interpl (/ (now) end) start-env)
            output (new i1 :time (now)
                        :dur .5
                        :amp 1)
	   wait nstart)))

;;; Define second process
(define (my-second-process start-env2 end)
(let ((rhy 1))
     (process for beg from 0 by rhy
     until (> (now) end)
     for nstart = (interpl (/ (now) end) start-env2)
            output (new i2 :time (now)
                        :dur .5
                        :amp 1)
	   wait nstart)))
	
;;; Render first instrument
(events (my-first-process start-env 60)
  "ins1.sco" 0)

;;; Render second instrument
(events (my-second-process start-env2 60)
  "ins2.sco" 0)



From bakanonaka at yahoo.com  Tue Jul 20 11:15:01 2004
From: bakanonaka at yahoo.com (kelly hirai)
Date: Tue, 20 Jul 2004 11:15:01 -0700 (PDT)
Subject: [CM] Re: Blender
In-Reply-To: <20040719190001.25437.68832.Mailman@cm-mail.stanford.edu>
Message-ID: <20040720181501.52862.qmail@web20025.mail.yahoo.com>

i've been keeping up with blender for a while. the
post open source version is alot diferent than the
pre. they added a real raytracer, tabbed control
panels. 

Blender's keyboard shortcuts are getting packed. ie.
u, alt-u ctl-u shift-u all have some function. i think
any modeler worth its beans is going to take some
study and practice [oh now don't wince, we're all into
music right?].

...

[a day later]

for that matter it would be cool if blender had a maya
keymapping setting as well. 

i would be particularly interested in if anyone has
thought about the physics part of blender, the game
component has colision detection. anyone considered
how to gleen data from that? what fromat it should
take?
midi? csound? what othere codex are out there? 

the offical blender 2.3 guide was released around
april. it has a quickstart tootorial. unfortunately
the game engine isn't covered in this book. i believe
there is another book comming on that subject. 

i'm also impressed with the level of abstraction
pov-ray has atained. the short code contest is mighty
cool.

also of note in the 3d-sound crossover, archive.org
has a movie with perry cook's physical models,
animated. unfortunately, the shapes listened to in the
movie are regular geometric solid primatives, and so
its unclear to me if he is using his STK models or if
they are volumetric mesh solids with waves passing
through them and a listening location. anyone seen
this and know how it was made?

kelly hirai


Message: 1
From: Juan Reyes <juanig at ccrma.Stanford.EDU>
To: planetCCRMA <planetccrma at ccrma.Stanford.EDU>
Cc: cmdist at ccrma.Stanford.EDU
Date: Mon, 19 Jul 2004 10:26:32 -0400
Subject: [CM] Re: [PlanetCCRMA] Blender for animations


I can't resist but making this comment for our sake
and for that of
other Unix users: Blender should contain more Emacs or
SND shortcuts.
Perhaps this will make that "learning slope" not so
steep. 

As Dave points out Blender results look very sexy
indeed.

I remember the days attempting visualizing the
sound-object under
acousmatique terms. Perhaps one should leave and
effort time to see 
that
object with plain eyes and not ears.

  --* Juan

On Sat, 2004-07-17 at 16:33, Dave Phillips wrote:
> Hi Juan:
> 
>   I've only toyed with earlier versions of Blender,
but I recently 
> attended a very good presentation about it at the
libre software 
meeting 
> in Bordeaux. Some of the demonstration images and
animations were 
> impressive, and Martin Poirier made its use seem
attainable by mere 
> mortals. I'd suggest spending the time to get
acquainted with its 
> basics, and try to use a recent release.
> 
> Best regards,
> 
> dp
> 
> 
> Juan Reyes wrote:
> 
> >Is anyone using Blender for animation stuff on
Linux. It seems to me
> >that it has steep learning curve because of its
rather complex
> >multi-button interface. 
> >
> >Results seems to be good though.
> >
> >Any alternatives, any suggestions ?
> >
> >  --* Juan Reyes
> >

> 
> _______________________________________________
> PlanetCCRMA mailing list
> PlanetCCRMA at ccrma.stanford.edu
>
http://ccrma-mail.stanford.edu/mailman/listinfo/planetccrma





=====
kelly hirai  http://ongaku.isa-geek.net/~khirai


		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!
http://promotions.yahoo.com/new_mail 



From anthony.kozar at utoledo.edu  Tue Jul 20 11:13:02 2004
From: anthony.kozar at utoledo.edu (Anthony Kozar)
Date: Tue, 20 Jul 2004 14:13:02 -0400
Subject: [CM] CLM-3 -> back to CLM-2 
In-Reply-To: <40FCFA35.4060203@ccrma>
Message-ID: <BD22D8DF.B631%anthony.kozar@utoledo.edu>

On 7/20/04 6:55 AM, Bill Schottstaedt<bil at ccrma.Stanford.EDU> etched in
stone:

> Removed old Mac OS/MPW support:
> ToolServer.Lisp, maclib.lisp, make-clm.cl, mac.help
> much of mac.lisp (except for ffi links)

I discovered this last night while trying to build it ...

So I switched back to trying to get the lastest clm-2 to install (MCL 4.3.1,
MacOS 9.1).  I created the empty config.h file and loaded all.lisp.  Got the
"cannot find clmLib" message so ran the makefile directly in MPW and
received the following messages:

MrC C Compiler 4.1.0f1c1
Copyright (C) 1994-1999 by Apple Computer, Inc.

#include <config.h>
                 ^
File "sndlib.h"; line 10 #Fatal error: unable to open input file 'config.h'
#-----------------------
### MPW Shell - Execution of tmp.make terminated.

Now obviously, the config.h that I created is in the clm-2 directory and
isn't being found in the system include path (why isn't this a user-style
include: "config.h" ?).  So I cheated and used the MrC option "-inclpath
ignoresys" which makes it treat <...> like "...".  Then I get these
messages:

int fileno(FILE *fp);
                  ^
File "sndlib.h"; line 646 #Error: ')' expected
#-----------------------
#-----------------------
  fd = OPEN(arg, O_RDONLY, 0);
                           ^
File "io.c"; line 564 #Error: undefined identifier 'O_RDONLY'
#-----------------------
  if ((fd = OPEN(arg, O_RDWR, 0)) == -1)
                              ^
File "io.c"; line 588 #Error: undefined identifier 'O_RDWR'
#-----------------------
int fileno(FILE *fp)
                  ^
File "io.c"; line 1354 #Error: ')' expected
#-----------------------
#-----------------------
  else
  ^
File "io.c"; line 1358 #Error: ';' or ',' expected
#-----------------------


And now I am not sure what to do next.

:(

Anthony Kozar
anthony.kozar at utoledo.edu
http://akozar.spymac.net/



From RickTaylor at Speakeasy.Net  Tue Jul 20 22:57:13 2004
From: RickTaylor at Speakeasy.Net (RickTaylor at Speakeasy.Net)
Date: Tue, 20 Jul 2004 23:57:13 -0600
Subject: [CM] Re: [PlanetCCRMA] Blender for animations
Message-ID: <dwwzkm4b.yzeuxzadsehr@localhost.localdomain>

Marco,
> I started to look at it last winter (and tried out different sofwtare
> available in the animation field, povray in particular) and it is
> quite good.  Just do not try to understand the logic behind certain
> shortcuts....:-) The mixer/sequencer is quite good (sound needs to be
> processed with other tools, clm and snd of course...) it really helps
> to put together a final work with audio and graphics.
> 
> In a, short, moment of fanaticism, i used blender for an architectural
> project....well not so crazy if you think that blender can export in dxf
> - the equivalent of .rtf in the CAD family.
> 
> Povray is just a scripting language, very easy to understand and
> probably faster to program than blender, if you prefer (like i do) the
> non-gui oriented world...(povray is not free software in the "pure"
> sense, it is just free, like in free beer...)

 My own vote would go to Blender with Yafray and Python scripting. 

 Pov is an excellent render engine. Ayam/Aqsis is a good system. K3d is pretty full featured. AL has some nice advantages. Art of Illusion is nice. http://www.coin3d.org/ Coin3d and Inventor are nice. http://oss.sgi.com/projects/inventor/ Moho is very nice. {not 3d} http://www.lostmarble.com/index.shtml {not free either}

 ...
This message has been brought to you in part by a grant from Columba.



From RickTaylor at Speakeasy.Net  Tue Jul 20 22:57:13 2004
From: RickTaylor at Speakeasy.Net (RickTaylor at Speakeasy.Net)
Date: Tue, 20 Jul 2004 23:57:13 -0600
Subject: [CM] Re: [PlanetCCRMA] Blender for animations
Message-ID: <dwwzkm4b.yzeuxzadsehr@localhost.localdomain>

Marco,
> I started to look at it last winter (and tried out different sofwtare
> available in the animation field, povray in particular) and it is
> quite good.  Just do not try to understand the logic behind certain
> shortcuts....:-) The mixer/sequencer is quite good (sound needs to be
> processed with other tools, clm and snd of course...) it really helps
> to put together a final work with audio and graphics.
> 
> In a, short, moment of fanaticism, i used blender for an architectural
> project....well not so crazy if you think that blender can export in dxf
> - the equivalent of .rtf in the CAD family.
> 
> Povray is just a scripting language, very easy to understand and
> probably faster to program than blender, if you prefer (like i do) the
> non-gui oriented world...(povray is not free software in the "pure"
> sense, it is just free, like in free beer...)

 My own vote would go to Blender with Yafray and Python scripting. 

 Pov is an excellent render engine. Ayam/Aqsis is a good system. K3d is pretty full featured. AL has some nice advantages. Art of Illusion is nice. http://www.coin3d.org/ Coin3d and Inventor are nice. http://oss.sgi.com/projects/inventor/ Moho is very nice. {not 3d} http://www.lostmarble.com/index.shtml {not free either}

 ...
This message has been brought to you in part by a grant from Columba.



From RickTaylor at Speakeasy.Net  Tue Jul 20 23:03:51 2004
From: RickTaylor at Speakeasy.Net (RickTaylor at Speakeasy.Net)
Date: Wed, 21 Jul 2004 00:03:51 -0600
Subject: [CM] Re: [PlanetCCRMA] Blender for animations
In-Reply-To: <dwwzkm4b.yzeuxzadsehr@localhost.localdomain>
Message-ID: <dwwzt4u4.yy4sqww6e50m@localhost.localdomain>

>  My own vote would go to Blender with Yafray and Python scripting. 


Forgot tsipp:
http://public.kitware.com/pipermail/vtkusers/2001-August/057104.html

 ...
This message has been brought to you in part by a grant from Columba.



From andersvi at extern.uio.no  Wed Jul 21 03:18:15 2004
From: andersvi at extern.uio.no (Anders Vinjar)
Date: Wed, 21 Jul 2004 12:18:15 +0200
Subject: [CM] csound score, 2 instruments
In-Reply-To: <B95BB006-DA6E-11D8-9A7F-00039367667A@york.ac.uk> (Fergal
 Dowling's message of "Tue, 20 Jul 2004 18:03:29 +0100")
References: <B95BB006-DA6E-11D8-9A7F-00039367667A@york.ac.uk>
Message-ID: <copwu0x64uw.fsf@bruno.localdomain>

>>> "FD" == Fergal Dowling <fd106 at york.ac.uk> writes:

    FD> Hi, I wish to write a csound score for that uses two
    FD> instruments, i1 and i2, and control their respective
    FD> start times using independent time varying envelopes. To
    FD> achieve this, I now output two separate .sco files, one
    FD> for each instrument, and then copy and paste one into the
    FD> other. How can I append the out the second files to the
    FD> first?

Passing a list of the processes you want to output to #'events
will merge them together in the same destination.

(events (list  #&seqone #&seqtwo) "together.sco" 0)

(events (list (my-first-process start-env 60)
	      (my-second-process start-env2 60))
  "ins3.sco" 0)

; Common Music 2.5.0 output on 21 Jul 2004, 12:12:19
i1 0 0.5 1
i2 0 0.5 1
i2 0.1 0.5 1
i2 0.20133334 0.5 1
i2 0.30401778 0.5 1
i2 0.40807134 0.5 1
i2 0.5135123 0.5 1
i2 0.6203591 0.5 1
i2 0.7286306 0.5 1
i2 0.83834565 0.5 1
i1 0.9 0.5 1
i2 0.94952357 0.5 1
i2 1.0621839 0.5 1
i2 1.1763463 0.5 1
...

If you actually need to append them for some reason you can
output the resulting subobjects to temporary seq's before
appending them together in a separate seq:

(events (my-first-process start-env 60)
	(new seq :name 'seqone) 0)

(events (my-second-process start-env2 60)
	      (new seq :name 'seqtwo) 0)

(new seq :name 'together)

(setf (subobjects #&together) (append (subobjects #&seqone) (subobjects #&seqtwo)))

(events #&together "together.sco" 0)

; Common Music 2.5.0 output on 21 Jul 2004, 12:15:15
i1 0 0.5 1
i1 0.9 0.5 1
i1 1.7879999 0.5 1
i1 2.6641598 0.5 1
i1 3.5286376 0.5 1
i1 4.381589 0.5 1
i1 5.223168 0.5 1
i1 6.0535254 0.5 1
...



From bil at ccrma.Stanford.EDU  Wed Jul 21 04:03:58 2004
From: bil at ccrma.Stanford.EDU (Bill Schottstaedt)
Date: Wed, 21 Jul 2004 04:03:58 -0700
Subject: [CM] CLM-3 -> back to CLM-2
In-Reply-To: <BD22D8DF.B631%anthony.kozar@utoledo.edu>
References: <BD22D8DF.B631%anthony.kozar@utoledo.edu>
Message-ID: <40FE4D9E.6090805@ccrma>

 > And now I am not sure what to do next.

I hate to say this, but I think you need to either get OSX
or use some other program -- the problems you mention appear
to be caused by the lack of a real config.h (i.e. one that
reflects your setup).