From achim.bornhoeft at web.de Wed Jul 4 03:01:03 2007 From: achim.bornhoeft at web.de (Achim Bornhoeft) Date: Wed, 04 Jul 2007 12:01:03 +0200 Subject: [CM] CLM style warnings on Intel Message-ID: <468B6FDF.6010300@web.de> Hello, when compiling CLM Instruments on an Intel OS X 10.4x Machine I got several style warnings. As an example I attached a simple sine instrument with its corresponding output. While compiling sometimes I got strange EOF errors with files coming from a PPC. It helped just to copy the content to a new file ... Achim -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: sine.ins URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: sine-warnings.txt URL: From bil at ccrma.Stanford.EDU Wed Jul 4 03:32:22 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 4 Jul 2007 03:32:22 -0700 Subject: [CM] CLM style warnings on Intel In-Reply-To: <468B6FDF.6010300@web.de> References: <468B6FDF.6010300@web.de> Message-ID: <20070704102415.M82180@ccrma.Stanford.EDU> This is from sbcl? Two of the "warnings" are that I'm using *name* for a lexical variable -- I say it is none of lisp's goddamn business what variable names I choose, and that sbcl should be fixed. The other has to do with eval-when -- I'll check whether it is needed in sbcl. I think you can turn off these messages with something like: #+sbcl (setf *compile-print* nil) #+sbcl (setf *compile-verbose* nil) #+sbcl (declaim (sb-ext:muffle-conditions sb-ext:compiler-note)) From achim.bornhoeft at web.de Wed Jul 4 05:58:42 2007 From: achim.bornhoeft at web.de (Achim Bornhoeft) Date: Wed, 04 Jul 2007 14:58:42 +0200 Subject: [CM] CLM style warnings on Intel In-Reply-To: <20070704102415.M82180@ccrma.Stanford.EDU> References: <468B6FDF.6010300@web.de> <20070704102415.M82180@ccrma.Stanford.EDU> Message-ID: <468B9982.5020805@web.de> I am running CLM with OpenMCL inside the CM.app (cm-2.10.0-app-osx-intel) in Aquamacs Emacs 1.0b. Achim Bill Schottstaedt schrieb: > This is from sbcl? Two of the "warnings" are that I'm using *name* for > a lexical variable -- I say it is none of lisp's goddamn business what > variable names I choose, and that sbcl should be fixed. The other > has to do with eval-when -- I'll check whether it is needed in sbcl. > I think you can turn off these messages with something like: > > #+sbcl (setf *compile-print* nil) > #+sbcl (setf *compile-verbose* nil) > #+sbcl (declaim (sb-ext:muffle-conditions sb-ext:compiler-note)) > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From taube at uiuc.edu Wed Jul 4 06:09:20 2007 From: taube at uiuc.edu (Rick Taube) Date: Wed, 4 Jul 2007 08:09:20 -0500 Subject: [CM] CLM style warnings on Intel In-Reply-To: <468B9982.5020805@web.de> References: <468B6FDF.6010300@web.de> <20070704102415.M82180@ccrma.Stanford.EDU> <468B9982.5020805@web.de> Message-ID: <48AE3A51-6844-4605-90BD-A6622842A51F@uiuc.edu> the lisp you sent the warnings from is sbcl, openmcl isnt working yet on the intel macs as far as i know. this is what I do to keep from hurling my computer out the window when i work in sbcl: (declaim (sb-ext:muffle-conditions style-warning sb-ext:compiler-note)) On Jul 4, 2007, at 7:58 AM, Achim Bornhoeft wrote: > I am running CLM with OpenMCL inside the CM.app (cm-2.10.0-app-osx- > intel) in Aquamacs Emacs 1.0b. > > Achim > > Bi From achim.bornhoeft at web.de Wed Jul 4 06:20:38 2007 From: achim.bornhoeft at web.de (Achim Bornhoeft) Date: Wed, 04 Jul 2007 15:20:38 +0200 Subject: [CM] CLM style warnings on Intel In-Reply-To: <48AE3A51-6844-4605-90BD-A6622842A51F@uiuc.edu> References: <468B6FDF.6010300@web.de> <20070704102415.M82180@ccrma.Stanford.EDU> <468B9982.5020805@web.de> <48AE3A51-6844-4605-90BD-A6622842A51F@uiuc.edu> Message-ID: <468B9EA6.9080302@web.de> It was not my machine on which I compiled the instruments. I simply assumed it was OpenMCL ... Thanks for your help. Achim Rick Taube schrieb: > the lisp you sent the warnings from is sbcl, openmcl isnt working yet on > the intel macs as far as i know. this is what I do to keep from hurling > my computer out the window when i work in sbcl: > > (declaim (sb-ext:muffle-conditions style-warning sb-ext:compiler-note)) > > > > > > > On Jul 4, 2007, at 7:58 AM, Achim Bornhoeft wrote: > >> I am running CLM with OpenMCL inside the CM.app >> (cm-2.10.0-app-osx-intel) in Aquamacs Emacs 1.0b. >> >> Achim >> >> Bi > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From error at balticom.lv Thu Jul 5 01:15:34 2007 From: error at balticom.lv (UnknownError) Date: Thu, 05 Jul 2007 08:15:34 +0000 Subject: [CM] alsa problem .. In-Reply-To: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> Message-ID: <1183623335.7571.20.camel@00110101> Hello. i'm using snd with hemmerfall-dsp ALSA driver (linux-2.6.20-gentoo-r8) alsa-lib version is 1.0.14rc2, this comes with the current stable gentoo portage. i have seen the same problem reported earlier on this list , that was with Snd version 8.5; i just compiled the version 9.1 and get this: :;status of hw:0 hw_params status of hw:0 ACCESS: MMAP_NONINTERLEAVED RW_NONINTERLEAVED FORMAT: S32_LE SUBFORMAT: STD SAMPLE_BITS: 32 FRAME_BITS: [128 384] CHANNELS: [4 12] RATE: [32000 192000] PERIOD_TIME: (333 256000] PERIOD_SIZE: [64 8192] PERIOD_BYTES: [2560 393216] PERIODS: 2 BUFFER_TIME: (666 512000] BUFFER_SIZE: [128 16384] BUFFER_BYTES: [2560 786432] TICK_TIME: 1000 sw_params status of hw:0 start_mode: DATA xrun_mode: STOP tstamp_mode: NONE period_step: 0 sleep_min: 0 avail_min: 0 xfer_align: 0 silence_threshold: 0 silence_size: 0 boundary: 0 ;Invalid argument: hw:0: access type RW_INTERLEAVED not available i looked, just in case., if the superuser gets no error, as there might be a 'permission denied' i a way .. no, it's not at all - the root gets the same. the person meantioning the same earlier had a different soundcard and that was more then 6 month ago :( very strange , i'd expect that's due to some change in ALSA, but doesn't seem so. well, may be, still .. is there a way to change the ACCESS type from LISP? .. i tried by analogy (define (mus-alsa-access) "MMAP_NONINTERLEAVED") but it didn't change anything . ---- all the best, Ilya D. From edeleflie at gmail.com Thu Jul 5 07:06:31 2007 From: edeleflie at gmail.com (e deleflie) Date: Fri, 6 Jul 2007 00:06:31 +1000 Subject: [CM] normalize-sound is an unbound variable Message-ID: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> I'm trying to use normalize-sound as documented here: http://ccrma.stanford.edu/software/snd/snd/sndscm.html#normalizesound but I'm getting an "unbound variable" error. I've included a load directive at the start of my script (as advised at the top of the above page): (load "/home/stuff/snd-9/dsp.scm") ... which apparently finds the file... but still no normalize-sound. Actually I cant find any definition for "normalize-sound" inside dsp.scm ... have I got the wrong dsp.scm? ... or something else .... Etienne From k.s.matheussen at notam02.no Thu Jul 5 07:28:46 2007 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Thu, 5 Jul 2007 16:28:46 +0200 (CEST) Subject: [CM] normalize-sound is an unbound variable In-Reply-To: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> References: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> Message-ID: On Fri, 6 Jul 2007, e deleflie wrote: > I'm trying to use normalize-sound as documented here: > > http://ccrma.stanford.edu/software/snd/snd/sndscm.html#normalizesound > > but I'm getting an "unbound variable" error. I've included a load > directive at the start of my script (as advised at the top of the > above page): > > (load "/home/stuff/snd-9/dsp.scm") > > ... which apparently finds the file... but still no normalize-sound. > Actually I cant find any definition for "normalize-sound" inside > dsp.scm ... > Me neither. :-) > have I got the wrong dsp.scm? ... or something else .... > If you need to specify position and duration, you can use "scale-selection-to". If not, "scale-to" should be enough. From edeleflie at gmail.com Thu Jul 5 07:47:26 2007 From: edeleflie at gmail.com (e deleflie) Date: Fri, 6 Jul 2007 00:47:26 +1000 Subject: [CM] normalize-sound is an unbound variable In-Reply-To: References: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> Message-ID: <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> > If you need to specify position and duration, you can use > "scale-selection-to". If not, "scale-to" should be enough. there is a slight complication which is that I am dealing with a 4 channel file ... and all channels have to be scaled by an equal amount ... to use scale-to I would have to first establish which channel has the highest amplitude, then go through the other channels and find their highest amplitude... divide them to find out what the equivalent scale-to should be for each independent channel and then apply the scale-to 4 times with 4 different values each time. normalize-sound, which advertises that it functions on a sound rather than a channel, would do all that for me (theoretically)... ... unless there is a scale-to version that can work on a sound instead of a channel? Etienne From edeleflie at gmail.com Thu Jul 5 07:52:53 2007 From: edeleflie at gmail.com (e deleflie) Date: Fri, 6 Jul 2007 00:52:53 +1000 Subject: [CM] normalize-sound is an unbound variable In-Reply-To: <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> References: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> Message-ID: <9a471d320707050752vf0bc7c4red4c3ec63acaf456@mail.gmail.com> found it in extensions.scm so there's 2 (potential) documentation bugs (unless I am misreading)... the doc for normalize-sound says that the code is in frame.scm ... and the top of that same doco page says that all these functions are in dsp.scm Etienne On 7/6/07, e deleflie wrote: > > If you need to specify position and duration, you can use > > "scale-selection-to". If not, "scale-to" should be enough. > > there is a slight complication which is that I am dealing with a 4 > channel file ... and all channels have to be scaled by an equal amount > ... > > to use scale-to I would have to first establish which channel has the > highest amplitude, then go through the other channels and find their > highest amplitude... divide them to find out what the equivalent > scale-to should be for each independent channel and then apply the > scale-to 4 times with 4 different values each time. > > normalize-sound, which advertises that it functions on a sound rather > than a channel, would do all that for me (theoretically)... > > ... unless there is a scale-to version that can work on a sound > instead of a channel? > > Etienne > From k.s.matheussen at notam02.no Thu Jul 5 07:56:51 2007 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Thu, 5 Jul 2007 16:56:51 +0200 (CEST) Subject: [CM] normalize-sound is an unbound variable In-Reply-To: <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> References: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> Message-ID: On Fri, 6 Jul 2007, e deleflie wrote: >> If you need to specify position and duration, you can use >> "scale-selection-to". If not, "scale-to" should be enough. > > there is a slight complication which is that I am dealing with a 4 > channel file ... and all channels have to be scaled by an equal amount > ... > > to use scale-to I would have to first establish which channel has the > highest amplitude, then go through the other channels and find their > highest amplitude... divide them to find out what the equivalent > scale-to should be for each independent channel and then apply the > scale-to 4 times with 4 different values each time. > > normalize-sound, which advertises that it functions on a sound rather > than a channel, would do all that for me (theoretically)... > > ... unless there is a scale-to version that can work on a sound > instead of a channel? I'm pretty sure scale-to works the way you describe. Just don't specify channel, and it should scale the whole sound. From bil at ccrma.Stanford.EDU Thu Jul 5 10:24:17 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 5 Jul 2007 10:24:17 -0700 Subject: [CM] normalize-sound is an unbound variable In-Reply-To: <9a471d320707050752vf0bc7c4red4c3ec63acaf456@mail.gmail.com> References: <9a471d320707050706m2eb00158mf539267f46c39ba1@mail.gmail.com> <9a471d320707050747s4ecdc396k44e2fbd192b6d05e@mail.gmail.com> <9a471d320707050752vf0bc7c4red4c3ec63acaf456@mail.gmail.com> Message-ID: <20070705172251.M50945@ccrma.Stanford.EDU> > the doc for normalize-sound says that the code is in > frame.scm ... and the top of that same doco page says that all these > functions are in dsp.scm Thanks for the bug report -- I've fixed the reference to frame.scm. In my version of sndscm.html, normalize-sound is listed in the extensions.scm section. From bil at ccrma.Stanford.EDU Thu Jul 5 12:12:47 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 5 Jul 2007 12:12:47 -0700 Subject: [CM] alsa problem .. In-Reply-To: <1183623335.7571.20.camel@00110101> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> Message-ID: <20070705190844.M86097@ccrma.Stanford.EDU> I don't know anything about this, but I notice a variable snd_pcm_access_t alsa_interleave in audio.c line 3971 -- I wonder what would happen if you changed it to some other value? I can easily bring it out to the Snd extension language if it's useful. From nando at ccrma.Stanford.EDU Thu Jul 5 12:23:11 2007 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: Thu, 05 Jul 2007 12:23:11 -0700 Subject: [CM] alsa problem .. In-Reply-To: <20070705190844.M86097@ccrma.Stanford.EDU> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> <20070705190844.M86097@ccrma.Stanford.EDU> Message-ID: <1183663391.17234.16.camel@cmn3.stanford.edu> On Thu, 2007-07-05 at 12:12 -0700, Bill Schottstaedt wrote: > I don't know anything about this, but I notice a variable > > snd_pcm_access_t alsa_interleave > > in audio.c line 3971 -- I wonder what would happen if you > changed it to some other value? I can easily bring it out to > the Snd extension language if it's useful. Hmm, I fuzzily seem to remember I never implemented interleaved support in the alsa driver (only a few - two - cards used it at the time), if nobody fixed this in between...... -- Fernando From josepadovani at yahoo.com.br Thu Jul 5 17:40:33 2007 From: josepadovani at yahoo.com.br (padovani) Date: Thu, 05 Jul 2007 21:40:33 -0300 Subject: [CM] linux, midi default player... caution: newbie! :) Message-ID: <468D8F81.8010509@yahoo.com.br> Hi, so... I'm happy that I can listen to all iterations I'm studying on "Notes from the Metalevel"... I can hear the examples that begin with (events (process-name args) "midifile.mid")... Now, I'm trying to understand which player cm is using to play the midi-events... I can play midifiles only with (play "midifile.mid"), but lines like (oss-play-midi-file file); (set-midi-file-versions! true); (set-midi-output-hook! #'oss-play-midi-file) return errors that I just can't understand... I'm using Ubuntu Studio (Feisty, 7.04)... Thanks if anyone can explain better these things and how portmidi and midishare are called/used(?) by cm to play things... Tx, Jos? H. Padovani From error at balticom.lv Fri Jul 6 09:53:59 2007 From: error at balticom.lv (UnknownError) Date: Fri, 06 Jul 2007 16:53:59 +0000 Subject: [CM] alsa problem .. In-Reply-To: <20070705190844.M86097@ccrma.Stanford.EDU> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> <20070705190844.M86097@ccrma.Stanford.EDU> Message-ID: <1183740839.7571.25.camel@00110101> On Thu, 2007-07-05 at 12:12 -0700, Bill Schottstaedt wrote: > I don't know anything about this, but I notice a variable > > snd_pcm_access_t alsa_interleave > > in audio.c line 3971 -- I wonder what would happen if you > changed it to some other value? I can easily bring it out to > the Snd extension language if it's useful. right, i reckon it's a very good idea to get all such parametrs adjustable in the listener . i'll try changing the variable in audio.c .. will let you know > From juanig at ccrma.Stanford.EDU Fri Jul 6 09:58:02 2007 From: juanig at ccrma.Stanford.EDU (Juan I Reyes) Date: Fri, 06 Jul 2007 12:58:02 -0400 Subject: [CM] linux, midi default player... caution: newbie! :) In-Reply-To: <468D8F81.8010509@yahoo.com.br> References: <468D8F81.8010509@yahoo.com.br> Message-ID: <1183741082.14080.40.camel@strawberry> > Now, I'm trying to understand which player cm is using to play the > midi-events... I can play midifiles only with (play "midifile.mid"), but > lines like I think you have to see what is the value of the variable *midi-player* : * *midi-player* In Ubuntu as well as in Fedora you can use Timidity as a MIDI player. You can also use pmidi. In the case of Timidity you can set the value *midi-player* as, (setf *midi-player* "timidity -Os -quite=2") provided timidity is on your path. Timidity reminds me of the MusicKit days on Next hardware :-) --* Juan Reyes From edeleflie at gmail.com Sat Jul 7 03:29:56 2007 From: edeleflie at gmail.com (e deleflie) Date: Sat, 7 Jul 2007 20:29:56 +1000 Subject: [CM] Syntax error in fixed argument declaration Message-ID: <9a471d320707070329s20b8c4d9y7b1f265d0b453f5@mail.gmail.com> Hi All's fine on my development machine that has Guile 1.8, but on my deploy machine (Guile 1.6.7), when I run my script I get: "Syntax error in fixed argument declaration" So I dug around and found that I should load 'fix-optargs.scm' first like this: (load "/opt/snd-9/fix-optargs.scm") (load "/opt/snd-9/extensions.scm") ... but now I get a different error (which just seems to concern the first few lines of the script): unbound-variable: (#f Unbound variable: ~S (define*) #f) Unbound variable: define* Any ideas how I can get Guile 1.6.7 to behave itself ? Etienne From bil at ccrma.Stanford.EDU Sat Jul 7 04:30:41 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 7 Jul 2007 04:30:41 -0700 Subject: [CM] Syntax error in fixed argument declaration In-Reply-To: <9a471d320707070329s20b8c4d9y7b1f265d0b453f5@mail.gmail.com> References: <9a471d320707070329s20b8c4d9y7b1f265d0b453f5@mail.gmail.com> Message-ID: <20070707112729.M6648@ccrma.Stanford.EDU> That error should only occur in Guile 1.4 or earlier, and define* is defined in fix-optargs.scm, so I don't understand what is happening. The Help:About Snd menu item will tell you which Guile Snd thinks it is running, and at the end of the info is a list of the files from Guile that Snd has loaded. I'd start there -- perhaps you're actually getting a much older version than you think. From edeleflie at gmail.com Sat Jul 7 20:35:32 2007 From: edeleflie at gmail.com (e deleflie) Date: Sun, 8 Jul 2007 13:35:32 +1000 Subject: [CM] Syntax error in fixed argument declaration In-Reply-To: <20070707112729.M6648@ccrma.Stanford.EDU> References: <9a471d320707070329s20b8c4d9y7b1f265d0b453f5@mail.gmail.com> <20070707112729.M6648@ccrma.Stanford.EDU> Message-ID: <9a471d320707072035g2f464244x81c32a105590319f@mail.gmail.com> yes, I had 1.4 installed .... and I overode it with 1.6.7 .... but didn't think to recompile SND .... after a recompile all is well. Etienne On 7/7/07, Bill Schottstaedt wrote: > That error should only occur in Guile 1.4 or earlier, and define* is > defined in fix-optargs.scm, so I don't understand what is happening. > The Help:About Snd menu item will tell you which Guile Snd thinks > it is running, and at the end of the info is a list of the files from > Guile that Snd has loaded. I'd start there -- perhaps you're actually > getting a much older version than you think. > > > From error at balticom.lv Sun Jul 8 01:08:32 2007 From: error at balticom.lv (Unk_Err_) Date: Sun, 08 Jul 2007 08:08:32 +0000 Subject: [CM] multiuser audio editing environment Message-ID: <1183882112.7571.68.camel@00110101> good morning list. I have been considering how useful would be to have multiple users working on the same project, especialy such as an audio track editing or some graphical production. there are a few ways of implimenting it, depending where the users located and how they would like to organise their work.. well, probably for an audio project it could good just to start with a single machine which would have two monitors connected to it and two keyborad + pointer devices attached to each screen. that's because of the real-time issues and the simplicity for the first try. there are more complex way to be considered later , e.g. using different machines and a network filesystem, or using DMX (distributed multi-head X server) . first thing i'd like to ask, if it would be doable with Snd - to have multiple view of the same sound file, so we run it on one screen and then do something so there is a new window poping up and we just drag it to the second screen for the second user to work on .. regarding the audio output, let's presume that at firts users will use the same interface and agree when each of them would like to hear the output .. this is not very convinient, but just for the first experiment will do fine. then after jackd can be used , i gues it can be set that for each next portion of output a new jack channel can be used, and that may be routed to users headphone according to the local needs. as i have also thought, this may be implemeted in such a way, that Snd is first invoked as sort of console process, withou a gui and then from there the two (or more) windows with the same file are opend, or something similar.. it might even be start with -separate switch and then programmed in such a way that each new widget comes out as double or triple .. do you think this could be done and would be quite usefull ? also another way i meantioned above - using a network filesystem .. it would be very simple to edit two copies separatly and then compare and merge the two versions.. possibly implementing something as cool as snddiff ;) but the metter of this experiment is to get people working together at the same time on the same thing. in such a way we would be able to break the usual barier of the very common issue, that only an individual is working on a computer based creative project! so the question is, if that would be possible to have the same audiofile being syncroniously edited by snd runing separately on two machines and the flie located on a sort of network fileserver. we might consider some extra reimplementations of the network filesystem and/or some means of communication between instances of Snd, so the work could be syncronised in a way. ------ best regards, ilay d/ From error at balticom.lv Sun Jul 8 01:41:13 2007 From: error at balticom.lv (Unk_Err_) Date: Sun, 08 Jul 2007 08:41:13 +0000 Subject: [CM] alsa problem .. In-Reply-To: <1183663391.17234.16.camel@cmn3.stanford.edu> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> <20070705190844.M86097@ccrma.Stanford.EDU> <1183663391.17234.16.camel@cmn3.stanford.edu> Message-ID: <1183884073.7571.72.camel@00110101> On Thu, 2007-07-05 at 12:23 -0700, Fernando Lopez-Lezcano wrote: > On Thu, 2007-07-05 at 12:12 -0700, Bill Schottstaedt wrote: > > I don't know anything about this, but I notice a variable > > > > snd_pcm_access_t alsa_interleave > > > > in audio.c line 3971 -- I wonder what would happen if you > > changed it to some other value? I can easily bring it out to > > the Snd extension language if it's useful. > i have tried to change alsa_interleave to MMAP_NONINTERLEAVE and to RW_NONINTERLEAVE, it compile fine , of course, but when tried to play a file, said device is busy . it wasn't busy at all. > Hmm, I fuzzily seem to remember I never implemented interleaved support > in the alsa driver (only a few - two - cards used it at the time), if > nobody fixed this in between...... > > -- Fernando > > From bil at ccrma.Stanford.EDU Sun Jul 8 10:06:05 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sun, 8 Jul 2007 10:06:05 -0700 Subject: [CM] multiuser audio editing environment In-Reply-To: <1183882112.7571.68.camel@00110101> References: <1183882112.7571.68.camel@00110101> Message-ID: <20070708170425.M58453@ccrma.Stanford.EDU> You can have the same sound open multiple times either in one Snd, or multiple Snds; so if the net access was set up, you could have people on different machines working on the same sound. Their edit sequences won't collide until someone saves his edits, changing the underlying sound; then all the others get the exploding bomb icon -- each could have the update-hook set up to grab their current edits via edit-list->function, update to the new version of the sound, then reapply the saved edits, but whether that is a sensible thing to do depends on the edits. I was about to write such an update-hook function, but got interested in snddiff instead -- a first very simple-minded version: (define (cross-correlate snd0 chn0 snd1 chn1) (let* ((len0 (frames snd0 chn0)) (len1 (frames snd1 chn1)) (ilen (max len0 len1)) (pow2 (inexact->exact (ceiling (/ (log ilen) (log 2))))) (fftlen (inexact->exact (expt 2 pow2))) (fftlen2 (/ fftlen 2)) (fftscale (/ 1.0 fftlen)) (rl1 (channel->vct 0 fftlen snd1 chn1)) (rl2 (channel->vct 0 fftlen snd0 chn0)) (im1 (make-vct fftlen)) (im2 (make-vct fftlen))) (fft rl1 im1 1) (fft rl2 im2 1) (let* ((tmprl (vct-copy rl1)) (tmpim (vct-copy im1)) (data3 (make-vct fftlen))) (vct-multiply! tmprl rl2) ; (* tempr1 tempr2) (vct-multiply! tmpim im2) ; (* tempi1 tempi2) (vct-multiply! im2 rl1) ; (* tempr1 tempi2) (vct-multiply! rl2 im1) ; (* tempr2 tempi1) (vct-add! tmprl tmpim) ; add the first two (vct-subtract! im2 rl2) ; subtract the 4th from the 3rd (vct-scale! (fft tmprl im2 -1) fftscale)))) (define (lag? snd0 chn0 snd1 chn1) ;; returns the probable lagtime between the two sounds (negative time means second sound is delayed) (let* ((corr (cross-correlate snd0 chn0 snd1 chn1)) (len (vct-length corr)) (pk (- (vct-peak corr) .000001)) (pos -1) (lag (do ((i 0 (1+ i))) ((or (= i len) (>= pos 0)) pos) (if (>= (vct-ref corr i) pk) (set! pos i))))) (if (= lag -1) 0 (if (< lag (/ len 2)) lag (- (- len lag)))))) (define (snddiff snd0 chn0 snd1 chn1) ;; this can currently find initial delays, scaling differences, and scattered individual sample differences (let ((lag (lag? snd0 chn0 snd1 chn1))) (if (> lag 0) (pad-channel 0 lag snd1 chn1) (if (< lag 0) (pad-channel 0 (- lag) snd0 chn0))) (let ((s0 (channel->vct 0 #f snd0 chn0)) (s1 (channel->vct 0 #f snd1 chn1))) (if (= (vct-peak (vct-subtract! s0 s1)) 0.0) (if (= lag 0) ; trailing zeros? "no difference" (format #f "no difference except ~A is delayed ~A samples" (if (> lag 0) "first" "second") (abs lag))) (let* ((pos (maxamp-position snd0 chn0)) (mx0 (sample pos snd0 chn0)) (mx1 (sample pos snd1 chn1)) ; use actual values to keep possible sign difference (scl (/ mx1 mx0))) (scale-channel scl 0 #f snd0 chn0) (set! s0 (channel->vct 0 #f snd0 chn0)) (if (< (vct-peak (vct-subtract! s0 s1)) 0.0001) (if (= lag 0) (format #f "second is ~A * first" scl) (format #f "~A is delayed ~A samples and multiplied by ~A" (if (> lag 0) "first" "second") (abs lag) (if (> lag 0) (/ 1.0 scl) scl))) (begin (if (not (= scl 1.0)) (undo 1 snd0 chn0)) (set! s0 (channel->vct 0 #f snd0 chn0)) (let ((diffs 0) (diff-data '()) (len (min (vct-length s0) (vct-length s1)))) (do ((i 0 (1+ i))) ((or (> diffs 10) (= i len))) (if (> (abs (- (vct-ref s0 i) (vct-ref s1 i))) .00001) (begin (set! diffs (1+ diffs)) (set! diff-data (cons (list i (vct-ref s0 i) (vct-ref s1 i)) diff-data))))) (if (< diffs 10) (format #f "different sample~A: ~A" (if (= diffs 1) "" "s") (reverse diff-data)) (begin "giving up")))))))))) From mperti at ciudad.com.ar Sun Jul 8 12:08:34 2007 From: mperti at ciudad.com.ar (M. Perticone) Date: Sun, 8 Jul 2007 16:08:34 -0300 Subject: [CM] pcsets ready-made code References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> <20070705190844.M86097@ccrma.Stanford.EDU> <1183663391.17234.16.camel@cmn3.stanford.edu> Message-ID: <000701c7c193$629e18f0$fa9fd4c9@piii> hello cm listers, has anyone already worked some code for pitch class sets? something involving morris' and lewin's concepts? i just want to avoid starting from scratch. any pointer greatly appreciated, thx, marcelo From taube at uiuc.edu Sun Jul 8 12:54:13 2007 From: taube at uiuc.edu (Rick Taube) Date: Sun, 8 Jul 2007 14:54:13 -0500 Subject: [CM] pcsets ready-made code In-Reply-To: <000701c7c193$629e18f0$fa9fd4c9@piii> References: <20061112200003.2015.13246.Mailman@cm-mail.stanford.edu> <1183623335.7571.20.camel@00110101> <20070705190844.M86097@ccrma.Stanford.EDU> <1183663391.17234.16.camel@cmn3.stanford.edu> <000701c7c193$629e18f0$fa9fd4c9@piii> Message-ID: <9E63DD9C-DDAA-4867-9CE7-8E7395A0B109@uiuc.edu> cm basically defines just prime-form: http://commonmusic.sourceforge.net/doc/dict/prime-form-fn.html Ill bet that Drew Krause's toolbox has set functions defined in it, there is a link to it on the cm homepage. otherwise, John Amuedo had a Lisp system called Set Slave a few years ago, you might find it somewhere on the web... On Jul 8, 2007, at 2:08 PM, M. Perticone wrote: > hello cm listers, > > has anyone already worked some code for pitch class sets? something > involving morris' and lewin's concepts? > i just want to avoid starting from scratch. > any pointer greatly appreciated, > thx, > marcelo > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From error at balticom.lv Mon Jul 9 07:35:57 2007 From: error at balticom.lv (Unk_Err_) Date: Mon, 09 Jul 2007 14:35:57 +0000 Subject: [CM] multiuser audio editing environment In-Reply-To: <20070708170425.M58453@ccrma.Stanford.EDU> References: <1183882112.7571.68.camel@00110101> <20070708170425.M58453@ccrma.Stanford.EDU> Message-ID: <1183991757.2111.16.camel@00110101> On Sun, 2007-07-08 at 10:06 -0700, Bill Schottstaedt wrote: > You can have the same sound open multiple times either in one Snd, or > multiple Snds; so if the net access was set up, you could have people > on different machines working on the same sound. Their edit sequences > won't collide until someone saves his edits, changing the underlying > sound; then all the others get the exploding bomb icon -- each could > have the update-hook set up to grab their current edits via edit-list->function, > update to the new version of the sound, then reapply the saved edits, but > whether that is a sensible thing to do depends on the edits. > hello. i have have looked at the snddiff code above, i'm going to try it propertly later . so right, i'll think of what update-hook could be .. the only thing is that so far none of my friends ever used Snd .. i probably need to introduce them to snd-ls ;) will see what it gets to in practice . people won't be able to work on the same region of a soundfile exactly at the same time anyway, but the can actualy devide between them - so each user works on his/her region about .. i first thought of a collective digital painting process, when people paint , say in Gimp, and each can easely add a stroke, especialy could be great for children! i'm gonna try it out using Xdmx (http://dmx.sf.net) so with a wavefroms it's different, but such things as controls could be adjusted simultanously in way that one user sets something roughly and the other fine-tunes that and so on .. > I was about to write such an update-hook function, but got interested > in snddiff instead -- a first very simple-minded version: > > > (define (cross-correlate snd0 chn0 snd1 chn1) > (let* ((len0 (frames snd0 chn0)) > (len1 (frames snd1 chn1)) > (ilen (max len0 len1)) > (pow2 (inexact->exact (ceiling (/ (log ilen) (log 2))))) > (fftlen (inexact->exact (expt 2 pow2))) > (fftlen2 (/ fftlen 2)) > (fftscale (/ 1.0 fftlen)) > (rl1 (channel->vct 0 fftlen snd1 chn1)) > (rl2 (channel->vct 0 fftlen snd0 chn0)) > (im1 (make-vct fftlen)) > (im2 (make-vct fftlen))) > (fft rl1 im1 1) > (fft rl2 im2 1) > (let* ((tmprl (vct-copy rl1)) > (tmpim (vct-copy im1)) > (data3 (make-vct fftlen))) > (vct-multiply! tmprl rl2) ; (* tempr1 tempr2) > (vct-multiply! tmpim im2) ; (* tempi1 tempi2) > (vct-multiply! im2 rl1) ; (* tempr1 tempi2) > (vct-multiply! rl2 im1) ; (* tempr2 tempi1) > (vct-add! tmprl tmpim) ; add the first two > (vct-subtract! im2 rl2) ; subtract the 4th from the 3rd > (vct-scale! (fft tmprl im2 -1) fftscale)))) > > (define (lag? snd0 chn0 snd1 chn1) > ;; returns the probable lagtime between the two sounds (negative time means second sound is > delayed) > (let* ((corr (cross-correlate snd0 chn0 snd1 chn1)) > (len (vct-length corr)) > (pk (- (vct-peak corr) .000001)) > (pos -1) > (lag (do ((i 0 (1+ i))) > ((or (= i len) > (>= pos 0)) > pos) > (if (>= (vct-ref corr i) pk) > (set! pos i))))) > (if (= lag -1) > 0 > (if (< lag (/ len 2)) > lag > (- (- len lag)))))) > > (define (snddiff snd0 chn0 snd1 chn1) > ;; this can currently find initial delays, scaling differences, and scattered individual sample > differences > (let ((lag (lag? snd0 chn0 snd1 chn1))) > (if (> lag 0) > (pad-channel 0 lag snd1 chn1) > (if (< lag 0) > (pad-channel 0 (- lag) snd0 chn0))) > > (let ((s0 (channel->vct 0 #f snd0 chn0)) > (s1 (channel->vct 0 #f snd1 chn1))) > (if (= (vct-peak (vct-subtract! s0 s1)) 0.0) > (if (= lag 0) ; trailing zeros? > "no difference" > (format #f "no difference except ~A is delayed ~A samples" (if (> lag 0) "first" "second") (abs > lag))) > > (let* ((pos (maxamp-position snd0 chn0)) > (mx0 (sample pos snd0 chn0)) > (mx1 (sample pos snd1 chn1)) ; use actual values to keep possible sign difference > (scl (/ mx1 mx0))) > (scale-channel scl 0 #f snd0 chn0) > (set! s0 (channel->vct 0 #f snd0 chn0)) > (if (< (vct-peak (vct-subtract! s0 s1)) 0.0001) > (if (= lag 0) > (format #f "second is ~A * first" scl) > (format #f "~A is delayed ~A samples and multiplied by ~A" > (if (> lag 0) "first" "second") > (abs lag) > (if (> lag 0) (/ 1.0 scl) scl))) > (begin > (if (not (= scl 1.0)) (undo 1 snd0 chn0)) > (set! s0 (channel->vct 0 #f snd0 chn0)) > (let ((diffs 0) > (diff-data '()) > (len (min (vct-length s0) (vct-length s1)))) > (do ((i 0 (1+ i))) > ((or (> diffs 10) > (= i len))) > (if (> (abs (- (vct-ref s0 i) (vct-ref s1 i))) .00001) > (begin > (set! diffs (1+ diffs)) > (set! diff-data (cons (list i (vct-ref s0 i) (vct-ref s1 i)) diff-data))))) > (if (< diffs 10) > (format #f "different sample~A: ~A" (if (= diffs 1) "" "s") (reverse diff-data)) > (begin > "giving up")))))))))) > > From johnhenrydale at gmail.com Mon Jul 9 10:21:21 2007 From: johnhenrydale at gmail.com (john henry dale) Date: Mon, 9 Jul 2007 13:21:21 -0400 Subject: [CM] just intonation Message-ID: Hello CLM list people, can anyone tell me how I would create an arpeggio that goes through a just intonation scale ? I have all the frequency values in a reference boook, but I am unsure of how to write the code for this is CLM ? Thanks, JHD -------------- next part -------------- An HTML attachment was scrubbed... URL: From bil at ccrma.Stanford.EDU Mon Jul 9 12:28:39 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 9 Jul 2007 12:28:39 -0700 Subject: [CM] just intonation In-Reply-To: References: Message-ID: <20070709192335.M10815@ccrma.Stanford.EDU> > can anyone tell me how I would create an arpeggio that goes through a just intonation scale ? First you need an instrument -- the fm-violin is a good choice. So, compile v.ins and load v. Next put your frequencies in an array, then (with-sound () (loop for i from 0 to 7 do (fm-violin i 1 (aref pitches i) .1))) scales.cl might be of interest. From taube at uiuc.edu Mon Jul 9 13:14:27 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 9 Jul 2007 15:14:27 -0500 Subject: [CM] just intonation In-Reply-To: <20070709192335.M10815@ccrma.Stanford.EDU> References: <20070709192335.M10815@ccrma.Stanford.EDU> Message-ID: <47351DD9-CE60-4BD3-BBEA-C39C8B6D0024@uiuc.edu> a "just" scale is simply a scale built from ratios in the harmonic series. you could create these scales using function that takes a fundamental and as many ratios as you like: (defun jscale (fund &rest ratios) (loop for r in ratios collect (* fund r))) (define (jscale fund . ratios) (if (null? ratios) '() (cons (* fund (car ratios)) (apply jscale fund (cdr ratios))))) ; just major scale (jscale 220.0 1 9/8 5/4 4/3 3/2 5/3 15/8 2) => (220.0 247.5 275.0 293.33334 330.0 366.66666 412.5 440.0) ; harry partch 43-tone just scale (jscale 220.0 1/1 81/80 33/32 21/20 16/15 12/11 11/10 10/9 9/8 8/7 7/6 32/27 6/5 11/9 5/4 14/11 9/7 21/16 4/3 27/20 11/8 7/5 10/7 16/11 40/27 3/2 32/21 14/9 11/7 8/5 18/11 5/3 27/16 12/7 7/4 16/9 9/5 20/11 11/6 15/8 40/21 64/33 160/81 2/1) => (220.0 222.75002 226.875 230.99998 234.66667 240.0 242.0 244.44446 247.5 251.42859 256.66666 260.74075 264.0 268.8889 275.0 280.0 282.85715 288.75 293.33334 297.0 302.5 308.0 314.28574 320.0 325.9259 330.0 335.2381 342.22223 345.71426 352.0 360.0 366.66666 371.25 377.14285 385.0 391.1111 396.0 400.0 403.33334 412.5 419.0476 426.6667 434.5679 440.0) From taube at uiuc.edu Mon Jul 9 13:18:34 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 9 Jul 2007 15:18:34 -0500 Subject: [CM] just intonation In-Reply-To: <47351DD9-CE60-4BD3-BBEA-C39C8B6D0024@uiuc.edu> References: <20070709192335.M10815@ccrma.Stanford.EDU> <47351DD9-CE60-4BD3-BBEA-C39C8B6D0024@uiuc.edu> Message-ID: sorry forgot the sound! (with-sound () (loop for p in (jscale 220.0 1 9/8 5/4 4/3 3/2 5/3 15/8 2) for start = 0 then (+ start .25) do (fm-violin start 1 (elt pitches i) .1))) From znmeb at cesmail.net Mon Jul 9 21:20:36 2007 From: znmeb at cesmail.net (M. Edward (Ed) Borasky) Date: Mon, 09 Jul 2007 21:20:36 -0700 Subject: [CM] just intonation In-Reply-To: <20070709192335.M10815@ccrma.Stanford.EDU> References: <20070709192335.M10815@ccrma.Stanford.EDU> Message-ID: <46930914.6040400@cesmail.net> Bill Schottstaedt wrote: >> can anyone tell me how I would create an arpeggio that goes through a just intonation scale ? > > First you need an instrument -- the fm-violin is a good choice. So, > compile v.ins and load v. Next put your frequencies in an array, > then > > (with-sound () (loop for i from 0 to 7 do (fm-violin i 1 (aref pitches i) .1))) > > scales.cl might be of interest. > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > If you're seriously interested in just intonation and microtonality (and fm-violins) I'd recommend getting a copy of William Sethares' "Tuning, Timbre, Spectrum, Scale", 2nd edition. He's got a whole chapter on scales that sound good with FM instruments, and I don't recall just intonation being one of them. From plewto at gmail.com Tue Jul 10 01:25:21 2007 From: plewto at gmail.com (Steven Jones) Date: Tue, 10 Jul 2007 03:25:21 -0500 Subject: [CM] problems with libclm.so while installing clm Message-ID: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> I'm trying to install clm-3 on FC5 system under SBCL using alsa. Loading all.lisp initially works but bombs out with the following message: Error opening shared object "/home/sj/documents/src/lisp/clm-3/libclm.so": cannot restore segment prot after reloc: Permission denied. For the record I have tried using the oss default instead of alsa and also GCL with similar results. Any help greatly appreciated. Thanks Steven Jones From bil at ccrma.Stanford.EDU Tue Jul 10 04:20:47 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 10 Jul 2007 04:20:47 -0700 Subject: [CM] problems with libclm.so while installing clm In-Reply-To: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> References: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> Message-ID: <20070710111823.M90688@ccrma.Stanford.EDU> > cannot restore segment prot after reloc: Permission denied. I haven't seen this error before, but Googling it leads to me to think you might have a problem with some security setting? From nando at ccrma.Stanford.EDU Tue Jul 10 09:38:43 2007 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: Tue, 10 Jul 2007 09:38:43 -0700 Subject: [CM] problems with libclm.so while installing clm In-Reply-To: <20070710111823.M90688@ccrma.Stanford.EDU> References: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> <20070710111823.M90688@ccrma.Stanford.EDU> Message-ID: <1184085523.2399.0.camel@cmn3.stanford.edu> On Tue, 2007-07-10 at 04:20 -0700, Bill Schottstaedt wrote: > > cannot restore segment prot after reloc: Permission denied. > > I haven't seen this error before, but Googling it leads to me to > think you might have a problem with some security setting? That's selinux, you can either disable it or change one of the fine tuning settings, I don't remember the exact name, something like 'permit relocation of text segments' or words to that effect. -- Fernando From plewto at gmail.com Tue Jul 10 12:30:27 2007 From: plewto at gmail.com (Steven Jones) Date: Tue, 10 Jul 2007 14:30:27 -0500 Subject: [CM] Re: problems with libclm.so while installing clm In-Reply-To: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> References: <58a9955a0707100125w17e0a1a0iaa808f676ffedf5d@mail.gmail.com> Message-ID: <58a9955a0707101230r10a8873cx2b5c2fed3d7660e@mail.gmail.com> Thanks for the responses. I changed /etc/sysconfig/selinux to contain the line SELINUX=permissive instead of "enforcing". That appears to have worked. From ruido_electronico at yahoo.es Tue Jul 10 13:24:08 2007 From: ruido_electronico at yahoo.es (ruben gomez) Date: Tue, 10 Jul 2007 22:24:08 +0200 (CEST) Subject: [CM] realtime application Message-ID: <609923.31504.qm@web27203.mail.ukl.yahoo.com> Hello: I want make one realtime application with a custom interface and one algorithmic sequencer plus synthesis and realtime MIDI out. I try Csound and now with python and puredata communicating via OSC but the system is not clean and fast because are two differents languages. I can program this application in CM totally in Clisp or Scheme or another implementation of Lisp? CM i'ts portable to Linux,Mac and Windows? By where I can begin? Thanks. (it excuse my english, i am spanish) --------------------------------- S? un Mejor Amante del Cine ?Quieres saber c?mo? ?Deja que otras personas te ayuden!. -------------- next part -------------- An HTML attachment was scrubbed... URL: From johnhenrydale at gmail.com Wed Jul 11 08:17:00 2007 From: johnhenrydale at gmail.com (john henry dale) Date: Wed, 11 Jul 2007 11:17:00 -0400 Subject: [CM] translating spirals into music CLM Message-ID: Hi, This is a rather conceptual question, but I am interested in trying to somehow represent the concept of a spiral in CLM. I figure this would probably involve the golden ratio and fibonacci series number, but I am stuck as to how to begin and how this "sound spiral" would actually sound. This corresponds to my earlier question about just intonation, because according to Alain Danielou ( in his book Music and the Power of Sound), the circle of fifths originated from a "spiral of fifths", which was modified by either slightly sharping or flatting certain notes to get equal temperament for the pianoforte; just intonation, according to him, was a truer representation of the "pure" musical intervals that occurred in nature, so to speak. I guess my goal here is try to somehow represent the concept of the spiral both melodically and rhythmically in CLM since this pattern seems to inform so much of how of and why things evolve the way they do. This could obviously get very complicated and beyond the timeframe I have to complete my project, so I'm hoping for short but potent bits of information here. Sorry to lay such a huge question on what is generally a fairly practical discussion list, but I need to find ways of narrowing the focus of this idea soon. Thanks much, JHD -------------- next part -------------- An HTML attachment was scrubbed... URL: From thomaslambert at altern.org Wed Jul 11 15:15:27 2007 From: thomaslambert at altern.org (Thomas Lambert) Date: Thu, 12 Jul 2007 00:15:27 +0200 (CEST) Subject: [CM] translating spirals into music CLM In-Reply-To: References: Message-ID: <4412.195.132.186.61.1184192127.squirrel@www.altern.org> From bil at ccrma.Stanford.EDU Wed Jul 11 15:56:03 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 11 Jul 2007 15:56:03 -0700 Subject: [CM] translating spirals into music CLM In-Reply-To: <4412.195.132.186.61.1184192127.squirrel@www.altern.org> References: <4412.195.132.186.61.1184192127.squirrel@www.altern.org> Message-ID: <20070711225018.M49017@ccrma.Stanford.EDU> Being of unsound mind (but the rest of me is ok), I once wrote a program to generate ideal comma ratios out to 2000 digits. Whenever I need a laugh, I run this program -- sort of like a box of chocolates (you need gmp): #include #include int main (int argc, char **argv) { MP_INT tt, a1, b1, a2, b2, a, b, tmp1, tmp2, tmp3, num, den, lscl, res; MP_RAT x, xtmp, xxtmp; int happy; FILE *fp; fp = fopen("data1", "w"); mpz_init_set_str(&tt, "1", 10); mpz_init_set_str(&a1, "0", 10); mpz_init_set_str(&res, "0", 10); mpz_init_set_str(&b2, "0", 10); mpz_init_set_str(&a2, "1", 10); mpz_init_set_str(&b1, "1", 10); mpz_init_set_str(&a, "0", 10); mpz_init_set_str(&b, "0", 10); mpz_init_set_str(&tmp1, "0", 10); mpz_init_set_str(&tmp2, "0", 10); mpz_init_set_str(&tmp3, "0", 10); #if 0 mpz_init_set_str(&num, "6931471805599453094172321", 10); mpz_init_set_str(&den, "4054651081081643819780131", 10); #endif mpz_init_set_str(&num, "69314718055994530941723212145817656807550013436025525412068000949339362196969 471560586332699641868754200148102057068573368552023575813055703267075163507596 193072757082837143519030703862389167347112335011536449795523912047517268157493 206515552473413952588295045300709532636664265410423915781495204374043038550080 194417064167151864471283996817178454695702627163106454615025720740248163777338 963855069526066834113727387372292895649354702576265209885969320196505855476470 330679365443254763274495125040606943814710468994650622016772042452452961268794 654619316517468139267250410380254625965686914419287160829380317271436778265487 756648508567407764845146443994046142260319309673540257444607030809608504748663 852313818167675143866747664789088143714198549423151997354880375165861275352916 610007105355824987941472950929311389715599820565439287170007218085761025236889 213244971389320378439353088774825970171559107088236836275898425891853530243634 214367061189236789192372314672321720534016492568727477823445353476481149418642 386776774406069562657379600867076257199184734022651462837904883062033061144630 073719489002743643965002580936519443041191150608094879306786515887090060520346 842973619384128965255653968602219412292420757432175748909770675268711581705113 700915894266547859596489065305846025866838294002283300538207400567705304678700 184162404418833232798386349001563121889560650553151272199398332030751408426091 479001265168243443893572472788205486271552741877243002489794540196187233980860 831664811490930667519339312890431641370681397776498176974868903887789991296503 619270710889264105230924783917373501229842420499568935992206602204654941510613 918788574424557751020683703086661948089641218680779020818158858000168811597305 618667619918739520076671921459223672060253959543654165531129517598994005600036 651356756905124592682574394648316833262490180382424082423145230614096380570070 255138770268178516306902551370323405380214501901537402950994226299577964742713 815736380172987394070424217997226696297993931270694", 10); mpz_init_set_str(&den, "10986122886681096913952452369225257046474905578227494517346943336374942932186 089668736157548137320887879700290659578657423680042259305198210528018707672774 106031627691833813671793736988443609599037425703167959115211455919177506713470 549401667755802222031702529468975606901065215056428681380363173732985777823669 916547921318181490200301038236301222486527481982259910974524908964580534670088 459650857484441190188570876474948670796130858294116021661211840014098255143919 487688936798494302255731535329685345295251459213876494685932562794416556941578 272310355168866102118469890439943063138255285736466882824988136822800634143910 786893251456437510204451627561934973982116941585740535361758900975122233797736 969687754354795135712982177017581242122351405810163272465588937249564919185242 960796684234647069377237252655082032078333928055892853146873095132606458309184 397496822230325765467533311823019649275257599132217851353390237482964339502546 074245824934666866121881436526565429542767610505477795422933973323401173743193 974579847018559548494059478353943841010602930762292228131207489306344534025277 732685627148001681871547243978207187803444678021617815841904282007672124325573 801436417887682616104101681872424068790890992987420815218323752894275273253407 100283575069506240396546275224430846258845085978625308322477453888506800348832 434049008399005808094356528212237038870203680454860077621424408869725941358436 599922621173967080495095279271436315464044462308915818536711960837030485352090 967262958241504035599512135545033224174847410033198148783245256933470494993730 165633666099190395712282284488167431215062856999387403881901274483956479103477 288597211985064942279698579166995641855126504150219155471966585692972660652357 329373683002783092177660538703046200766158494670022601175679751800393479176327 784493514263496836003755785716070049818151918437343829093474666045775065927367 012111537058249647984793040420582396475385785096062609338991470612013024310826 0518262958640076003059494321166880446106134684533980", 10); mpz_sub(&den, &den, &num); mpz_init_set_str(&lscl, "17312340490667563", 10); /* (/ 1.0 (log (expt 2.0 (/ 1.0 1200)))) */ mpq_init(&x); mpq_init(&xtmp); mpq_init(&xxtmp); mpq_set_num(&x, &num); mpq_set_den(&x, &den); happy = 1; while (happy) { mpz_mul(&tmp1, &a1, &tt); mpz_add(&a, &tmp1, &a2); mpz_mul(&tmp1, &b1, &tt); mpz_add(&b, &tmp1, &b2); mpz_mul(&tmp2, &den, &b); mpz_mul(&tmp1, &num, &a); mpz_sub(&tmp3, &tmp1, &tmp2); mpz_abs(&tmp1, &tmp3); #if 0 mpz_mul(&res, &tmp1, &lscl); /* printf("%s %s %s\n", mpz_get_str(NULL, 10, &res), mpz_get_str(NULL, 10, &b), mpz_get_str (NULL, 10, &a)); */ fprintf(fp, "%s %s %s\n", mpz_get_str(NULL, 10, &res), mpz_get_str(NULL, 10, &b), mpz_get_str (NULL, 10, &a)); #endif fprintf(fp, "%s %s\n", mpz_get_str(NULL, 10, &b), mpz_get_str(NULL, 10, &a)); fflush(fp); mpq_set_num(&xtmp, &tt); mpq_sub(&xxtmp, &x, &xtmp); mpq_inv(&x, &xxtmp); mpq_get_num(&tmp1, &x); mpq_get_den(&tmp2, &x); mpz_div(&tt, &tmp1, &tmp2); mpz_set(&a2, &a1); mpz_set(&b2, &b1); mpz_set(&a1, &a); mpz_set(&b1, &b); happy = mpz_cmp(&num, &b); } fclose(fp); } the printout starts: 1 1 2 1 5 3 12 7 41 24 53 31 306 179 665 389 15601 9126 31867 18641 79335 46408 111202 65049 190537 111457 10590737 6195184 10781274 6306641 53715833 31421748 171928773 100571885 225644606 131993633 397573379 232565518 6189245291 3620476403 6586818670 3853041921 65470613321 38297853692 137528045312 80448749305 753110839881 440541600217 which should be obvious to a microtuning devotee. From taube at uiuc.edu Wed Jul 11 16:01:17 2007 From: taube at uiuc.edu (Rick Taube) Date: Wed, 11 Jul 2007 18:01:17 -0500 Subject: [CM] translating spirals into music CLM In-Reply-To: References: Message-ID: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> johannes quint has an example of shepard tones in his course materials. the s.t. are the most obvious spiral sound i can think of off hand. it would be trivial to rewrite the code for fm-violin http://www.johannes-quint.de/cm/html/shepard.html On Jul 11, 2007, at 10:17 AM, john henry dale wrote: > Hi, > This is a rather conceptual question, but I am interested in trying > to somehow represent the concept of a spiral in CLM. I figure this > would probably involve the golden ratio and fibonacci series > number, but I am stuck as to how to begin and how this "sound > spiral" would actually sound. This corresponds to my earlier > question about just intonation, because according to Alain Danielou > ( in his book Music and the Power of Sound), the circle of fifths > originated from a "spiral of fifths", which was modified by either > slightly sharping or flatting certain notes to get equal > temperament for the pianoforte; just intonation, according to him, > was a truer representation of the "pure" musical intervals that > occurred in nature, so to speak. I guess my goal here is try to > somehow represent the concept of the spiral both melodically and > rhythmically in CLM since this pattern seems to inform so much of > how of and why things evolve the way they do. This could obviously > get very complicated and beyond the timeframe I have to complete my > project, so I'm hoping for short but potent bits of information > here. Sorry to lay such a huge question on what is generally a > fairly practical discussion list, but I need to find ways of > narrowing the focus of this idea soon. > Thanks much, > JHD > From bil at ccrma.Stanford.EDU Wed Jul 11 16:20:48 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 11 Jul 2007 16:20:48 -0700 Subject: [CM] translating spirals into music CLM In-Reply-To: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> Message-ID: <20070711231905.M46801@ccrma.Stanford.EDU> There's also an example of shepard tones in the ws.scm section of sndscm.html, complete with a purty picture that took me about 10 times longer to make than the sound. From bil at ccrma.Stanford.EDU Wed Jul 11 16:28:13 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 11 Jul 2007 16:28:13 -0700 Subject: [CM] Snd 9.2 Message-ID: <20070711232646.M11544@ccrma.Stanford.EDU> Snd 9.2 many Forth/Ruby improvements thanks to Mike. RIFF bugfix and minGW configure info thanks to Steve Beet. snddiff.scm virtual-filter-channel (examp.scm) max-virtual-ptrees (greatly expanded virtual editor) added optional snd chn args to apply-ladspa and with-tag to mix-channel power spectral density as transform (green.scm) clm: mus-increment on gens with a notion of frequency (phase increment); to make room for this, asymmetric-fm ratio and sine-summation b moved to mus-offset. exp envs optimized Thanks!: Mike Scholz, Plutek, Etienne Deleflie, Steve Beet, Kjetil Matheussen, unknown_error Checked: gtk 2.11.2|3|4|5, cairo 1.4.8|10, Fedora 7, sbcl 1.0.6|7 From plewto at gmail.com Thu Jul 12 00:43:55 2007 From: plewto at gmail.com (Steven Jones) Date: Thu, 12 Jul 2007 02:43:55 -0500 Subject: [CM] CLM Neophyte with play question Message-ID: <58a9955a0707120043l7b99b51y7ed86954aacb5636@mail.gmail.com> Just started playing with CLM and have a few questions. Using the fm-violin examples I have been able to generate sound files and play them using an external play utility. So far however I have been unable to play sounds from within CLM. Both with-sound and play simply return the sound file name without producing any audio output. On a possibly related note, searching the docs I tried (setf (voulme) 1.0) If I then enter (volume) the result is NIL. I'm using a SBCL on FC5 system with alsa. sndplay is a symbolic link to play. Thanks for any help Steven Jones From shevegen at linuxmail.org Thu Jul 12 04:19:12 2007 From: shevegen at linuxmail.org (Roebe XXX) Date: Thu, 12 Jul 2007 12:19:12 +0100 Subject: [CM] Dir naming scheme for snd package Message-ID: <20070712111912.5B4B42478C@ws5-3.us4.outblaze.com> Hi guys, sorry to bother you, but I wanted to have at least tried to :-) Right now the dir name of snd is a little bit bothersome (to me, of course), could I suggest that if i.e. snd version would be: snd-9.6 both the tarball becoming: snd-9.6.tar.bz2 and the directory name of snd to: snd-9.6 as well? Right now, if you extract i.e. snd-9.2 it becomes snd-9 but i believe the name should be the same. Yes it is no problem for me to repackage the dir at all to be 9.2 too, but I also wanted to try to point this out, maybe it could be changed. Thanks for reading anyway. :-) = -- Powered by Outblaze From bil at ccrma.Stanford.EDU Thu Jul 12 06:37:01 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 12 Jul 2007 06:37:01 -0700 Subject: [CM] CLM Neophyte with play question In-Reply-To: <58a9955a0707120043l7b99b51y7ed86954aacb5636@mail.gmail.com> References: <58a9955a0707120043l7b99b51y7ed86954aacb5636@mail.gmail.com> Message-ID: <20070712133352.M91394@ccrma.Stanford.EDU> Snd and CLM use the same audio output code (audio.c), so if one works, the other ought to. The first thing I'd check is that you're actually using the same audio system. I think OSS is the default still in both, so you would need to use the --with-alsa switch in Snd, and push :alsa on *features* in CLM. You can check the two mus-config.h files to see what you got (the macro HAVE_ALSA). If those are the same, the next thing I'd do is look at the output of mus-audio-describe. From bil at ccrma.Stanford.EDU Thu Jul 12 07:04:19 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 12 Jul 2007 07:04:19 -0700 Subject: [CM] Dir naming scheme for snd package In-Reply-To: <20070712111912.5B4B42478C@ws5-3.us4.outblaze.com> References: <20070712111912.5B4B42478C@ws5-3.us4.outblaze.com> Message-ID: <20070712135551.M50651@ccrma.Stanford.EDU> > Right now, if you extract i.e. snd-9.2 it becomes > snd-9 but i believe the name should be the same. I'm tempted to remove the ".2" or even the whole number -- I dislike this convention. But, I suppose if I go along with the naming scheme part way, I should go whole hog. From juanig at ccrma.Stanford.EDU Thu Jul 12 15:30:42 2007 From: juanig at ccrma.Stanford.EDU (Juan I Reyes) Date: Thu, 12 Jul 2007 18:30:42 -0400 Subject: [CM] translating spirals into music CLM In-Reply-To: <20070711231905.M46801@ccrma.Stanford.EDU> References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> Message-ID: <1184279442.3713.40.camel@strawberry> > There's also an example of shepard tones in the ws.scm section > of sndscm.html, complete with a purty picture that took me about 10 > times longer to make than the sound. Here is my CLM rendition of Shepard Tones based on Bill's ws.scm and from Charles Dodge's Computer Music text. (definstrument shepard (beg dur amp &key (incr .000001) ) (let* ((start (floor (* beg *srate*))) (end (+ start (floor (* dur *srate*)))) (x 0.0) (arr (make-array 12))) ;; we'll create a 12 oscil bank ;; (do ((i 0 (1+ i))) ;; initialize oscils ((= i 12)) (setf (aref arr i) (make-oscil :frequency 0.0)) ) ;; (run (loop for i from start to end do (let ((y 0.00) (oscbank 0.0)) ;; sum signal of twelve oscillators (do ((i 0 (1+ i))) ((= i (length arr))) ;; get amplitude and oscillator phase offset (let ((phoffset (+ x (/ i 12)))) (if (> phoffset 1) (setf phoffset (- phoffset 1))) (setf y (- 4.0 (* 8.0 phoffset))) (incf oscbank (* (exp (* -0.5 y y)) (oscil (aref arr i) (hz->radians (expt 2.0 (* (- 1 phoffset) 12.0)))))) ;; to go up ;;;(* phoffset 12.0) )) (incf x incr) (outa i (* amp oscbank)) ))) )) ;;;(with-sound () (shepard 0 10 0.1)) From taube at uiuc.edu Mon Jul 16 09:08:08 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 11:08:08 -0500 Subject: [CM] future of cm (long) Message-ID: <6DF14DA3-0B40-4241-9D7A-159EA1369CD3@uiuc.edu> Those using CM might be interested in learning that over the past few months Todd Ingalls and i have been (quietly) working away on a new C+ +/JUCE & Lisp environment for algorithmic composition called Grace (for Graphical Realtime Algorithmic Composition Environment). Although the app is still rough it is now a point where it can function as an (unstable) replacement for the current CM in Emacs +Slime. I am "announcing" Grace now because -- in the best case scenerio -- we will use it for the ICMC workshop in August and would therefore welcome anyone that can build it from CVS sources to try it out a bit to give us feedback: bug reports, bug fixes, ideas for better functionality etc. Needless to say, in the not-too-distant future the goal is to deliver Grace as a binary drag'n drop app that runs exactly the same on Windows, Linux and Macos with basically no configuration or setup. The alpha version of Grace is up and running on Linux and OSX. We are looking for someone with good C++/Windows chops to help us make whatever (hopefully small) changes are required to get Grace up and running on XP. If you comfortable in C++ and know about Window's compilers and would like to help out, PLEASE send us a note! The really great audio/graphics capabilities of JUCE will allow Grace to focus on integrating text-based algorithmic composition with graphics and realtime sound generation, rather than on pure programming as in the current Emacs/Slime environment. over time i will slowly be moving as much of CM's underlying tools (including SAL) as makes sense into C++ so that the same features are available outside of Lisp. CM's existing lisp code will always remain and i will continue to fix and improve lisp CM sources. But -- since that code is quite stable now -- most of my energy/time will be spent working on new tools that, for whatever reason, cant be provided in crossplatform lisp. Athough Grace is not ready for prime time a lot of essential features are there, at least in some capacity. RIght now the app contains: o A very cool splash screen (hey no point without one of these!) o An Editor with good support for SAL and Lisp hacking: syntax highlighting, Emacs emulation including Tab key syntactic indentation, word/sexpr motion (C-M-f etc), eval services and symbol help lookup in Sal buffers, errr..Editor Windows o Menu based loading of ASDF defined lisp system. o Ready-to-use algocomp tutorials available from the Help menu. o A Plotter window for data visualization, sequencing, and envelope design. plotter is only at the very beginning stage, bascially you can add some points, admire them and then use the Export command to turn the data into text (code) inside an editor window via paste. [Plotter development will be the main focus of the next (second stage) of work along with RTS sound output. any ideas you have on what a good visualization tool would include will be seriouly considered. for sure there will SAL mouse-hooks for algorithmicaly genereating material on mouse down clicks, also computing point layers from sal code, realtime plotter playback etc.] On osx or linux you can build Grace from cvs sources if you have an existing libjuce.a (version 4.3) and recent versions of gnu autoconf tools: cvs checkout grace cd grace ./configure --with-jucedir=/path/to/my/jucedir make make install cd /path/to/Grace/resourcedir cvs checkout cm PLEASE read the grace/INSTALL file before you attempt to build. if the sources dont build on your machine then please help us fix it by sending us the changes that get it working. otherwise wait till the dust settles and we can release binary drag'n drops. The file TODO has the current list of known bugs and a list of things that remain to be done. many many thanks to David Psenicka for adding the ./configure & automake magic! --rick From vze26m98 at optonline.net Mon Jul 16 09:45:39 2007 From: vze26m98 at optonline.net (Charles Turner) Date: Mon, 16 Jul 2007 12:45:39 -0400 Subject: [CM] [OT] Ladspa Jack Insert? Message-ID: Hi all- Apologies for the off-topic post, but I thought I could get a quick and definitive answer here w/o having to subscribe to anything new... Maybe I'm blind and stupid, but I can't seem to find a "Jack-insert" in LADSPA format. Jack-OSX has a VST equivalent. Is there something about the LADSPA architecture that I'm missing, or is there one out there and I'm just too lame to find it? Best and thanks! Charles Turner From joshp at u.washington.edu Mon Jul 16 10:51:51 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 10:51:51 -0700 Subject: [CM] grace Message-ID: Hi Rick, Just saw your email about grace. A couple of things came up when building on a MacBook Pro. First, I had to point the configure script to g++ for some reason: export CXX=g++; ./configure --with-jucedir=/Users/joshp/src/juce/ Then it configured fine (attached is the Makefile). However, make fails with: g++ -g -O2 -o grace Buffer.o Console.o Editor.o Grace.o Plotter.o Syntab.o Syntax.o FontList.o Layer.o Resources.o -L/Users/joshp/src/ juce//bin -framework Carbon -framework CoreServices -framework CoreAudio -framework CoreMidi -framework ApplicationServices - framework AGL -framework QuickTime -framework IOKit -ljuce /usr/bin/ld: Undefined symbols: LispConnection::isLispRunning() LispConnection::chooseASDF() LispConnection::sendLispSexpr(juce::String, int) LispConnection::isLispStartable() LispConnection::loadASDF(ASDF*) LispConnection::stopLisp() LispConnection::startLisp() LispConnection::LispConnection(ConsoleWindow*) ConfigureLispView::ConfigureLispView(LispConnection*) collect2: ld returned 1 exit status make[2]: *** [grace] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2 Any ideas? Thanks, Josh ?****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Makefile Type: application/octet-stream Size: 18966 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From joshp at u.washington.edu Mon Jul 16 10:55:16 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 10:55:16 -0700 Subject: [CM] Re: grace In-Reply-To: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: Sorry... On Jul 16, 2007, at 10:52 AM, cmdist-request at ccrma.Stanford.EDU wrote: > export CXX=3Dg++; ./configure --with-jucedir=3D/Users/joshp/src/juce/ should have been: export CXX=g++; ./configure --with-jucedir=/Users/joshp/src/juce/ ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From taube at uiuc.edu Mon Jul 16 11:21:50 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 13:21:50 -0500 Subject: [CM] grace In-Reply-To: References: Message-ID: it looks to me like -- for whatever reason -- the file src/Lisp.cpp is not getting compiled and/or included in the linking set.look in build/Makefile.am and check to see that its on the list. if it is see if it is at least getting compiled. not sure why you should have to export g++ by hand the ./configure looks for it. i wonder if you have some old version of automake. i think you need 1.9.6 or higher, not sure > Hi Rick, > > Just saw your email about grace. A couple of things came up when > building on a MacBook Pro. > > First, I had to point the configure script to g++ for some reason: > > export CXX=g++; ./configure --with-jucedir=/Users/joshp/src/juce/ > > Then it configured fine (attached is the Makefile). > > However, make fails with: > > g++ -g -O2 -o grace Buffer.o Console.o Editor.o Grace.o > Plotter.o Syntab.o Syntax.o FontList.o Layer.o Resources.o -L/Users/ > joshp/src/juce//bin -framework Carbon -framework CoreServices - > framework CoreAudio -framework CoreMidi -framework > ApplicationServices -framework AGL -framework QuickTime -framework > IOKit -ljuce > /usr/bin/ld: Undefined symbols: > LispConnection::isLispRunning() > LispConnection::chooseASDF() > LispConnection::sendLispSexpr(juce::String, int) > LispConnection::isLispStartable() > LispConnection::loadASDF(ASDF*) > LispConnection::stopLisp() > LispConnection::startLisp() > LispConnection::LispConnection(ConsoleWindow*) > ConfigureLispView::ConfigureLispView(LispConnection*) > collect2: ld returned 1 exit status > make[2]: *** [grace] Error 1 > make[1]: *** [all-recursive] Error 1 > make: *** [all] Error 2 > > Any ideas? > > Thanks, > > Josh > > > ****************************************** > Joshua Parmenter > University of Washington > Center for Digital Arts and Experimental Media > School of Music > Seattle, Washington 98195 > > http://www.realizedsound.net/josh/ > http://www.dxarts.washington.edu > > > From joshp at u.washington.edu Mon Jul 16 11:30:24 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 11:30:24 -0700 Subject: [CM] grace In-Reply-To: References: Message-ID: <85C12016-D709-45CA-9C5E-5AC337969C5A@u.washington.edu> Here is the list from Makefile.am. I see it there, but no .o file or .Po file seems to be generated. I have automake 1.10 from MacPorts. The computer is a 2.4 GHz MacBook Pro with OS 10.4.10 Thanks, Josh srcdir=@top_srcdir@/src VPATH=@top_srcdir@/src bin_PROGRAMS = grace # C/C++ files nodist_grace_SOURCES = Audio.h Buffer.h Console.h Editor.h Grace.h Plotter.h Syntab.h Syntax.h FontList.h Layer.h Resources.h Lisp.h \ Audio.cpp Buffer.cpp Console.cpp Editor.cpp Grace.cpp Plotter.cpp Syntab.cpp Syntax.cpp FontList.cpp Layer.cpp Resources.cpp Lisp.cpp # Lisp files nodist_pkgdata_DATA = grace.asd asdf.lisp grace.lisp gray.lisp socketserver.lisp # Compile instructions AM_CPPFLAGS = -DNDEBUG @OS_CFLAGS@ @JUCE_CFLAGS@ @X_CFLAGS@ grace_LDADD = @JUCE_LIBDIRS@ @DARWIN_LIBS@ @X_LIBS@ @JUCE_LIBS@ # Extra clean instructions if COMPILE_DARWIN clean-local: rm -rf Grace.app endif # Doc files nobase_data_DATA = @DOCUMENT_FILES@ # Make "all" also copy binary under app on OS X if COMPILE_DARWIN all-local: @mkdir -p Grace.app/Contents/MacOS cp grace Grace.app/Contents/MacOS/grace endif Best, Josh On Jul 16, 2007, at 11:21 AM, Rick Taube wrote: > it looks to me like -- for whatever reason -- the file src/Lisp.cpp > is not getting compiled and/or included in the linking set.look in > build/Makefile.am > > and check to see that its on the list. if it is see if it is at > least getting compiled. > not sure why you should have to export g++ by hand the ./configure > looks for it. > i wonder if you have some old version of automake. i think you need > 1.9.6 or higher, not sure > > >> Hi Rick, >> >> Just saw your email about grace. A couple of things came up when >> building on a MacBook Pro. >> >> First, I had to point the configure script to g++ for some reason: >> >> export CXX=g++; ./configure --with-jucedir=/Users/joshp/src/juce/ >> >> Then it configured fine (attached is the Makefile). >> >> However, make fails with: >> >> g++ -g -O2 -o grace Buffer.o Console.o Editor.o Grace.o >> Plotter.o Syntab.o Syntax.o FontList.o Layer.o Resources.o -L/ >> Users/joshp/src/juce//bin -framework Carbon -framework >> CoreServices -framework CoreAudio -framework CoreMidi -framework >> ApplicationServices -framework AGL -framework QuickTime -framework >> IOKit -ljuce >> /usr/bin/ld: Undefined symbols: >> LispConnection::isLispRunning() >> LispConnection::chooseASDF() >> LispConnection::sendLispSexpr(juce::String, int) >> LispConnection::isLispStartable() >> LispConnection::loadASDF(ASDF*) >> LispConnection::stopLisp() >> LispConnection::startLisp() >> LispConnection::LispConnection(ConsoleWindow*) >> ConfigureLispView::ConfigureLispView(LispConnection*) >> collect2: ld returned 1 exit status >> make[2]: *** [grace] Error 1 >> make[1]: *** [all-recursive] Error 1 >> make: *** [all] Error 2 >> >> Any ideas? >> >> Thanks, >> >> Josh >> >> >> ****************************************** >> Joshua Parmenter >> University of Washington >> Center for Digital Arts and Experimental Media >> School of Music >> Seattle, Washington 98195 >> >> http://www.realizedsound.net/josh/ >> http://www.dxarts.washington.edu >> >> >> ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From taube at uiuc.edu Mon Jul 16 11:32:29 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 13:32:29 -0500 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: if nothing works, try this handmade Makefile and tell me if that works. i cobbled it up after a fit of depression caused by reading the automake manual make -f Makefile.osx JUCEDIR=/path/to/juce make -f Makefile.osc install -------------- next part -------------- A non-text attachment was scrubbed... Name: Makefile.osx Type: application/octet-stream Size: 2736 bytes Desc: not available URL: From taube at uiuc.edu Mon Jul 16 11:41:12 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 13:41:12 -0500 Subject: [CM] grace In-Reply-To: <85C12016-D709-45CA-9C5E-5AC337969C5A@u.washington.edu> References: <85C12016-D709-45CA-9C5E-5AC337969C5A@u.washington.edu> Message-ID: <656319F0-7080-4139-98EE-D3394D9EB3B5@uiuc.edu> josh forgot to say you will have to put that Makefile.osx in the build/ subdir From joshp at u.washington.edu Mon Jul 16 11:42:53 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 11:42:53 -0700 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: <6BAE271E-52A4-4335-8CDC-7C9CAA8E2915@u.washington.edu> Not really working either :( d-128-95-92-42:~/src/grace joshp$ make -f Makefile.osx JUCEDIR=/Users/ joshp/src/juce make: *** No rule to make target `Audio.cpp', needed by `Audio.o'. Stop. Thanks, Josh On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: > if nothing works, try this handmade Makefile and tell me if that > works. i cobbled it up after a fit of depression caused by reading > the automake manual > > make -f Makefile.osx JUCEDIR=/path/to/juce > > make -f Makefile.osc install > > ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From testcase at asu.edu Mon Jul 16 11:44:05 2007 From: testcase at asu.edu (todd ingalls) Date: Mon, 16 Jul 2007 11:44:05 -0700 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: when i check out a fresh copy i have usually had to run autoconf to get the right Makefiles. On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: > if nothing works, try this handmade Makefile and tell me if that > works. i cobbled it up after a fit of depression caused by reading > the automake manual > > make -f Makefile.osx JUCEDIR=/path/to/juce > > make -f Makefile.osc install > -------------- next part -------------- A non-text attachment was scrubbed... Name: Makefile.osx Type: application/octet-stream Size: 2736 bytes Desc: not available URL: -------------- next part -------------- From joshp at u.washington.edu Mon Jul 16 11:45:01 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 11:45:01 -0700 Subject: [CM] grace In-Reply-To: <656319F0-7080-4139-98EE-D3394D9EB3B5@uiuc.edu> References: <85C12016-D709-45CA-9C5E-5AC337969C5A@u.washington.edu> <656319F0-7080-4139-98EE-D3394D9EB3B5@uiuc.edu> Message-ID: <684CB9D0-5086-4A0C-9700-52425C6D668A@u.washington.edu> Got it! d-128-95-92-42:~/src/grace/build joshp$ make -f Makefile.osx JUCEDIR=/ Users/joshp/src/juce/ CXX=g++ g++ -c -I../src -I/Users/joshp/src/juce/ -g -O2 -DNDEBUG - DJUCE_MAC ../src/Audio.cpp g++ -c -I../src -I/Users/joshp/src/juce/ -g -O2 -DNDEBUG - DJUCE_MAC ../src/Lisp.cpp g++ -o grace Audio.o Buffer.o Console.o Editor.o FontList.o Grace.o Layer.o Lisp.o Plotter.o Resources.o Syntab.o Syntax.o -L/Users/joshp/ src/juce//bin -framework Carbon -framework CoreServices -framework CoreAudio -framework CoreMidi -framework ApplicationServices - framework AGL -framework QuickTime -framework IOKit -ljuce Still had to set the CXX var though. Seems to boot fine though! Josh On Jul 16, 2007, at 11:41 AM, Rick Taube wrote: > josh forgot to say you will have to put that Makefile.osx in the > build/ subdir ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From joshp at u.washington.edu Mon Jul 16 11:55:19 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 11:55:19 -0700 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> So... I had to hand make and copy grace.asd to Grace.app/Contents/ Resources/grace I had to make the Resources and grace directories. Then, when I tried start lisp: Launching /usr/local/bin/sbcl --eval '(load "/Users/joshp/src/grace/ build/Grace.app/Contents/Resources/grace/grace.asd")' --eval '(asdf:oos (quote asdf:load-op) "grace")' --eval '(grace:start-server 8000 "/Users/joshp/Library/Caches/Grace/temp0grace")' Polling socket server..................... =:( Connection failed. The LISP config wouldn't let me point to the source outside the directory. Thanks, Josh On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: > if nothing works, try this handmade Makefile and tell me if that > works. i cobbled it up after a fit of depression caused by reading > the automake manual > > make -f Makefile.osx JUCEDIR=/path/to/juce > > make -f Makefile.osc install > > ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From joshp at u.washington.edu Mon Jul 16 11:55:27 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 11:55:27 -0700 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> Message-ID: <87FD7F2A-C6ED-4162-B0D8-6B0314508057@u.washington.edu> Thanks... I'll try that also. Josh On Jul 16, 2007, at 11:44 AM, todd ingalls wrote: > when i check out a fresh copy i have usually had to run autoconf to > get the right Makefiles. > > > On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: > >> if nothing works, try this handmade Makefile and tell me if that >> works. i cobbled it up after a fit of depression caused by reading >> the automake manual >> >> make -f Makefile.osx JUCEDIR=/path/to/juce >> >> make -f Makefile.osc install >> >> > ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From taube at uiuc.edu Mon Jul 16 12:04:34 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 14:04:34 -0500 Subject: [CM] Re: grace In-Reply-To: <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> Message-ID: <67BD3564-20DD-4A40-A133-E761F61B6792@uiuc.edu> On Jul 16, 2007, at 1:55 PM, Joshua Parmenter wrote: > So... I had to hand make and copy grace.asd to Grace.app/Contents/ > Resources/grace you need more than that file! if you are using the handmade Makefile it has an install rule: cd build make -f Makefile.osx install other wise do this cd build mkdir -p Grace.app/Contents/Resources/grace cp ../src/*.{asd,lisp} Grace.app/Contents/Resources/grace cd Grace.app/Contents/Resources/ cvs checkout cm > > Then, when I tried start lisp: > > Launching /usr/local/bin/sbcl --eval '(load "/Users/joshp/src/grace/ > build/Grace.app/Contents/Resources/grace/grace.asd")' --eval > '(asdf:oos (quote asdf:load-op) "grace")' --eval '(grace:start- > server 8000 "/Users/joshp/Library/Caches/Grace/temp0grace")' > Polling socket server..................... =:( > Connection failed. it looks like /usr/local/bin/sbcl isnt booting, or the .asd isnt being loaded for some reason? start the app from the terminal as described in INSTALL and then look to see whats happening to sbcl as it boots. > > The LISP config wouldn't let me point to the source outside the > directory. not sure what you mean? you can set the Lisp systems directory to any directory in the config windows. Grace will look there first for the grace/grace.asd file, then under the resrouce direcory. you can load any ASDF "by hand" using Lisp>Load System>Load... menu item sbcl works fine on my macs and my planetccrma box. > Thanks, > > Josh > > > > On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: > >> if nothing works, try this handmade Makefile and tell me if that >> works. i cobbled it up after a fit of depression caused by reading >> the automake manual >> >> make -f Makefile.osx JUCEDIR=/path/to/juce >> >> make -f Makefile.osc install >> >> > > ****************************************** > Joshua Parmenter > University of Washington > Center for Digital Arts and Experimental Media > School of Music > Seattle, Washington 98195 > > http://www.realizedsound.net/josh/ > http://www.dxarts.washington.edu > > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From joshp at u.washington.edu Mon Jul 16 12:13:54 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 12:13:54 -0700 Subject: [CM] Re: grace In-Reply-To: <67BD3564-20DD-4A40-A133-E761F61B6792@uiuc.edu> References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> <67BD3564-20DD-4A40-A133-E761F61B6792@uiuc.edu> Message-ID: ergh... forgot that step. Sorry. LISP now starts and evaluate, but it can't find CM Now, where does it expect all my CM code to be? Also... is there a way to save the LISP configure options? Thanks, Josh On Jul 16, 2007, at 12:04 PM, Rick Taube wrote: > > On Jul 16, 2007, at 1:55 PM, Joshua Parmenter wrote: > >> So... I had to hand make and copy grace.asd to Grace.app/Contents/ >> Resources/grace > > you need more than that file! if you are using the handmade > Makefile it has an install rule: > > cd build > make -f Makefile.osx install > > other wise do this > > cd build > mkdir -p Grace.app/Contents/Resources/grace > cp ../src/*.{asd,lisp} Grace.app/Contents/Resources/grace > cd Grace.app/Contents/Resources/ > cvs checkout cm > >> >> Then, when I tried start lisp: >> >> Launching /usr/local/bin/sbcl --eval '(load "/Users/joshp/src/ >> grace/build/Grace.app/Contents/Resources/grace/grace.asd")' --eval >> '(asdf:oos (quote asdf:load-op) "grace")' --eval '(grace:start- >> server 8000 "/Users/joshp/Library/Caches/Grace/temp0grace")' >> Polling socket server..................... =:( >> Connection failed. > > it looks like /usr/local/bin/sbcl isnt booting, or the .asd isnt > being loaded for some reason? > > start the app from the terminal as described in INSTALL and then > look to see whats happening to sbcl as it boots. > >> >> The LISP config wouldn't let me point to the source outside the >> directory. > > not sure what you mean? you can set the Lisp systems directory to > any directory in the config windows. Grace will look there first > for the grace/grace.asd file, then under the resrouce direcory. > you can load any ASDF "by hand" using Lisp>Load System>Load... menu > item > > sbcl works fine on my macs and my planetccrma box. > > > >> Thanks, >> >> Josh >> >> >> >> On Jul 16, 2007, at 11:32 AM, Rick Taube wrote: >> >>> if nothing works, try this handmade Makefile and tell me if that >>> works. i cobbled it up after a fit of depression caused by >>> reading the automake manual >>> >>> make -f Makefile.osx JUCEDIR=/path/to/juce >>> >>> make -f Makefile.osc install >>> >>> >> >> ****************************************** >> Joshua Parmenter >> University of Washington >> Center for Digital Arts and Experimental Media >> School of Music >> Seattle, Washington 98195 >> >> http://www.realizedsound.net/josh/ >> http://www.dxarts.washington.edu >> >> >> >> _______________________________________________ >> Cmdist mailing list >> Cmdist at ccrma.stanford.edu >> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From taube at uiuc.edu Mon Jul 16 12:18:32 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 16 Jul 2007 14:18:32 -0500 Subject: [CM] Re: grace In-Reply-To: References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> <67BD3564-20DD-4A40-A133-E761F61B6792@uiuc.edu> Message-ID: <4DE1FF3E-CBF4-4739-B932-9A76E1149425@uiuc.edu> On Jul 16, 2007, at 2:13 PM, Joshua Parmenter wrote: > Now, where does it expect all my CM code to be? See the INSTALL. it looks for a 'lisp systems directory, which defaults to the resources directory. use Console>Lisp>Configure Lisp> window to set the Lisp systems directory, its in the lower part of that window. Or use Console>Lisp>Load System>Load... to 'hand load' any system, just use the dialog to find the .asd file. its not savable yet, we need a Preferences... feature badly! sorry. From joshp at u.washington.edu Mon Jul 16 12:25:18 2007 From: joshp at u.washington.edu (Joshua Parmenter) Date: Mon, 16 Jul 2007 12:25:18 -0700 Subject: [CM] Re: grace In-Reply-To: <4DE1FF3E-CBF4-4739-B932-9A76E1149425@uiuc.edu> References: <20070716175202.10903.75348.Mailman@cm-mail.stanford.edu> <6F7F5047-5A0D-4E07-88AF-CD9B39418D3F@u.washington.edu> <67BD3564-20DD-4A40-A133-E761F61B6792@uiuc.edu> <4DE1FF3E-CBF4-4739-B932-9A76E1149425@uiuc.edu> Message-ID: Great... that has it up and running! Many thanks, Josh On Jul 16, 2007, at 12:18 PM, Rick Taube wrote: > > On Jul 16, 2007, at 2:13 PM, Joshua Parmenter wrote: > >> Now, where does it expect all my CM code to be? > > See the INSTALL. it looks for a 'lisp systems directory, which > defaults to the resources directory. > use Console>Lisp>Configure Lisp> window to set the Lisp systems > directory, its in the lower part of that window. > Or use Console>Lisp>Load System>Load... to 'hand load' any > system, just use the dialog to find the .asd file. > > its not savable yet, we need a Preferences... feature badly! sorry. > > ****************************************** Joshua Parmenter University of Washington Center for Digital Arts and Experimental Media School of Music Seattle, Washington 98195 http://www.realizedsound.net/josh/ http://www.dxarts.washington.edu From applemk at stanford.edu Fri Jul 20 13:00:37 2007 From: applemk at stanford.edu (Mark Applebaum) Date: Fri, 20 Jul 2007 13:00:37 -0700 Subject: [CM] Stanford University Faculty Composition Search Message-ID: Dear Colleagues, I am pleased to announce that the Department of Music at Stanford University is inviting applications for the position of tenure-track composer at the rank of Assistant Professor. The receipt deadline for applications is October 1, 2007. Attached is a .pdf of the formal announcement. The prose appears below as well. I would be grateful if you would circulate this widely. (My apologies to individuals for any cross-listings.) Many thanks, Mark Applebaum, Search Committee Chair ****************************************** Notice of Vacancy COMPOSER The Department of Music at Stanford University is inviting applications for the position of tenure-track composer at the rank of Assistant Professor. The envisaged starting date is September 1, 2008. Candidates should be, or show every promise of becoming, distinguished composers. The successful candidate should also have strong teaching qualifications as well as interest and expertise in one or more additional areas of performance and/or research, including, but not limited to, music theory, music performance, computer music, and non-Western music. Teaching responsibilities will involve courses primarily in analysis and composition, at both undergraduate and graduate levels, as well as the supervision of individual composition research. Courses may also be taught in the candidate's additional areas of expertise. Stanford University is an affirmative-action, equal-opportunity employer. Letters of application, together with a curriculum vitae, list of works and performances, and three to five letters of reference, should be received on or before October 1, 2007, and sent to: Dr. Mark Applebaum, Composer Search Committee Chair Attn: Debbie Barney Department of Music Stanford University Stanford, CA 94305-3076 Other supporting materials will be requested at a later date. Please do NOT send scores or recordings at this stage of the search. ******************************************* -- ******************************************* Mark Applebaum, Ph.D. Assistant Professor of Composition & Theory Department of Music Stanford University Stanford, CA 94305-3076 (650)723-1656 http://www.markapplebaum.com ******************************************* -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ComposerSearch_1.pdf Type: application/pdf Size: 83162 bytes Desc: not available URL: From j.v.o.koskinen at gmail.com Sun Jul 22 07:42:39 2007 From: j.v.o.koskinen at gmail.com (Ville Koskinen) Date: Sun, 22 Jul 2007 17:42:39 +0300 Subject: [CM] Snd: matching utterances with a DTW-like algorithm Message-ID: <392b6b160707220742x7c820b16j7f847af299d1ccd5@mail.gmail.com> Hello, I've been pondering the possibility of scripting Snd to automatically detect reader errors from spoken word audio. Usually the reading is like this: "At the beginning of July, during a spell of exceptionally hot WOTHER, ... ... , during a spell of exceptionally hot weather ..." That is, most of the time the errors in the reading are followed by a period of silence, and after that there is an overlap to the already recorded part. I've found out that in speech recognition an algorithm called "dynamic time warping" is often used to match utterances to templates. With DTW you should be able to match a spectrogram of "speech" to the spectrogram of "sspeechh". I wonder if Snd could be scripted to: 1) create a local spectrogram from the data at the beginning of the reader's correction (assuming that long pause is always followed by a correction) 2) go backwards to beginnings of previous sentences (this I can already do) and to create local spectrograms from these positions 3) compare past spectrograms to the one from the correction in order to detect the overlap. I wonder if anyone has tried to implement anything like this (or a subset)? Any suggestions? Kind Regards, Ville Koskinen From taube at uiuc.edu Mon Jul 23 11:13:57 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 23 Jul 2007 13:13:57 -0500 Subject: [CM] cvs grace Message-ID: <53F70F00-1021-47FB-A2AE-7A9C312DAA14@uiuc.edu> grace cvs now has a "preferences" feature working that store settings between sessions and makes configuring lisp more simple. on linu preferences are stored in: ~/.Grace/grace.prefs on darwin the preferences are stored in: ~/Library/Preferences/grace.prefs also new: * Native titlebars are now turned on by default so Grace looks more like a "native" app out-of-the-box. Use Windows>Native Titlebars to toggle beween native and JUCE style. If you want to toggle between styles on lunix you will need the svn version of JUCE as there is a workaround to an X windows issue that toggling titlebars triggers. * A new File>Open Recent> submenu holds (upto) 10 recently opened files * A "Launch Lisp at Startup" checkbox is now part of Configure Lisp dialog and stored in the preference file. * Console>Lisp>Load File> now works. * Recent ASDF systems and Lisp files you load are now remembered between sessions * ASDF and LISP info are now internally Xml objects so it will be soon be possible to import customized configuration. From benmca at gmail.com Tue Jul 24 09:56:35 2007 From: benmca at gmail.com (Ben McAllister) Date: Tue, 24 Jul 2007 09:56:35 -0700 Subject: [CM] CM 1.x Message-ID: <860f9f110707240956r6d743e1cn50ee386c28e2728@mail.gmail.com> Hi - I've got some old pieces I'd like to revist which were written against CM 1.x (1.3 I think). Is there a resting place somewhere online for these older versions of CM? Thanks - Ben Mcallister http://www.listenfaster.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From bil at ccrma.Stanford.EDU Tue Jul 24 13:13:20 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 24 Jul 2007 13:13:20 -0700 Subject: [CM] CM 1.x In-Reply-To: <860f9f110707240956r6d743e1cn50ee386c28e2728@mail.gmail.com> References: <860f9f110707240956r6d743e1cn50ee386c28e2728@mail.gmail.com> Message-ID: <20070724201230.M36283@ccrma.Stanford.EDU> I think they are accessible at ccrma-ftp:/pub/Lisp/cm/sources. cm 1.2 to the latest. From benmca at gmail.com Tue Jul 24 13:21:37 2007 From: benmca at gmail.com (Ben McAllister) Date: Tue, 24 Jul 2007 13:21:37 -0700 Subject: [CM] CM 1.x In-Reply-To: <20070724201230.M36283@ccrma.Stanford.EDU> References: <860f9f110707240956r6d743e1cn50ee386c28e2728@mail.gmail.com> <20070724201230.M36283@ccrma.Stanford.EDU> Message-ID: <860f9f110707241321v557b37desc1900ec7f75ce145@mail.gmail.com> Thanks so much! Ben On 7/24/07, Bill Schottstaedt wrote: > > I think they are accessible at ccrma-ftp:/pub/Lisp/cm/sources. > cm 1.2 to the latest. > > > -- -- http://www.listenfaster.com http://www.wizardprison.com http://www.myspace.com/cmoncmonband -------------- next part -------------- An HTML attachment was scrubbed... URL: From landspeedrecord at gmail.com Tue Jul 24 17:56:23 2007 From: landspeedrecord at gmail.com (Landspeedrecord) Date: Tue, 24 Jul 2007 20:56:23 -0400 Subject: [CM] translating spirals into music CLM In-Reply-To: <1184279442.3713.40.camel@strawberry> References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> Message-ID: I don't know if it is moot at this point, but I have been thinking about music spiraling and shephard tones were the first thing I thought of, but I rejected it as an analog because I felt they aren't really a spiral any more than watching the stripes on a barber pole twist around is. Since then I have not come up with much except that it could be done using panning and volume/reverb (with a little doppler effect to add spice?) - this would simulate a circular motion through the listening field in front of the listener with the volume rising on on pass from left to right and falling on the trip back from right to left. Reverb would rise as the volume fell to simulate distance from the listener as the sound was "receeding". The problem is there are lots of musical parameters that can be assigned to x and y axes but very few that together imply "directionality" which seems to me to be a necessary component of visually perceiving a spiral. As a spiral is drawn in 2D the eye doesn't see it as two seperate orthagonal vectors but as one contracting circular movement around a central point. Blah blah blah. I'm done ranting. I am just bugged because I haven't made much headway on the idea. Any thoughts or ideas anyone can throw my way? On 7/12/07, Juan I Reyes wrote: > > > There's also an example of shepard tones in the ws.scm section > > of sndscm.html, complete with a purty picture that took me about 10 > > times longer to make than the sound. > > Here is my CLM rendition of Shepard Tones based on Bill's ws.scm and > from Charles Dodge's Computer Music text. > > (definstrument shepard (beg dur amp &key > (incr .000001) > ) > (let* ((start (floor (* beg *srate*))) > (end (+ start (floor (* dur *srate*)))) > (x 0.0) > (arr (make-array 12))) ;; we'll create a 12 oscil bank > ;; > (do ((i 0 (1+ i))) ;; initialize oscils > ((= i 12)) > (setf (aref arr i) (make-oscil :frequency 0.0)) ) > ;; > (run > (loop for i from start to end do > (let ((y 0.00) > (oscbank 0.0)) > ;; sum signal of twelve oscillators > (do ((i 0 (1+ i))) > ((= i (length arr))) > ;; get amplitude and oscillator phase offset > (let ((phoffset (+ x (/ i 12)))) > (if (> phoffset 1) (setf phoffset (- phoffset 1))) > (setf y (- 4.0 (* 8.0 phoffset))) > (incf oscbank (* (exp (* -0.5 y y)) > (oscil (aref arr i) > (hz->radians (expt 2.0 (* (- 1 phoffset) 12.0)))))) > ;; to go up ;;;(* phoffset 12.0) > )) > (incf x incr) > (outa i (* amp oscbank)) ))) > )) > > ;;;(with-sound () (shepard 0 10 0.1)) > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From glauberalex at uol.com.br Tue Jul 24 21:23:44 2007 From: glauberalex at uol.com.br (Glauber Prado) Date: Wed, 25 Jul 2007 01:23:44 -0300 Subject: [CM] translating spirals into music CLM In-Reply-To: References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> Message-ID: <46A6D050.2000806@uol.com.br> maybe lsystem for the score? :D take a look for some inspiration: http://br.youtube.com/watch?v=PZNvjbXPVSI Landspeedrecord escreveu: > I don't know if it is moot at this point, but I have been thinking > about music spiraling and shephard tones were the first thing I > thought of, but I rejected it as an analog because I felt they aren't > really a spiral any more than watching the stripes on a barber pole > twist around is. > > Since then I have not come up with much except that it could be done > using panning and volume/reverb (with a little doppler effect to add > spice?) - this would simulate a circular motion through the listening > field in front of the listener with the volume rising on on pass from > left to right and falling on the trip back from right to left. Reverb > would rise as the volume fell to simulate distance from the listener > as the sound was "receeding". > > The problem is there are lots of musical parameters that can be > assigned to x and y axes but very few that together imply > "directionality" which seems to me to be a necessary component of > visually perceiving a spiral. As a spiral is drawn in 2D the eye > doesn't see it as two seperate orthagonal vectors but as one > contracting circular movement around a central point. > > Blah blah blah. I'm done ranting. I am just bugged because I haven't > made much headway on the idea. Any thoughts or ideas anyone can throw > my way? > On 7/12/07, Juan I Reyes wrote: >> >> > There's also an example of shepard tones in the ws.scm section >> > of sndscm.html, complete with a purty picture that took me about 10 >> > times longer to make than the sound. >> >> Here is my CLM rendition of Shepard Tones based on Bill's ws.scm and >> from Charles Dodge's Computer Music text. >> >> (definstrument shepard (beg dur amp &key >> (incr .000001) >> ) >> (let* ((start (floor (* beg *srate*))) >> (end (+ start (floor (* dur *srate*)))) >> (x 0.0) >> (arr (make-array 12))) ;; we'll create a 12 oscil bank >> ;; >> (do ((i 0 (1+ i))) ;; initialize oscils >> ((= i 12)) >> (setf (aref arr i) (make-oscil :frequency 0.0)) ) >> ;; >> (run >> (loop for i from start to end do >> (let ((y 0.00) >> (oscbank 0.0)) >> ;; sum signal of twelve oscillators >> (do ((i 0 (1+ i))) >> ((= i (length arr))) >> ;; get amplitude and oscillator phase offset >> (let ((phoffset (+ x (/ i 12)))) >> (if (> phoffset 1) (setf phoffset (- phoffset 1))) >> (setf y (- 4.0 (* 8.0 phoffset))) >> (incf oscbank (* (exp (* -0.5 y y)) >> (oscil (aref arr i) >> (hz->radians (expt 2.0 (* (- 1 >> phoffset) 12.0)))))) >> ;; to go up ;;;(* phoffset >> 12.0) >> )) >> (incf x incr) >> (outa i (* amp oscbank)) ))) >> )) >> >> ;;;(with-sound () (shepard 0 10 0.1)) >> >> _______________________________________________ >> Cmdist mailing list >> Cmdist at ccrma.stanford.edu >> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist >> > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From johnhenrydale at gmail.com Tue Jul 24 22:51:58 2007 From: johnhenrydale at gmail.com (john henry dale) Date: Wed, 25 Jul 2007 01:51:58 -0400 Subject: [CM] translating spirals into music CLM In-Reply-To: References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> Message-ID: Ok, does anyone have any insight on how/what I would use in CLM to send panning information to a 5 channel speaker system with a tweeter tree speaker sitting above the audience in the center of the room ? Would that tweeter tree be a Z axis for panning coordinates ? I have basically figured out the timing, tonal and rhythmic components for the "spiral composition" by using fibonacci numbers as time marker points in the composition to increase the speed of the rhythm and the note frequencies logarithmically. But what I can't figure out is how to map sound output to a 5 channel system such that the sound will seemingly go in circles around the room (and audience), starting from sub-audible frequencies (1 hz, 2 hz, 3 hz, 5 hz, 8 hz etc) from "below" the floor, so to speak, and continuing to spiral "upward" until reaching a climax point (which will have all 5 channels going at full amplitude) and then "descending" back down again. Would it perhaps be better to use an 9 channel system to do this, with 4 speakers on the floor and 4 at ear height ? anyway, i looked am assuming that loc-sig is where i would start with this but would truly appreciate any tips or shortcuts people may have in terms of how to coordinate the rhythm and timing of the composition with the panning effects I mentioned earlier. Thanks, John Henry Dale -------------- next part -------------- An HTML attachment was scrubbed... URL: From bil at ccrma.Stanford.EDU Wed Jul 25 11:22:54 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 25 Jul 2007 11:22:54 -0700 Subject: [CM] translating spirals into music CLM In-Reply-To: References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> Message-ID: <20070725181300.M12604@ccrma.Stanford.EDU> > Ok, does anyone have any insight on how/what I would use in CLM to send > panning information to a 5 channel speaker system with a tweeter tree > speaker sitting above the audience in the center of the room ? If the individual notes are in one place, you could do something like: (with-sound (:channels 5) (let ((loc (make-locsig :channels 4 :output *output*)) (osc (make-oscil 440.0)) (j 0)) (run (lambda () (do ((i 0 (1+ i))) ((= i 360)) (do ((k 0 (1+ k))) ((= k 1000)) (let ((sig (* .5 (oscil osc)))) (locsig loc j sig) (out-any j sig 4 *output*) (set! j (1+ j)))) (move-locsig loc (exact->inexact i) 1.0)))))) This sends out 360 notes a degree at a time -- locsig is used for the circle in 4 chans, and the 5th chan is just out-any. dlocsig can move a single sound (panning, Doppler, etc). I don't know how convincing this would be with only 5 speakers. From johnhenrydale at gmail.com Thu Jul 26 03:37:02 2007 From: johnhenrydale at gmail.com (john henry dale) Date: Thu, 26 Jul 2007 11:37:02 +0100 Subject: [CM] translating spirals into music CLM In-Reply-To: <20070725181300.M12604@ccrma.Stanford.EDU> References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> <20070725181300.M12604@ccrma.Stanford.EDU> Message-ID: Thanks very much for that Bill, can you tell me how I write this if I were to use the 9 speaker set up I described earlier? Would I need a soundcard with 5 (or 9) outs to be able to achieve this 360 degree effect ? Thanks, JH On 7/25/07, Bill Schottstaedt wrote: > > > Ok, does anyone have any insight on how/what I would use in CLM to send > > panning information to a 5 channel speaker system with a tweeter tree > > speaker sitting above the audience in the center of the room ? > > If the individual notes are in one place, you could do something like: > > (with-sound (:channels 5) > (let ((loc (make-locsig :channels 4 :output *output*)) > (osc (make-oscil 440.0)) > (j 0)) > (run > (lambda () > (do ((i 0 (1+ i))) > ((= i 360)) > (do ((k 0 (1+ k))) > ((= k 1000)) > (let ((sig (* .5 (oscil osc)))) > (locsig loc j sig) > (out-any j sig 4 *output*) > (set! j (1+ j)))) > (move-locsig loc (exact->inexact i) 1.0)))))) > > This sends out 360 notes a degree at a time -- locsig is used for the > circle > in 4 chans, and the 5th chan is just out-any. dlocsig can move a single > sound (panning, Doppler, etc). I don't know how convincing this would be > with only 5 speakers. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bil at ccrma.Stanford.EDU Thu Jul 26 07:39:46 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 26 Jul 2007 07:39:46 -0700 Subject: [CM] translating spirals into music CLM In-Reply-To: References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> <20070725181300.M12604@ccrma.Stanford.EDU> Message-ID: <20070726143230.M10013@ccrma.Stanford.EDU> > can you tell me how I write this if I were to use the 9 speaker set up I described earlier? > Would I need a soundcard with 5 (or 9) outs to be able to achieve this 360 degree effect ? Use 2 4-channel locsigs, one for the floor, the other for the ceiling; write the locsig output to a frame, rather than directly to the output file, so that you can divvy it up to arbitrary output channels. I don't know about the other question -- I think to get a good sound placement illusion you need more speakers than you can imagine -- someone in SF has a listening space with hundreds of speakers -- a sign that this problem needs more thought... (It is odd, to me anyway, that the sound placement illusion is the only audio illusion I've ever heard that can be as compelling as the visual illusions -- shepard tones never work unless you're willing to go along, but a cricket in a box can be incredibly hard to locate -- as you (unwittingly) move around the box, it seems to be moving around with you, but to find it, you have to willfully ignore your ears. And yet, in computers, nothing in this regard has ever struck me as convincing.) From nando at ccrma.Stanford.EDU Thu Jul 26 10:44:37 2007 From: nando at ccrma.Stanford.EDU (Fernando Lopez-Lezcano) Date: Thu, 26 Jul 2007 10:44:37 -0700 Subject: [CM] translating spirals into music CLM In-Reply-To: References: <07BEA964-26E6-4DDC-B8E9-1A0B49398BF2@uiuc.edu> <20070711231905.M46801@ccrma.Stanford.EDU> <1184279442.3713.40.camel@strawberry> Message-ID: <1185471877.31654.5.camel@cmn3.stanford.edu> On Wed, 2007-07-25 at 01:51 -0400, john henry dale wrote: > Ok, does anyone have any insight on how/what I would use in CLM to > send panning information to a 5 channel speaker system with a tweeter > tree speaker sitting above the audience in the center of the room ? You could also take a look at dlocsig (which comes with clm - wrote it a long time ago). With it you can define "paths" an object will take when moving in space and then use dlocsig to render that movement. Dlocsig is sort of a unit generator you can include in your own instruments. Or you can use the example soundfile ins to move a soundfile around. It is not easy to use :-) but it can deal with arbitrary speaker configurations. It can render using VBAP (amplitude panning) or first order Ambisonics. -- Fernando > Would that tweeter tree be a Z axis for panning coordinates ? I have > basically figured out the timing, tonal and rhythmic components for > the "spiral composition" by using fibonacci numbers as time marker > points in the composition to increase the speed of the rhythm and the > note frequencies logarithmically. But what I can't figure out is how > to map sound output to a 5 channel system such that the sound will > seemingly go in circles around the room (and audience), starting from > sub-audible frequencies (1 hz, 2 hz, 3 hz, 5 hz, 8 hz etc) from > "below" the floor, so to speak, and continuing to spiral "upward" > until reaching a climax point (which will have all 5 channels going at > full amplitude) and then "descending" back down again. Would it > perhaps be better to use an 9 channel system to do this, with 4 > speakers on the floor and 4 at ear height ? anyway, i looked am > assuming that loc-sig is where i would start with this but would truly > appreciate any tips or shortcuts people may have in terms of how to > coordinate the rhythm and timing of the composition with the panning > effects I mentioned earlier. Thanks, > John Henry Dale From k.s.matheussen at notam02.no Sat Jul 28 07:17:00 2007 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Sat, 28 Jul 2007 16:17:00 +0200 (CEST) Subject: [CM] New version of snd-ls Message-ID: Hi, I just uploaded a new version (0.9.8.0_beta) of snd-ls at the usual place, http://www.notam02.no/arkiv/src/snd/ . It contains Snd v9.3 from 30-Jul-20007. (downloaded 27-Jul-2007, weird. :-) ) I think this one should fix the problem for those who got an error about the missing rt_readin_tag during startup. From taube at uiuc.edu Sat Jul 28 11:22:48 2007 From: taube at uiuc.edu (Rick Taube) Date: Sat, 28 Jul 2007 13:22:48 -0500 Subject: [CM] grace build uses scons Message-ID: <6E169E02-78AB-441A-AFA7-775E96BB99E0@uiuc.edu> ive switched grace from gnu's configure/make to SCons: cvs checkout grace cd grace scons JUCEDIR=/path/to/juce scons install of course, you need to have scons for this to work: http://www.scons.org/download.php http://www.python.org/download/ you only need to specify the JUCEDIR the first time you build. ive tested on planetccrma, mac/ppc mac/intel. From bil at ccrma.Stanford.EDU Sat Jul 28 11:47:11 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 28 Jul 2007 11:47:11 -0700 Subject: [CM] New version of snd-ls In-Reply-To: References: Message-ID: <20070728184613.M84280@ccrma.Stanford.EDU> > It contains Snd v9.3 from 30-Jul-20007. > (downloaded 27-Jul-2007, weird. ) I can explain that -- July 14 is my birthday. From edeleflie at gmail.com Sun Jul 29 02:59:56 2007 From: edeleflie at gmail.com (e deleflie) Date: Sun, 29 Jul 2007 19:59:56 +1000 Subject: [CM] applying HRTFs for binaural rendering Message-ID: <9a471d320707290259o1997a0a5q7ddb8d3d59956313@mail.gmail.com> Hi all, I'm wondering if anyone knows of any HRTF rendering functions written in snd? I'm looking for something where you pass in the sound, the azimuth and elevation, left or right ear, etc. etc... I've found this: http://ccrma.stanford.edu/~gsell/220b/hrtf-file-rewrites.ins ... but it doesn't look recent or maintained... Eteienne From k.s.matheussen at notam02.no Sun Jul 29 10:06:16 2007 From: k.s.matheussen at notam02.no (Kjetil Svalastog Matheussen) Date: Sun, 29 Jul 2007 19:06:16 +0200 (CEST) Subject: [CM] New version of snd-ls In-Reply-To: <20070728184613.M84280@ccrma.Stanford.EDU> References: <20070728184613.M84280@ccrma.Stanford.EDU> Message-ID: On Sat, 28 Jul 2007, Bill Schottstaedt wrote: > > It contains Snd v9.3 from 30-Jul-20007. > > (downloaded 27-Jul-2007, weird. ) > > I can explain that -- July 14 is my birthday. > I don't understand... (You don't want to release a new version of Snd on friday the thirteenth? (day after your birthday)) From bil at ccrma.Stanford.EDU Sun Jul 29 10:34:19 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sun, 29 Jul 2007 10:34:19 -0700 Subject: [CM] New version of snd-ls In-Reply-To: References: <20070728184613.M84280@ccrma.Stanford.EDU> Message-ID: <20070729172714.M76229@ccrma.Stanford.EDU> No, it was just a joke -- an elephant joke -- sorry. To atone, I've finished a complete rewrite of the CLM docs: sndclm.html. From k.s.matheussen at notam02.no Sun Jul 29 16:04:56 2007 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Mon, 30 Jul 2007 01:04:56 +0200 (CEST) Subject: [CM] New version of snd-ls In-Reply-To: References: Message-ID: On Sat, 28 Jul 2007, Kjetil S. Matheussen wrote: > > Hi, > > I just uploaded a new version (0.9.8.0_beta) of snd-ls at the usual place, > http://www.notam02.no/arkiv/src/snd/ . > 0.9.8.1_beta is uploaded. It contains a small fix for a bug that caused the startup of Snd to fail in case it couldn't open any previously used soundfile. Thanks to Ville Koskinen for reporting the bug. > I think this one should fix the problem for those who got an error about the > missing rt_readin_tag during startup. > Anyone knows if its working now? I can't reproduce the problem myself. From taube at uiuc.edu Sun Jul 29 17:27:30 2007 From: taube at uiuc.edu (taube at uiuc.edu) Date: Sun, 29 Jul 2007 19:27:30 -0500 (CDT) Subject: [CM] cm moving to subversion... Message-ID: <20070729192730.ASF34595@expms6.cites.uiuc.edu> over the next day or two i plan to switch the cm's soureforge cvs repository over to subversion (svn). in anticipation of this i have already updated the File Release pages so the latest tarballs (except for CM and Grace) are now identical to what is currently in CVS. Once i switch over to svn, only the cm and grace modules will be imported back into the repository, at least initially so the Release page will be the only place to get these modules. until the dust settles Ive put current snapshots of grace and cm on pinhead: cm: http://pinhead.music.uiuc.edu:/~hkt/cm-2.11.1.tgz grace: http://pinhead.music.uiuc.edu:/~hkt/grace-0.0.1.tgz ill also use the switchover as an opportunity to make some (modest) rearrangement of files and directories under the existing cm and grace trees. i hope this wont cause anyone too much pain (including me). the good news is that, if anything, svn is even easier to use than cvs -- for example no cvsroot or cvs login nonsense -- and the svn command set is bascally the same as cvs for simple things like file checkout. --rick From andersvi at extern.uio.no Mon Jul 30 02:04:52 2007 From: andersvi at extern.uio.no (andersvi at extern.uio.no) Date: 30 Jul 2007 11:04:52 +0200 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: <20070729192730.ASF34595@expms6.cites.uiuc.edu> Message-ID: Hello. I see there are files and snippets related to lispworks in the CM-sources, but they seem somewhat old. Have anyone been using recent versions of CM w. lispworks? The CLM sources removed lispworks-support somewhere around 2005 and the release of CLM-3. What hacks would be needed to get it going again? Guess is to plough through sndlib2clm.lisp and fill in. I dont have any lispworks-license, but am considering getting one for some other project. Any comments on how these apps (CM/CLM) performed/performs in lispworks? Concerning lispworks, are they following standards closely and stay reasonably backwards-compatible, or are they like some vendors using new releases to build obstacles for existing users? -anders From taube at uiuc.edu Mon Jul 30 03:47:36 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 30 Jul 2007 05:47:36 -0500 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: References: Message-ID: <42DA3874-59FC-49D2-B9A2-E84B92789889@uiuc.edu> anders, as far as i know CM works with Lispworks, at least it did when I last tried two years ago. Im happy to add whatever patches are necessary if its out of date. as for sndlib2clm.lisp -- why not add a cffi binding instead of one specifically for lispworks? On Jul 30, 2007, at 4:04 AM, andersvi at extern.uio.no wrote: > Hello. > > I see there are files and snippets related to lispworks in the > CM-sources, but they seem somewhat old. > > Have anyone been using recent versions of CM w. lispworks? > > The CLM sources removed lispworks-support somewhere around 2005 and > the release of CLM-3. What hacks would be needed to get it going > again? Guess is to plough through sndlib2clm.lisp and fill in. > > I dont have any lispworks-license, but am considering getting one for > some other project. Any comments on how these apps (CM/CLM) > performed/performs in lispworks? > > Concerning lispworks, are they following standards closely and stay > reasonably backwards-compatible, or are they like some vendors using > new releases to build obstacles for existing users? > > -anders > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From bil at ccrma.Stanford.EDU Mon Jul 30 04:38:57 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 30 Jul 2007 04:38:57 -0700 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: References: <20070729192730.ASF34595@expms6.cites.uiuc.edu> Message-ID: <20070730112857.M45607@ccrma.Stanford.EDU> > The CLM sources removed lispworks-support somewhere around 2005 and > the release of CLM-3. What hacks would be needed to get it going > again? I still have the Lispworks code (mostly from 2000), and would be willing to add it back in if you're interested in it. When it was removed, there was a bug that shared libraries would not load. Has anything changed in the FFI since 2000? > as for sndlib2clm.lisp -- why not add a cffi binding instead of one > specifically for lispworks? CFFI is not an option for CLM -- no support for arrays, which is crucial. Also, all the other FFI code already exists, so nothing is gained by adding a frail dependency (last change: Sept 2006 -- every time I look, for example at sourceforge, it seems that 99% of the projects are dead). From rm at seid-online.de Mon Jul 30 04:54:09 2007 From: rm at seid-online.de (Ralf Mattes) Date: Mon, 30 Jul 2007 13:54:09 +0200 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: <20070730112857.M45607@ccrma.Stanford.EDU> References: <20070729192730.ASF34595@expms6.cites.uiuc.edu> <20070730112857.M45607@ccrma.Stanford.EDU> Message-ID: <1185796449.6021.13.camel@localhost.localdomain> On Mon, 2007-07-30 at 04:38 -0700, Bill Schottstaedt wrote: > CFFI is not an option for CLM -- no support for arrays, which is crucial. What about mem-aref et al.? "The mem-aref function is similar to mem-ref but will automaticallycalculate the offset from an index." Or, from an example in the fine (really!) documentation: CFFI> (with-foreign-object (array :int 10) (dotimes (i 10) (setf (mem-aref array :int i) (random 100))) (loop for i below 10 collect (mem-aref array :int i))) => (22 7 22 52 69 1 46 93 90 65) > Also, all the other FFI code already exists, so nothing is gained by adding > a frail dependency (last change: Sept 2006 -- every time I look, > for example at sourceforge, it seems that 99% of the projects > are dead). Maybe that's because you look at the wrong places? ;-) CFFI is hosted on common-lisp.net and the last change I saw was this: /LISP/cffi$ darcs pull [snipp] Sat Jul 28 00:16:28 CEST 2007 Luis Oliveira * Make libtest work with MSVC8 Shall I pull this patch? (29/29) [ynWvpxqadjk], or ? for help: y Finished pulling and applying. Doesn't look to smelly too me. Seriously: CFFI is one of the supported backends for SWIG so i doubt it'll go stale anytime soon. HTH Ralf Mattes > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at uiuc.edu Mon Jul 30 05:37:58 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 30 Jul 2007 07:37:58 -0500 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: <20070730112857.M45607@ccrma.Stanford.EDU> References: <20070729192730.ASF34595@expms6.cites.uiuc.edu> <20070730112857.M45607@ccrma.Stanford.EDU> Message-ID: <32583D78-E96C-4A87-9FBA-758F8BC38508@uiuc.edu> yes support for passing lisp arrays doesnt appear to be there but you can allocate foreign arrays and set/get values. the cffi project itself is actively supported, stable, tested, and really good! On Jul 30, 2007, at 6:38 AM, Bill Schottstaedt wrote: >> The CLM sources removed lispworks-support somewhere around 2005 and >> the release of CLM-3. What hacks would be needed to get it going >> again? > > I still have the Lispworks code (mostly from 2000), and would be > willing to > add it back in if you're interested in it. When it was removed, > there was > a bug that shared libraries would not load. Has anything changed > in the > FFI since 2000? > >> as for sndlib2clm.lisp -- why not add a cffi binding instead of one >> specifically for lispworks? > > CFFI is not an option for CLM -- no support for arrays, which is > crucial. > Also, all the other FFI code already exists, so nothing is gained > by adding > a frail dependency (last change: Sept 2006 -- every time I look, > for example at sourceforge, it seems that 99% of the projects > are dead). > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From bil at ccrma.Stanford.EDU Mon Jul 30 06:21:28 2007 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 30 Jul 2007 06:21:28 -0700 Subject: [CM] CM/CLM w. Lispworks In-Reply-To: <32583D78-E96C-4A87-9FBA-758F8BC38508@uiuc.edu> References: <20070729192730.ASF34595@expms6.cites.uiuc.edu> <20070730112857.M45607@ccrma.Stanford.EDU> <32583D78-E96C-4A87-9FBA-758F8BC38508@uiuc.edu> Message-ID: <20070730130743.M66363@ccrma.Stanford.EDU> Ok -- if someone writes the CFFI code, I'll add it to my versions, but I'm keeping the existing FFI code. The fewer absolute dependencies, the better. From taube at uiuc.edu Mon Jul 30 07:34:28 2007 From: taube at uiuc.edu (Rick Taube) Date: Mon, 30 Jul 2007 09:34:28 -0500 Subject: [CM] svn migration working Message-ID: ive successfully migrated the sourceforge cm repository to subversion. to download CM do (all one line): svn co https://commonmusic.svn.sf.net/svnroot/commonmusic/trunk/cm cm to download Grace do: svn co https://commonmusic.svn.sf.net/svnroot/commonmusic/trunk/grace grace cvs access to the old repository will contine to work but these sources wont be updated in the future. it will take me a day or so to update the cm homepage at sourceforge to reference the svn code instead of cvs but the 'browse svn' on the commonmusic Project page is already working. -- You can get Subversion for the Mac at: http://metissian.com/projects/macosx/subversion/ and for windows (tortoisesvn) at: http://tortoisesvn.net/ Subversion documentation: http://svnbook.red-bean.com/en/1.1/index.html