CLM news
To: cmdist
bil
Mon, 23 Jan 95 08:29:25 PST
Recent changes to CLM include better SMS support (sms.lisp),
a rudimentary stab at LPC (lpc.c/h/lisp), and some new examples
of using the convolve generator (cnv.ins), including reverb
through convolution with exponentially decaying Gaussian noise,
as recommended by J.A.Moorer at the dawn of computing.
From: Rick Taube <hkt@guido.zkm.de>
Date: Tue, 24 Jan 95 15:13:00 GMT+0100
To: cmdist@ccrma.Stanford.EDU
Subject: graphic version of Common Music
Cc: algo-comp@heinous.isca.uiowa.edu, hkt@guido.zkm.de
The latest cm.sea.bin on ccrma-ftp.stanford.edu and ftp.zkm.de contains a new and completely graphical version of Common Music for the Macintosh under MCL 2.0.1. The graphical interface,
called Capella, is designed to completely replace the
existing command interpreter, but the two can actually be
run in parallel. To learn about the interface, just build
the system as usual, then look under Help in the main Stella
menu. Choosing Load Tutorial from the help menu should get you
started. Here is the About text that comes with the system:
----
Capella is a new graphical interface for Common Music. People
familiar with the system should find Capella to be much faster,
easier and clearer than using the old command line interpreter.
Someone new to algorithmic composition in Common Music should
find their learning curve substantially reduced. Capella
currently runs on the Macintosh under MCL 2.0.1. The eventual
aim is to support several window systems using at least one
public domain Lisp. The first candidate is probably CMUCL
using X11 on the SGI.
Capella is developed by two people, Rick Taube (hkt@zkm.de) and
Tobias Kunze (tkunze@ccrma.stanford.edu). This is the first
release; don't use it if you are a nervous soul or expect bug
free code. We would be happy to receive your bug reports,
criticisms and ideas.
---
Rick Taube
hkt@zkm.de
Zentrum fuer Kunst u. Medientechnologie
Karlsruhe, Deutschland
Date: Tue, 14 Feb 95 08:19:58 PST
From: bil (To: cmdist)
To: cmdist
Subject: clm news, request for exotic sounds
The new clm no longer tries to find cmus.h -- the latter was
causing nothing but confusion. Also many new header types
supported, but I need test cases -- if anyone has examples
of alaw data, the newer IRCAM headers (any data format), or
AVR, SPPACK, PSION, or MAUD headers, I'd greatly appreciate
a copy. Various bugs in the clm-within-stella-in-gcl load
process fixed, I hope.
Date: Tue, 28 Feb 95 09:17:13 PST
From: bil (To: cmdist)
To: cmdist
Subject: clm news
I've added singer.ins, an implementation of Perry Cook's
physical model of the vocal tract. And by popular request,
the hidden factor of 2 has been removed from the expansion
calculation in make-expand -- I can't remember why it was
there to begin with.
Date: Wed, 1 Mar 1995 09:17:53 +6000
From: Joan Ordinas <jordinas@xtec.es>
Subject: XMIDI distribution
To: Common Music Users <cmdist@ccrma.Stanford.EDU>
Cc: Xavier Serra <xavier@phonos.upf.es>
Dear friends,
Thanks to the usefull service that David Betz has done to
the informatic comunity by elaborating the XLISP language,
some years ago I implemented a tool, which I hope will
serve to potenciate and dinamize the research in computer
assisted musical composition. As a matter on fact, I have
added to XLISP 2.1 a docen of new functions that allow the
management of MIDI messages in real time. For this purpose I
have developed an extense library in C language that solves
the low level problems inherents to the programing of
interface MIDI Roland MPU-401 in IMB PC computer.
You can found the new version of XLISP, which I have named
XMIDI in order to distinguish it from the conventional
XLISP, with the URL:
ftp://ftp.xtec.es/pub/musica/xmidi.zip
This version acompanies the interpret with a variated
selection of LISP functions, most of them compatibles with
COMMON LISP, as well as with examples of MIDI functions,
and a simple editor of public use, which maturally can be
replaced by another, according to the user preferences.
Three functions that permit the new uses the edition of
programs and the on-line requests about the documentation of
the language, (ED), (DOC) and (XDOC), will without a doubt
be of great help to work with XLISP and thus, will
contribute to its diffussion.
I would apreciate you to spread the distribution of XMIDI by
the channels you consider most apropiate.
For your interest, I list the new MIDI functions and variables:
!
*MPU-EXIST*
*RESOLUTION*
ACCEPTABLE-CHANNELS
COPY-MESSAGE
CURRENT-TIME
DOC
DOWNLOAD
ED
FLUSH-INPUT
INPUT
INPUT-TIME
MAKE-MESSAGE
MESSAGE-CHANNEL
MESSAGE-DATA1
MESSAGE-DATA2
MESSAGE-P
MESSAGE-TYPE
PENDING-EVENTS
PRINT-MESSAGE
REAL-TIME
RECEIVE-MESSAGE
SEND-EXCLUSIVE
SEND-MESSAGE
SEND-NOTE
STOP
UPLOAD
WAIT
XDOC
+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
| Joan Josep Ordinas Rosa | Administrador de sistemes |
| jordinas@xtec.es | Programa d'Informatica Educativa |
+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
From: Rick Taube <hkt@guido.zkm.de>
Date: Mon, 20 Mar 95 14:16:42 GMT+0100
To: Fernando Izuzquiza <doenado@phonos.upf.es>
Subject: Re: numbers
Cc: cmdist@ccrma.Stanford.EDU
>Can somebody give me the number address of the following sites? :
>* ccrma-ftp.stanford.edu
36.49.0.93
> * guido.zkm.de
192.101.28.17
but note that when your nameserver is working, you should not use guido.zkm.de but rather ftp.zkm.de. id recommend waiting a day or two before ftp'ing cm -- we are just a few bugs away from a new release, ill post to cmdist when its done.
hkt
From: Rick Taube <hkt@guido.zkm.de>
Date: Fri, 24 Mar 95 14:50:22 GMT+0100
To: cmdist@ccrma.Stanford.EDU
Subject: New release of Common Music
Cc: algo-comp@heinous.isca.uiowa.edu, ram+@cs.cmu.edu,
a0156@rs1.rrz.uni-koeln.de, hkt@guido.zkm.de
I've placed a new release of Common Music on ccrma-ftp.stanford.edu
and ftp.zkm.de. The major new additions are:
o Three new ports in public domain lisps:
1) NeXT/Intel in GCL-1.1, includes MIDI real time.
2) SGI/IRIX5.3 in CMU Common Lisp
3) SGI/IRIX5.3 in CLISP
o Capella GUI on Macintosh has been improved and highly optimized:
1) New envelope editor supports various output formats, copy/paste.
2) Listing browsers now work well with longer (2000+) sequences.
3) Controls completely reimplemented to work better with color.
4) New "Play..." item in MIDI submenu for playing Midifiles.
5) New "Move" Worksheet for positional editing: swapping, rotating,
shuffling, reversing, and so forth.
o New :MIX syntax for Oyvind Hammer's (oyvindha@notam.uio.no)
realtime mixing program "mix" on the SGI, implemented by Anders
Vinjar (anders.vinjar@notam.uio.no). Get mix from
notam.uio.no:/pub/sgi.
o README's completely rewritten and reorganized.
o To accommodate the ever increasing number of ports, the build script
has undergone some simplification and the build process has changed
slightly. Here are the incompatible changes:
1) For MIDI on NeXTstep anywhere, you must install MusicKit_4.1.1
2) Nextstep 2.0 is no longer supported.
3) AKCL is no longer supported (use GCL instead).
4) The lisp build program no longer calls unix "make" to
automatically compile c programs. You get to do this
yourself, see the new main README for further information.
-hkt
From: Rick Taube <hkt@guido.zkm.de>
Date: Sat, 25 Mar 95 12:31:24 GMT+0100
To: cmdist@ccrma.Stanford.EDU
Subject: tut mir leid...
...but i made two archiving mistakes yesterday. stop reading if you didnt get cm.sea.bin or gcl-next.tar.gz yesterday. if you got gcl-next.tar.gz, please get it again -- i forgot to update the NeXT.defs file for the MusicKit_4.1.1 midi lib. if you got cm.sea.bin or cm.sea.hqx, i deleted code that makes a cross-hair cursor in the new envelope editor. and the world kept spinning! but since there is no point to a graphic envelope editor unless its spiffy, here is the patch. either put it in in a file called cminit.lisp, or replace the current defmethod in capella:mac:enved.lisp then recompile that file then do build.lisp again. the archives on ftp.zkm and ftp-ccrma have already been updated.
(in-package :stella)
(defmethod ccl:view-cursor ((view plot-view) point)
(declare (ignore point))
(if (slot-value view 'guide?)
(cross-hair-cursor)
ccl:*arrow-cursor*))
From: Rick Taube <hkt@guido.zkm.de>
Date: Thu, 30 Mar 95 09:58:49 GMT+0100
To: cmdist@ccrma.Stanford.EDU
Subject: cmpatches.lisp
Cc: hkt@guido.zkm.de
If you use a mac, the file cmpatches.lisp on ftp.zkm.de and ftp-ccrma.stanford.edu contains some fixes to the last release. transfer using ftp's ascii mode. see the top of the file for instructions on how to auto load.
Fixes:
1 copy/paste in listing browser works again.
2 expunging one algorithm doesnt expunge them all.
3 changing the class of an object updates the display.
4 lambda icon correctly activates for algorithms again.
5 clicking on Inspect button in an Info window now brings up mcl's inspector.
Date: Mon, 3 Apr 95 10:24:31 PDT
From: bil (To: cmdist)
To: cmdist
Subject: CLM/CMN news
The new CLM may not be entirely backwards compatible -- I changed
the way booleans are handled, so if your instruments stop working,
recompile them. Also read-position is now setf-able. CLM can
be built without any FFI, but it's slow! This made it possible to
port it to Clisp. New interpolating comb filter-related generators
in zd.ins, more examples of expand in expsrc.ins.
CMN runs under Linux (and gcl).
Date: Fri, 2 Jun 95 09:24:38 GMT-0700
From: bil (To: cmdist)
To: cmdist
Subject: CLM news
I've added bessel.lisp supporting bessel functions in run (in C and
Lisp). Also since the last message, parallel instruments work now
in Lisp; a better example of using these (implementing sharable
vibrato streams) is in vibstr.cl and vibexamp.cl; various new
byte and array oriented Lisp functions added to C-side run support;
"generic functions" added for easier access to various unit
generator internal fields (i.e. frequency, phase, channel, data,
size-of, read-position, read-forward, read-backward, and
read-direction), all setf-able; finally, if you're using stella
and loop and notice that CLM instruments are not working correctly,
load CLM's version of loop instead -- this will be fixed in
later versions of stella/clm.
Date: Fri, 23 Jun 95 14:24:45 +0100
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: clm/cm on SGI
Hello,
I'm trying to use CLM/CM on the SGI (Indy or Indigo2)
I use it normally on Mac and NeXT(black hard.).
Could somebody tell me wich files do I need to install it succesfully on an SGI and where to find them? Many, many thanks on advance, friends & coleagues.
Doenado, el Ur
doenado@phonos.upf.es
hronhir@harrison.upf.es
Date: Fri, 23 Jun 95 06:27:18 GMT-0700
From: bil
To: Doenado el Ur <doenado@phonos.upf.es>
Subject: Re: clm/cm on SGI
Cc: cmdist
We're in the process of clearing out the dry-rot from a previous
port of CLM to the SGI in ACL, and I hope to have it running
very soon (especially since I recently bought an SGI). CMN
runs ok, I think (in ACL 4.2). As soon as CMUCL (or GCL?) are
running on the SGI, I'll tackle porting to them. I'll send
cmdist a note when this is in a usable state.
Date: Sat, 24 Jun 95 18:43:54 +0100
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: SGI clm/cm (II)
Cc: doenado@phonos.upf.es
Hello,
Bill Schottstaedt told me about the actual general state of clm for SGI, but reffering to the ACL version (thank you very much again). But, in /clm/README.clm file, there is a text explaining how to install clm under ACL and under GCL :
I would like to know if clm is now really running under GCL on the SGI or not, because I think that GCL is free (is that thru?), and I do not have a release of ACL.
Thank you in advance,
Date: Sat, 24 Jun 95 12:18:02 -0700
From: <tkunze@ccrma.Stanford.EDU>
To: Doenado el Ur <doenado@phonos.upf.es>
Subject: Re: SGI clm/cm (II)
Cc: cmdist@ccrma.Stanford.EDU
> I would like to know if clm is now really running under GCL
> on the SGI or not, because I think that GCL is free (is that
> thru?), and I do not have a release of ACL.
currently, gcl-2.1 runs on sgi machines under irix 5.2 and
higher, but without a possibility of saving dynamically
loaded images. Clm, cm and cmn (?), however, seem to compile
fine under gcl-2.1 and it is expected that LINKING their
binaries with gcl (as opposed to using the lisp snapshot
facility) would equally yield an executable, but it hasn't
been tested so far. The bottom line thus reads like: I am
not aware of anyone having a working cm, clm or cmn image
that's been built under gcl on a sgi.
Some folks here are getting sgi's these days, so things might
well change/improve soon...
Date: Tue, 25 Jul 95 08:43:56 PDT
From: bil (To: cmdist)
To: cmdist
Subject: CLM on SGI
CLM now runs on the SGI, Irix 5.3, in GCL-2.1 and ACL 4.2.
run* added to make it easier to get results back from the
run loop. filtering and sampling-rate conversion added to
fasmix. Various obscure NeXT-related functions removed.
Date: Tue, 8 Aug 95 10:13:04 -0700
From: <hkt@ccrma.Stanford.EDU>
To: cmdist@ccrma.Stanford.EDU
Subject: new cm release
Ive placed a new release of Common Music on ftp.zkm.de and ccrma-ftp.stanford.edu.
Lots of new stuff: a fast, automatic interface to CLM definstruments, MIDI processing optimized, 1 and 2D cellular automata, tendency masking and some new non-linear random number generators, new composition utility functions like defmode, rescale, with-past-values, with-vars-snapshotted. stella.rtf tutorial revamped. See changes.text for more info.
-hkt
>From hkt@ccrma.Stanford.EDU Tue Aug 8 14:25:33 1995
Date: Tue, 8 Aug 95 14:25:04 -0700
To: cmdist@ccrma.Stanford.EDU
Subject: cm release and GCL
If you ftp'ed the newest release of cm and you use GCL please ftp it again. There were several problems in build.lisp that affected building CM in an image with PCL already installed. Sorry!
-hkt
Date: Thu, 17 Aug 95 12:15:03 -0700
From: John Candlish <candlish@oregon.uoregon.edu>
To: cmdist@ccrma.Stanford.EDU
Subject: where to find pcl-gcl-1.1.tar.gz ??
I had a copy of this, but now I can't find it.
Thanx
jCandlish
Date: Fri, 18 Aug 95 07:10:44 GMT-0700
From: bil (To: cmdist)
To: cmdist
Subject: Re: where to find pcl-gcl-1.1.tar.gz ??
I put a copy on ccrma-ftp:/pub/Lisp.
Date: Fri, 18 Aug 95 23:13:56 +0100
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: CM17Aug95-install-failed
Cc: doenado@phonos.upf.es
Trying to install the new cm under NeXT (3.2) and ACL (3.1) I had 2 problems:
1. Before compiling, I tryed to make the midi object (as recommended in README.ACL) and no ".o" was created, and then later, when compiling, I had a message saying I forgot to compile that midi object. I removed MIDI syntax in build.lisp to see if there was any other problem and then... :
2. The compilation stoped here:
Fast loading /dist/Lisp/cm-new/stella/edit.fasl.
; Fast loading /dist/Lisp/cm-new/stella/select.fasl.
; --- Compiling file /dist/Lisp/cm-new/stella/algo.lisp ---
; Compiling MAKE-OBJECT
; Compiling OBJECT
; Compiling EXPAND-CONTAINER
; Compiling THREAD
; Compiling HEAP
; Compiling MERGE
; Compiling ALGORITHM
; Compiling MUTE
; Compiling GENERATOR
; Compiling ADD-BINDING
; Compiling WRAP-ITEM
; Compiling WALK-ITEM
; Compiling WALK-ALGORITHM
Error: Symbol "*TOPLEVEL-ENVIRONMENT*" not found in the WALKER package. [file position = 10787]
[changing package from "USER" to "STELLA"]
[1] <cl>
----------------------------------------------------------------------
3. Also, I did not find the usual unix dir. for compiling the MusicKit things. It is no longer necessary to do what README.ACL recommends???
Had somebody the same experience?
Can somebody help?
Many thanks in advance.
Doenado, el Ur
PHONOS
Barcelona, Spain
Date: Fri, 18 Aug 1995 16:35:35 -0700 (PDT)
From: Richard Karpen <karpen@u.washington.edu>
To: Doenado el Ur <doenado@phonos.upf.es>
Cc: cmdist@ccrma.Stanford.EDU, doenado@phonos.upf.es
Subject: Re: CM17Aug95-install-failed
Yes, this is a problem! Rick Taube is aware of it but will be unavailable
I think for a couple of weeks, so if someone else knows
how to fix this.....? He suggested trying out the build.lisp
file in the cm directory at ccrma. Can someone at ccrma put
that on the ftp server?
Date: Sat, 19 Aug 95 06:44:03 GMT-0700
From: bil (To: cmdist)
To: cmdist
Subject: Re: CM17Aug95-install-failed
I put our local copy of cm's build/build.lisp on ccrma-ftp:/pub/Lisp --
if there's still a problem come Monday, I'll poke at it.
I believe walker:*toplevel-environment* can be replaced by nil
except in Clisp.
Date: Sat, 19 Aug 95 22:40:42 +0100
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: CMinst-failed(II)
Cc: doenado@phonos.upf.es
I tryed with build.lisp of CCRMA and nothing changes.
Thanks for the help.
Doenado el Ur
Date: Sat, 19 Aug 1995 14:52:09 -0700 (PDT)
From: Richard Karpen <karpen@u.washington.edu>
To: Doenado el Ur <doenado@phonos.upf.es>, hkt@ccrma.Stanford.EDU,
bil@ccrma.Stanford.EDU
Cc: cmdist@ccrma.Stanford.EDU
Subject: Re: CMinst-failed(II)
I have a fix to get CM to compile under NeXTStep using ACL,
at least until Rick is settled in and can make it "official"
Richard Karpen
****************************************************
Edit the file stella/algo.lisp
In the funtion walk-vars
change the line
walker:*toplevel-environment*
to
nil
In the function expand-algorithm
change the line
(walker:macroexpand-all
to
(nil
Date: Sun, 20 Aug 95 19:03:29 +0100
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: cm...
Cc: doenado@phonos.upf.es
Many thanks to Bill Schottstaedt and Richard Karpen for the fixings.
MIDI and MKit problems are still there, but I sopose that it's better to wait for hkt.
Thanx again,
Doenado, el Ur
From: Tobias Kunze <tkunze@radial.kgw.TU-Berlin.DE>
Date: Tue, 22 Aug 1995 11:35:45 -0600
In-Reply-To: John Candlish <candlish@oregon.uoregon.edu> "where to find pcl-gcl-1.1.tar.gz ??" (Aug 17, 12:15pm)
To: cmdist@ccrma.Stanford.EDU
Subject: Re: where to find pcl-gcl-1.1.tar.gz ??
> I had a copy of this, but now I can't find it.
>
> Thanx
> jCandlish
yes. pcl-gcl-1.1 is no longer on the net. and unfortunately,
gcl-2.1 does NOT run on the next. we have a copy on the
ccrma server, but I don't know how you (and maybe others
as well?) may access it. I'd contact nando@ccrma and ask
him what to do.
Date: Tue, 5 Sep 95 09:28:18 GMT-0700
From: bil (To: cmdist)
Subject: CLM news
I translated Perry Cook's physical modelling toolkit to CLM-ese:
prc-toolkit95.lisp -- has examples of various simple instruments.
scale.cl has a translation of the 1100 or so scales collected
by John Chalmers and Manuel Op de Coul. mat.c translates from
matlab data files to NeXT sound files. Various other headers
and data formats added. On the SGI and NeXT there's a new function
for recording: record. Added *clm-file-search-list* to make it
easier to deal with sound files on many directories.
From: Philip Lewis <lewispsc@sable.ox.ac.uk>
Subject: Capella
To: cmdist@ccrma.Stanford.EDU
Date: Sun, 10 Sep 1995 10:34:56 +0000 (BST)
I have been trying the Capella tutorial, and when I click to get the
Mix window, it only appears part way, and I get an error message:
** Break during event processing! Standin event processor created. ***
**** Original event processor will continue when break loop exits. ****
> Error in process Initial: value NIL is not of the expected type NUMBER.
> While executing: RESIZE-OUTPUT-PANE
I built a version of CommonMusic with changes.txt dated 21 June 1995
with MCL 3.0 on a PowerBook 520. There were a number of error messages
which I did not note, all of the general kind, as I remember, "Method
so-and-so already exists. Replace?" which I did. What should I do?
From: "Tobias Kunze" <tkunze@cmsgia.Stanford.EDU>
Date: Sun, 10 Sep 1995 15:55:38 -0700
To: Philip Lewis <lewispsc@sable.ox.ac.uk>
Subject: Re: Capella
Cc: cmdist@ccrma.Stanford.EDU
On Sep 10, 10:34am, Philip Lewis wrote:
> Subject: Capella
> [...]
The short story is: the version of Capella that's currently available
doesn't run yet under MCL 3.0 due to the vast amount of changes in the
class library and its new process scheme. We should have a new release
out the door soon, though, that fixes the mix/sequence worksheet problem
by replacing them with a comletely redesigned output/layout window.
I can't promise a date yet, however, for two reasons: (1) Stanford Software
Licensing apparently still hasn't got a site license for MCL 3.0 and
all I have is an expired beta version and (2) I'll be out of town for the
next 4 weeks...
In the meanwhile, I'd suggest to either use Mcl 2.0.1 for Common Music
or simply don't use the mix and sequence worksheets in 3.0.
You'll see an anouncement of the new release on this list as soon as
it's finished.
Date: Thu, 5 Oct 95 07:55:52 GMT-0700
From: bil (To: cmdist)
To: cmdist
Subject: CLM news
The new version of CLM supports n-channel IO and computed input
to the various generators that were file-oriented (src, resample,
etc). These changes are such that the new CLM is not entirely
backwards compatible -- when you update, it's probably safest to
delete all the CLM-related .fasl and .o files first. I also
changed some function names; primarily phase to angle, since
the name "phase" is causing confusion in some Lisps. See initmus.lisp
for a list of the changes.
Date: Sat, 07 Oct 1995 20:15:00 +0900
From: =?ISO-2022-JP?B?GyRCQ11DZjkvR24bKEI=?= <VYU00430@niftyserve.or.jp>
Subject: from Takenaka
To: cmdist-request@ccrma.Stanford.EDU
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-2022-jp
Date: Sat, 7 Oct 1995
To: Yasuhiro Takenaka<VYU00430@niftyserve.or.jp>
From: Heinrich Taube <cmdist-request@ccrma.stanford.edu>
************************************************************************
Dear Prof. Heinrich Taube,
According to your letter you described, MCL 3.0 sounds like unavailable
for Algo. Comp.
README in cm.sea.bin says that Common Music is known to work in the foll
owing Lisp ports:
MCL 2.0.1, MAC/System 7.5. Could you tell me what other version (MCL) is
available for Algo. Comp. and also how to get it? Now, I need MCL to wor
k Algo. Comp.
Thank you in advance for your help in answering these questions. Hoping
to hear from you soon.
Yours sincerely
TAKENAKA Yasuhiro
************************************************************************
Date: Fri, 13 Oct 95 11:42:43 -500
From: cweston@iastate.edu (Craig Weston)
To: cmdist@ccrma.Stanford.EDU
Subject: CLM on Mac/MCL
I'd really like to be able to run CLM on my mac (PPC 8100/80 w/ MCL
2.0.1).
Anyone who's doing it, I'd be grateful if we could e-mail privately and
talk about getting it running, performance, i/o, etc.
Thanks,
__________________________________________________________________
|Craig Weston--Assistant Professor of Music Theory, Composition, |
| & Electronic/Computer Music, Iowa State University|
| |
|e-mail: cweston@iastate.edu |
|WWW: http://www.public.iastate.edu/~cweston/homepage.html |
|________________________________________________________________|
From: "Tobias Kunze" <tkunze@cmsgia.Stanford.EDU>
Date: Sat, 14 Oct 1995 13:54:59 -0700
Subject: Re: CLM on Mac/MCL
Cc: cmdist@ccrma.Stanford.EDU
a quick review on how things are on the mac for now:
- mcl3 will go native in the first half of 96
- mcl will (or does already) support the linking of shared libraries
- metrowerks cw (and maybe others) will (or already does) support
the generation of shared libraries
given this, it might be feasible to port clm to the mac platform.
however (caution: my opinion), it'll still be a nightmare to develop
instruments. you'll have to
1) write (or edit) the lisp instrument code
2) let clm write a c file
3) open the c file in metrowerks
4) set options to compile it as a shared library (how many dialogs
is this??)
5) switch back to mcl
6) link the library
7) run with-sound
8) repeat steps 1-7 for ->EVERY<- bug or change you want to make
im not sure i have them all. there might be other steps to take.
of course, if someone talks digitool and metrowerks to support
appropriate apple events and is willing to write a lisp interface
to such a beast, it could be all done by a lisp script.
If you need only a couple of fixed instruments, however, all you'd
have to do is to link them once and have fun!
Date: Sun, 5 Nov 1995 08:54:14 -0800
From: CShehadi@eworld.com
To: cmdist@ccrma.Stanford.EDU
Subject: cm crashes
Hello...
I'm an intermediate-level C programmer trying to do project in Lisp with
Common Music.
I'm using System 7.5.1 on a PowerMac 6100/AV with Midi Manager version 2.0.2.
I'm using the minimum possible extensions to run Common Music.
I can get Common Music built on my machine using MCL 2.01 and it works ok for
a few hours but after a while it seems to get corrupted - or something.
I get unpredictable type 1 and type 3 crashes within 2 minutes of using
Common Music. I can't get anything to work in Common Music for any extended
period of time.
Unfortunately, because I'm new to Common Music and Lisp I really don't know
where to start to solve this problem.
Has anyone else had similar problems?
These are complete shots in the dark, but...
-Is there a chance that my hard disk driver could be conflicting with Common
Music?
I've run Norton Utilities and the disk checks out as ok - but there could
still be a conflict with the driver...or is this not possible?
-Could there be something in my code that corrupts Common Music? Or is CM
usually pretty indestructible?
-Is there a "support file" that Common Music uses that might get corrupted
after a crash, that I can easily replace without having to rebuild the whole
program?
Any help or advice as to where to look for a solution would be appreciated...
-Charlie Shehadi
cshehadi@eworld.com
Date: Sun, 5 Nov 1995 09:53:38 -0800 (PST)
From: Richard Karpen <karpen@u.washington.edu>
To: CShehadi@eworld.com
Cc: cmdist@ccrma.Stanford.EDU
Subject: Re: cm crashes
I run my PPC Macs with the "modern memory manager" off and virtual memory
off. Both of these are in the memory control panel. MCL and therefore
CM crash with these two "features" on.
From: "Tobias Kunze" <tkunze@cmsgia.Stanford.EDU>
Date: Sun, 5 Nov 1995 14:36:47 -0800
In-Reply-To: CShehadi@eworld.com
"cm crashes" (Nov 5, 8:54am)
References: <951105085403_17864545@hp1.online.apple.com>
X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail)
To: CShehadi@eworld.com, cmdist@ccrma.Stanford.EDU
Subject: Re: cm crashes
> I'm using System 7.5.1 on a PowerMac 6100/AV with Midi Manager version 2.0.2.
> I'm using the minimum possible extensions to run Common Music.
>
> I can get Common Music built on my machine using MCL 2.01 and it works ok for
> a few hours but after a while it seems to get corrupted - or something.
>
> I get unpredictable type 1 and type 3 crashes within 2 minutes of using
> Common Music. I can't get anything to work in Common Music for any extended
> period of time.
Hi, Charles-
MCL 2.0.1 dates back to the dark ages where ppc's weren't even around and
the IIfx the top-of-the-line product. You should thus make sure you've
loaded ALL available patches to the vanilla 2.0.1 image, especially the
MMU2,MMU3, MMU4, and MMU5 patch. I think you can still get them at
digitools ftp site ftp.digitool.com. unpack and follow the patch instructions.
ALso, make sure the modern memory manager is OFF in the memory control panel.
Once you've done that, everything should run fine.
[Type 1 errors are bus errors, Type 3 address errors, that is quite unspecific,
im afraid.]
> -Is there a chance that my hard disk driver could be conflicting with Common
> Music?
> I've run Norton Utilities and the disk checks out as ok - but there could
> still be a conflict with the driver...or is this not possible?
im almost 150% sure it doesn't have anything to do with it. unless your main
image is on a bad disk sektor or the like.
> -Could there be something in my code that corrupts Common Music? Or is CM
> usually pretty indestructible?
only if you call the os directly. then it is quite easy to crash the
machine. after all, it's still only a mac :)
> -Is there a "support file" that Common Music uses that might get corrupted
> after a crash, that I can easily replace without having to rebuild the whole
> program?
no. rest assured.
-Tobias
From: "Tobias Kunze" <tkunze@cmsgia.Stanford.EDU>
Date: Sun, 5 Nov 1995 14:41:40 -0800
In-Reply-To: Richard Karpen <karpen@u.washington.edu>
"Re: cm crashes" (Nov 5, 9:53am)
References: <Pine.OSF.3.91j.951105095140.15884A-100000@saul1.u.washington.edu>
X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail)
To: Richard Karpen <karpen@u.washington.edu>, CShehadi@eworld.com
Subject: Re: cm crashes
Cc: cmdist@ccrma.Stanford.EDU
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
On Nov 5, 9:53am, Richard Karpen wrote:
> Subject: Re: cm crashes
>
> I run my PPC Macs with the "modern memory manager" off and virtual memory
> off. Both of these are in the memory control panel. MCL and therefore
> CM crash with these two "features" on.
>
>-- End of excerpt from Richard Karpen
not right. virtual memory is fine (and needed ESPECIALLY on the
ppc unless you have 32 megs of RAM). just make sure you don't try
to cheat the system with gizmos like Ram Doubler or speed doubler
or brain doubler or ...
Date: Mon, 6 Nov 95 17:29:26 GMT
From: Doenado el Ur <doenado@phonos.upf.es>
To: cmdist@ccrma.Stanford.EDU
Subject: i*link Support
Cc: doenado@phonos.upf.es
Hello,
In a source file of CLM is suggested the possibility of supporting the i*link card (for the NS/Intel systems) from CLM (because i*link comes with with some NStep soft).
That would be usefull for many people. I encourage all you to do it (i don't know how@)
Doenado, el Ur
* * * * * * *
Date: Mon, 6 Nov 95 08:48:32 PST
From: bil (To: cmdist)
To: cmdist
Subject: Re: i*link Support for clm
>In a source file of CLM is suggested the possibility of supporting the i*link card (for the >NS/Intel systems) from CLM (because i*link comes with with some NStep soft).
>That would be usefull for many people. I encourage all you to do it (i don't know how@)
I just removed support for the ilink, Turtle Beach, and Ariel
PC56d cards from CLM because no one was using it; these cards
actually slow CLM down on any currently-available PC hardware;
that is, the synthesis is a lot faster on the main processor
than on the 56000. Once the QP cards die, I'll probably
throw away all the 56000 code in CLM. (Motorola recently
announced the availability of the 80 MHz 56002, but that's
still slow compared to any normal main processor, and even
when I get "real-time" support going, it will not use the
56000).
Content-Type: text/plain
Mime-Version: 1.0 (NeXT Mail 3.3 v118.2)
From: Leigh Smith <leigh@antechinus.cs.uwa.oz.au>
Date: Wed, 8 Nov 95 00:39:11 +0800
To: cmdist@ccrma.Stanford.EDU
Subject: Re: i*link Support for clm
References: <9511061648.AA29028@ ccrma.Stanford.EDU >
> Once the QP cards die, I'll probably
> throw away all the 56000 code in CLM. (Motorola recently
> announced the availability of the 80 MHz 56002, but that's
> still slow compared to any normal main processor, and even
> when I get "real-time" support going, it will not use the
> 56000).
>
Given that the Frankenstein box (8 56002's per card) has been demoed at ICMC,
won't there be some call for arrays of DSP/NSP processing, still? I understand that maintaining that duality of signal and main processor is a problem that may well weigh out against such parallel approaches in the long run, but gee, I'd like to be signal processing with a frankenstein card and Common Lisp in real time, soon... Besides, we can be assured that whatever processing power the main CPU posesses will be put into smoother user interfaces & graphics etc by our O.S. manufacturers and probably not left idle for us musicians :-) Perhaps the real-time Dig. Sign. uP support in the future is better left to the MusicKit/SynthBuilder type of program with high level OOP messaging to Lisp?
Really just thinking aloud
Leigh
---
Leigh Smith Computer Science, University of Western Australia
+61-9-380-1945 leigh@cs.uwa.edu.au (NeXTMail/MIME)
"In a world where success means gaining time, thinking has a single but
irredeemable fault: it's a waste of time" - Jean-Francois Lyotard
Date: Tue, 7 Nov 95 10:23:40 PST
From: bil (To: cmdist)
To: cmdist
Subject: Re: frankenstein and clm
As I understand it, this box has only analog output, and each dsp's
output is added to the output stream as an analog signal. This means
that the thing has to run "sample synchronous", unlike the QP board.
This limits the actual speed increase and the density, so to speak,
of the overall output. It will be a real improvement for MusicKit
users, but much less of one for CLM users. I believe it would be
only a week or so of work to add support for it in CLM, and may do
so as an experiment. More interesting, I think, are the various
multi-processor machines coming on the market that allow you to
split up a C program so that it can run on parallel cpu's -- this
kind of thing would be easy to add to CLM, and not require special
purpose assembly code or drivers (presumably).
Date: Tue, 7 Nov 95 12:09:50 -0800
From: david@jaffe.com (David A. Jaffe)
To: bil@ccrma.Stanford.EDU (Bill Schottstaedt)
Subject: Re: frankenstein and clm
Cc: cmdist@ccrma.Stanford.EDU
>As I understand it, this box has only analog output, and each dsp's
>output is added to the output stream as an analog signal.
This is how the current prototype works, but we all wants digital output, so that's in the plan for the "real" version. In that version, we will chain the DSP serial ports so that a set of DSPs can be made to run in series. In this case there's an implied delay between each DSP on the order of the sound buffer size (which we're planing to make smaller) and any DSPs that are digitally chained would be "sample synchronous" (actually "buffer synchronous").
>I believe it will be only a week or so of work to add support for it in CLM, and may do so as an experiment.
If you still have the old DSP support around, it should be trivial to support the Frankenstein in the same way you supported the other cards. If you're talking about real-time, it may be longer.
To: cmdist@ccrma.Stanford.EDU
Subject: Where can I find MidiQT?
Date: Fri, 10 Nov 1995 00:08:22 +0900
From: Seiji ADACHI <adachi@hip.atr.co.jp>
I am new to the common music environment (well, since this summer
workshop), and want to make my powerbook sound without any external
midi devices. Does anyone tell me where one can find MidiQT extension,
which should be used with QuickTime2.x. I could figure out the extension
is a shareware and should be somewhere in the net, but didn't find the
specific location.
Seiji
Date: Fri, 10 Nov 1995 00:01:26 +0800
From: "Gregory A. McCourt" <mccourt@technet.sg>
Apparently-To: cmdist@ccrma.Stanford.EDU
~s Re: i*link Support for clm
>
> >In a source file of CLM is suggested the possibility of supporting the i*link card (for the
> >NS/Intel systems) from CLM (because i*link comes with with some NStep soft).
> >That would be usefull for many people. I encourage all you to do it (i don't know how@)
>
> I just removed support for the ilink, Turtle Beach, and Ariel
> PC56d cards from CLM because no one was using it; these cards
> actually slow CLM down on any currently-available PC hardware;
> that is, the synthesis is a lot faster on the main processor
> than on the 56000. Once the QP cards die, I'll probably
> throw away all the 56000 code in CLM. (Motorola recently
> announced the availability of the 80 MHz 56002, but that's
> still slow compared to any normal main processor, and even
> when I get "real-time" support going, it will not use the
> 56000).
i agree. clm should be made totally independent of spacific dsp chips
so that it can be made more portable to different platforms. this
would also result in a cleaner and more compact codebase to
maintain and build on.
greg
G r e g o r y A . M c C o u r t <mccourt@technet.sg>
fax: + 6 2 . 2 7 4 . 3 7 9 7 8 0
P u s a t S t u d i I n d o n e s i a / F a k u l t a s S a s t r a
U n i v e r s i t a s G a d j a h M a d a
B u l a k s u m u r , Y o g y a k a r t a 5 5 2 8 1
I n d o n e s i a.
Date: Thu, 9 Nov 1995 11:10:35 -0800
From: tkunze@cmsgia.Stanford.EDU (Tobias Kunze)
To: Seiji ADACHI <adachi@hip.atr.co.jp>
Subject: Re: Where can I find MidiQT?
Cc: cmdist@ccrma.Stanford.EDU
I use QuickTime MIDI Player, which is an app instead of an init
and thus less errorprone. Here's the readme
--------------------------------------------------------
QuickTime MIDI Player 1.1
Copyright © 1995 by Terry Greeniaus
Many thanks to Kevin Patfield for his insightful comments and suggestions on how to improve QuickTime MIDI Player. Anyone else who has ideas, feel free to send me e-mail. My address is printed at the bottom of this file and in the title screen and About
QuickTime MIDI Player is an application which will allow you to play MIDI data through your Macintosh speakers using QuickTime 2.0 and the MIDI Manager. To use QuickTime MIDI Player, you will require both QuickTime 2.0 and the Apple MIDI Manager. This s
QuickTime MIDI Player allows you to set up the way that the instruments will sound on the Macintosh. Each different instrument is attached to a track. You can have up to 100 different tracks, although you probably won't be able to play that many instrum
[...]
Shareware fees:
QuickTime MIDI Player is shareware and if you use it, please send in your registration fee of $15.00, so that you will recieve future upgrades and I will be able to contrinue supporting this program. I know that peaople cannot always afford the shareware
Terry Greeniaus
89 Lancaster Cres.
St. Albert, Alberta
Canada
T8N 2N9
Disclaimer:
QuickTime MIDI Player is distributed "as is." I make no guarantees or warranties on the quality of the product and I cannot be held liable for any problems that this software may create. This program has not been tested on any other computer but my Maci
QuickTime MIDI Player may be distributed freely, as long as this documentation is included with the file. It may not be sold individually, but may be placed on CD's, disks, BBS's, etc. where there is no extra charge required to aquire this file other tha
I can also be reached through e-mail at tgreen@ersys.edmonton.ab.ca
Terry Greeniaus
June 10, 1995
Date: Mon, 20 Nov 95 09:57:58 -0600
From: Heinrich Taube <hkt@cmp-nxt.music.uiuc.edu>
To: cmdist@ccrma.Stanford.EDU
Subject: new cm release (finally...)
I've placed a new release of Common Music on ccrma-ftp and ftp.zkm.de. This release introduces some major new functionality and ports (MCL 3.0 and CLISP 1995-08-12). Ive tested as best I can but there may still be a few bugs lurking. I also redesigned the lower level stream interface, the result is that there is now no need to distinguish between writing files and "listening" -- in fact Listen etc are no longer needed. The Open commnd now remembers the current output stream and commands Mix and Seq send output to the current stream, no matter what it is.
Ive started saving cm.zip files for DOS again, unfortunately I dont have a DOS machine to test on so let me know if there are any problems...
The main points:
1 Ports to MCL 3.0 and CLISP 1995-08-12. The CLISP port includes real time MIDI support on NextStep (thank you Leigh Smith!)
2 New script for building the system (Tobias Kunze). Load build/make-cm.lisp and answer the questions. The script will save your setting as a local configuration file for future builds. See the new build/README for more info.
3 Many new features in Macintosh graphical interface. Main ones: (a) New main menu and submenus make it easy to create objects and output streams. (b) COMMAND-U keystroke sends musical output to the current output stream from any browser or listing selection. (c) New layout editor for designing output arrangements (Tobias Kunze). Load capella.tutorial to learn how to use it.
4 Output in Stella is much simpler. The OPEN command now makes the last opened output stream the default stream for processing. New commands Mix and Seq operate on the open stream. Mix and Seq replace Listen, Write, SListen, SWrite etc. Avoid using Listen and SListen, they are no longer needed and will go away in a future release.
5 $slot notation no longer needed in expressions.
6 Documentation, tutorials and READMEs updated.
-hkt
Date: Wed, 22 Nov 95 11:07:53 -0600
From: Heinrich Taube <hkt@cmp-nxt.music.uiuc.edu>
To: cmdist@ccrma.Stanford.EDU
Subject: Re: cm for MCL 3.0
there are few problems in the port to mcl 3.0 (i usd a beta release of mcl3.0 product and there are some minor differences between that and the real mcl that digitool delivers.)
hold off building in mcl 3.0 for a few days until i can get the problem(s) straightened out.
-rick
Date: Thu, 23 Nov 1995 10:36:50 GMT
From: anders.vinjar@notam.uio.no
To: cmdist@ccrma.Stanford.EDU
Subject: pathnames in new cm-dist
Reply-To: Anders Vinjar <anders.vinjar@notam.uio.no>
Mime-Version: 1.0
Hi,
for those of you wanting to use the pathnames constructor supposed to be
in the latest cm distribution, there are unfortunately some glaring
ommisions in the code which you'll have to fix.
Until the next distribution of cm comes along, you can find a corrected
version of the affected files, as well as the missing docs at the
following place:
ftp://notam.uio.no/pub/temp/paths.tar.gz
-anders
Date: Thu, 23 Nov 95 15:05:09 -0600
From: Heinrich Taube <hkt@cmp-nxt.music.uiuc.edu>
To: cmdist@ccrma.Stanford.EDU
Subject: new cm archives - fixes and csound improvements
Cc: hkt@cmp-nxt.music.uiuc.edu
Ive placed new archives on ccrma-ftp and ftp.zkm.de. These fix a few problems in MCL3.0p2 and include new support for Csound and CMN.
Fixes:
o Cm now builds in MCL3.0p2 without an error in low.lisp.
o Help menu now finds stella command help texts.
o Capella tutorial loads in 8 seconds rather than 35.
o Mix and Seq now pass play options along to post-processing routines.
o Pathnames item stream now works and is documented in dictionary.rtf
Csound Improvements:
o New note class CSound-Note implements automatic score file output for its subclasses. If your note object inherits from csound-note you dont have to write any output methods. See the updated stella/examples/i1.lisp for more information.
o Importing .sco files now works. Only straight scores at the moment -- no dot-pfields, no octcps statements, no nothing, just blanco absolute values. (Thank you Anders Vinjar.) See updated Csound entry in dictionary.rtf for import options.
CMN improvements:
o Post-processing routines now use cm/site/showcmn to display .eps files. Showcmn is currently set up for NextStep.
Date: Sat, 02 Dec 1995 18:47:26 -0400
From: Charles Shehadi <cshehadi@panix.com>
X-Mailer: Mozilla 2.0b2 (Macintosh; I; PPC)
Mime-Version: 1.0
To: cmdist@ccrma.Stanford.EDU
Subject: some small troubles
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Hello all...
I'm having some trouble with my MCL/Common Music code...
I'm pretty sure there are some simple solutions to these problems, but =
I don't know enough LISP yet to solve them...
The following method definition is supposed to draw the contents of a =
"tool palette" window for an application I'm writing. ("tool-palette" =
is a subclass of the MCL class "windoid"). =
**The method doesn't cause an error in MCL 3.0p2:
(defmethod view-draw-contents :before ((tplt tool-palette))
(let ((pict-hdl (slot-value tplt 'pict-hdl)))
(when pict-hdl
(rlet ((rect :rect
:topleft #@(0 0)
:bottomright #@(49 241)))
(#_DrawPicture pict-hdl rect)))))
**When I try to evaluate it in Common Music. I get two errors:
I think they involve the following lines of code:
**Error number 1: =
(#_DrawPicture pict-hdl rect)))))
> Error: File "ccl:interfaces;index;traps.idx" does not exist.
> While executing: CCL::IF-DOES-NOT-EXIST
> Type Command-. to abort.
See the Restarts=8A menu item for further choices.
I have a file called traps.idx in my MCL:Libraries:interfaces;index =
folder. Can I just copy that file to a location in the Common Music =
folder, or is fixing this problem more involved than that?
**Error number 2:
(rlet ((rect :rect
:topleft #@(0 0)
:bottomright #@(49 241)))
> Error: While compiling (VIEW-DRAW-CONTENTS BEFORE (TOOL-PALETTE)) :
> #1=3D(RECT :RECT :TOPLEFT 0 :BOTTOMRIGHT 15794225) is not a symbol=
or lambda expression in the form (#1#) .
> Type Command-. to abort.
See the Restarts=8A menu item for further choices.
I'm not sure what to make of this....it works in MCL but not in Common =
Music? Are some files I should "require" or something like that?
Thanks in advance for any help...
-Charlie
From: "Tobias Kunze" <tkunze@cmsgia.Stanford.EDU>
Date: Sat, 2 Dec 1995 17:47:12 -0800
In-Reply-To: Charles Shehadi <cshehadi@panix.com>
"some small troubles" (Dec 2, 6:47pm)
References: <30C0D77E.72E0@panix.com>
X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail)
To: cmdist@ccrma.Stanford.EDU, Charles Shehadi <cshehadi@panix.com>
Subject: Re: some small troubles
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
[1] your logical pathnames are messed up. the "ccl:interfaces;" path
doesn't lead to your actual index path. see the file site:pathname-
translations.lisp
[2] the right call to rlet would be
(ccl:rlet ((:rect r :topleft #@(0 0) :bottomright #@(49 241)))
...)
or the like. check the mcl docs.
Date: Thu, 7 Dec 95 14:39:54 -0600
From: Heinrich Taube <hkt@cmp-nxt.music.uiuc.edu>
To: cmdist@ccrma.Stanford.EDU
Subject: Notice to PCI PowerMac MCL users
Cc: brp@sound.music.mcgill.ca
This message will be of interest for anyone running CM on a Power PC. (As far as I know the actual PPC release of MCL 3.0 is still scheduled for early '96.)
-hkt
---
Begin forwarded message:
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 7 Dec 1995 13:48:02 -0500
To: announce-mcl@digitool.com
From: slh@digitool.com (Stephen L. Hain)
Subject: Notice to PCI PowerMac MCL users
In the December Developer CD, Apple provides a control panel that turns
off the new dynamic emulator on PCI Power Macs. Doing so improves the
performance of MCL 3.0, MCL 2.0 and presumably the forthcoming Apple Dylan
Technology Release. You'll find a copy of it on the Digitool FTP site
at ftp.digitool.com/pub/mcl/patches/ as Emulator.bin or Emulator.hqx. The
README file there has been updated with a note about the utility.
- Steve Hain
Digitool, Inc.
______________________________________________________________________________
One Main Street 7th Floor Cambridge, MA 02142 USA
Internet: slh@digitool.com AppleLink: digitool
World Wide Web: http://www.digitool.com/
Tel: 617 441-5000 Fax: 617 576-7680
Date: Fri, 8 Dec 1995 16:08:53 -0700
From: karl@index.uwyo.edu (Karl Branting)
To: cmdist@ccrma.Stanford.EDU
Subject: Installation under ACLPC v2.0 and CLISP (DOS)
I found it easy to install common music under Allegro CL 4.2 on an
SGI, but I'm having difficulty installing it on a PC under either
clisp (dos) or aclpc v2.0. In attempting to run make-cm.lsp in aclpc I
found that QUERY caused a string conversion error. I got around this
problem by explicitly assinging values to each variable (e.g., bd?,
etc.), but then I get an enigmatic aclpc error while compiling tl.lsp:
Function #<function 2 #x85F4957C> called with incorrect number of
arguments in #<x85F4952C>
In attempting to execute make-cm.lsp under clisp, I find that
HANDLER-BIND (which is used in loading loop) is undefined. If I
comment it out and load loop.lsp by hand, I get the following error in
loading tl.fas:
#<CLOS:STANDARD-METHOD ...># has &REST or &KEY but #<GENERIC-FUNCTION
EDIT-OBJECT> hasn't.
I notice that there has been some reference to CLISP 1995-08-12 in
cmdist. However, the CMU AI archive's version seems much old than this.
In any event, I would be very grateful for any suggestions from anyone
who has recently succeeded in installing common music under aclpc v2.0
or clisp (dos). Has anyone installed common music under aclp v3.0?
Thanks,
Karl Branting
==========================================================================
Department of Computer Science Internet: karl@index.uwyo.edu
University of Wyoming Tel: (307) 766-4258
P.O. Box 3682 FAX: (307) 766-4036
Laramie, WY 82071, USA URL: http://ai.uwyo.edu/~karl
==========================================================================
Date: Sat, 9 Dec 95 09:16:30 -0600
From: Heinrich Taube <hkt@cmp-nxt.music.uiuc.edu>
To: karl@index.uwyo.edu (Karl Branting)
Subject: Re: Installation under ACLPC v2.0 and CLISP (DOS)
Cc: cmdist@ccrma.Stanford.EDU
I think some of the problems are due to not using the latest version of CLISP. You can get the DOS binaries from
http://sayre.sysc.pdx.edu:8001/clisp
or
ftp://ma2s2.mathematik.uni-karlsruhe.de:/pub/lisp/clisp
It seems that the first address always has the very latest versions and that the second address is updated from time to time. If you use the http address take binaries/dos.zip dated 01-oct-95.
I will mail you a newer version of the make-cm.lisp script in a subsequent message to you personally. Once I know it works for DOS i will add it to a patch file i hope to release in a few days.