From bil at ccrma.Stanford.EDU Wed Oct 15 03:52:36 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 15 Oct 2008 03:52:36 -0700 Subject: [CM] Snd 10.1 Message-ID: <20081015105128.M3230@ccrma.Stanford.EDU> S7 is now the default extension language. added rhypars.cms by Juan Reyes in clm.c, I removed the reader/writer locks (rdin and rdout gens); these were not working the way I had hoped they would. Much work on threads -- with-threaded-sound has been rewritten 3 or 4 times. gtkglext (openGL support in gtk) has been unmaintained for a long time, and by this point use of --with-gl with --with-gtk can somehow cause Snd to log you out! So, it's not highly recommended. I wish gtk had gone to OpenGL, not cairo. I'll probably remove the gtkglext support soon. I'm about to remove Gauche support. Gauche 0.8.14 segfaults immediately in Snd; Shiro says he didn't change anything, but argument handling has clearly changed. I poked around in Gauche, but didn't find anything. checked: gtk 2.14.2|3, cairo 1.8.0, autoconf 2.63, sbcl 1.0.21, Gauche 0.8.14 Thanks!: Benjamin Um, Juan Reyes From unaudio at gmail.com Wed Oct 15 08:57:52 2008 From: unaudio at gmail.com (Vytautas Jancauskas) Date: Wed, 15 Oct 2008 08:57:52 -0700 Subject: [CM] [OT] jack and common lisp Message-ID: This is off topic but since people here have experience with both lisp and linux audio i was hoping someone will be able to advice me. What is the best way to have audio playback from SBCL? I have written a set of jack bindings with cffi and they sorta work except that sbcl spazzes out all the time (to many callbacks? garbage collector?). The way i did it is to define a callback that simply generates a single tone and then register that wih jack. Does anyone has experience with this? How would you do something like that. I'm using linux btw. Thanks in advance. -------------- next part -------------- An HTML attachment was scrubbed... URL: From k.s.matheussen at notam02.no Wed Oct 15 09:04:47 2008 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Wed, 15 Oct 2008 18:04:47 +0200 (CEST) Subject: [CM] [OT] jack and common lisp In-Reply-To: References: Message-ID: On Wed, 15 Oct 2008, Vytautas Jancauskas wrote: > This is off topic but since people here have experience with both lisp and > linux audio i was hoping someone will be able to advice me. What is the best > way to have audio playback from SBCL? I have written a set of jack bindings > with cffi and they sorta work except that sbcl spazzes out all the time (to > many callbacks? garbage collector?). The way i did it is to define a > callback that simply generates a single tone and then register that wih > jack. Does anyone has experience with this? How would you do something like > that. I'm using linux btw. Thanks in advance. > Are you running really common lisp code inside the jack process callback? That's cool! Perhaps there is a way to temporarily turn off the garbage collector in sbcl? From kroger.lists at gmail.com Wed Oct 15 09:10:15 2008 From: kroger.lists at gmail.com (Pedro =?utf-8?Q?Kr=C3=B6ger?=) Date: Wed, 15 Oct 2008 13:10:15 -0300 Subject: [CM] [OT] jack and common lisp In-Reply-To: (Kjetil S. Matheussen's message of "Wed, 15 Oct 2008 18:04:47 +0200 (CEST)") References: Message-ID: <87abd5alg8.fsf@gmail.com> "Kjetil S. Matheussen" writes: > Are you running really common lisp code inside the jack process callback? > That's cool! you bet! I'd love to take a look at the code. > Perhaps there is a way to temporarily turn off the garbage collector in > sbcl? you can do that with (sb-ext:gc-off) Pedro From luke at balooga.com Wed Oct 15 11:54:13 2008 From: luke at balooga.com (Luke Crook) Date: Wed, 15 Oct 2008 11:54:13 -0700 Subject: [CM] [OT] jack and common lisp In-Reply-To: References: Message-ID: On Wed, 15 Oct 2008 08:57:52 -0700, Vytautas Jancauskas wrote: > hat is the best > way to have audio playback from SBCL? I have written a set of jack > bindings > with cffi and they sorta work except that sbcl spazzes out all the time > (to > many callbacks? garbage collector?). The way i did it is to define a > callback that simply generates a single tone and then register that wih > jack. Does anyone has experience with this? How would you do something > like > that. I'm using linux btw. Thanks in advance. I maintain a set of CFFI bindings to the SDL_mixer library. SDL_mixer also uses a callback mechanism into a Lisp process (SBCL on Linux in my case) to fill the audio buffers. Seems to work on Linux just fine. It is a little flakey on win32, but I think this is due to the SBCL/win32 port not supporting threads 100%. The bindings work fine in Lispworks and CLISP though. How often is the callback called? - Luke From taube at uiuc.edu Wed Oct 15 12:59:40 2008 From: taube at uiuc.edu (Heinrich Taube) Date: Wed, 15 Oct 2008 14:59:40 -0500 Subject: [CM] Grace 1.1.1 available Message-ID: Binaries of Grace 1.1.1 are available at: http://camil.music.uiuc.edu/Software/grace/ Grace 1.1.1 * New messages for the MidiIn Port: send "mp:inchans" ... send "mp:intypes" ... mp:inchans enables input only on the specified channel(s) mp:intypes enables input only on the specifed type(s) Use #t to enable all channels/types or #f to disable all channels/types. You can also set these graphically using the MidiIn>Configure... dialog. * MidiSeq Save and Save As... dialog now allows the time format of the midi file to be specied. Select "ticks per quarter" to save the file in a format suitable for Sibelius or Finale. * MidIn's selected input device is now saved as preference * MidiIn's Configure... information is now saved as preferece * New function 'export-spear-frames' saves specta in a text file suitable for loading into Spear. * Fixed bug in 'segs' function that prevented geometric series from working * Fixed Midi Seq channel in for score capture mode. * WINDOWS: Fixed Editor bug involving unescaped DOS directory delimiter when loading files via the Load... dialog. From k.s.matheussen at notam02.no Wed Oct 15 13:33:10 2008 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Wed, 15 Oct 2008 22:33:10 +0200 (CEST) Subject: [CM] [OT] jack and common lisp In-Reply-To: References: Message-ID: On Wed, 15 Oct 2008, Luke Crook wrote: > On Wed, 15 Oct 2008 08:57:52 -0700, Vytautas Jancauskas > wrote: > >> hat is the best >> way to have audio playback from SBCL? I have written a set of jack >> bindings >> with cffi and they sorta work except that sbcl spazzes out all the time >> (to >> many callbacks? garbage collector?). The way i did it is to define a >> callback that simply generates a single tone and then register that wih >> jack. Does anyone has experience with this? How would you do something >> like >> that. I'm using linux btw. Thanks in advance. > > > I maintain a set of CFFI bindings to the SDL_mixer library. SDL_mixer also > uses a callback mechanism into a Lisp process (SBCL on Linux in my case) > to fill the audio buffers. Seems to work on Linux just fine. It is a > little flakey on win32, but I think this is due to the SBCL/win32 port not > supporting threads 100%. The bindings work fine in Lispworks and CLISP > though. How often is the callback called? > It's a little bit harder with jack than SDL, I would think, since jack enjoys very much to throw out clients that doesn't supply data fast enough. Running jack in "soft mode" might help the situation a little bit. At least earlier "soft mode" made alsa "handle" xruns much much faster (by ignoring them). Another thing could be not to lock down memory. Some programs requiring large amounts of memory, like vmware, and sometimes also Snd, fails completely if memory is locked, but I don't know if that is the reason here. But frankly, since Sbcl's garbage collector most likely was never designed to work very well in hard realtime, the best solution is probably to send data from sbcl to jack via a lockless ringbuffer. From cycle.code.media at gmail.com Wed Oct 15 16:15:49 2008 From: cycle.code.media at gmail.com (James Baker) Date: Thu, 16 Oct 2008 10:15:49 +1100 Subject: [CM] [OT] jack and common lisp In-Reply-To: References: Message-ID: <428d2dc0810151615g406f699bmab53835f0a184945@mail.gmail.com> On Thu, Oct 16, 2008 at 2:57 AM, Vytautas Jancauskas wrote: > This is off topic but since people here have experience with both lisp and > linux audio i was hoping someone will be able to advice me. What is the best > way to have audio playback from SBCL? I have written a set of jack bindings > with cffi and they sorta work except that sbcl spazzes out all the time (to > many callbacks? garbage collector?). The way i did it is to define a > callback that simply generates a single tone and then register that wih > jack. Does anyone has experience with this? How would you do something like > that. I'm using linux btw. Thanks in advance. > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > > Have you seen this? http://www.valkama.se/article/lisp/sbcl-sy-1/ " A very rudimentary synthesizer made in CL for SBCL. At only 333 lines it is very compact and easy to understand. It utilizes JACK for audio output and ALSA for midi input. " From luke at balooga.com Wed Oct 15 19:09:12 2008 From: luke at balooga.com (Luke Crook) Date: Wed, 15 Oct 2008 19:09:12 -0700 Subject: [CM] [OT] jack and common lisp In-Reply-To: References: Message-ID: On Wed, 15 Oct 2008 13:33:10 -0700, Kjetil S. Matheussen wrote: > It's a little bit harder with jack than SDL, I would think, since > jack enjoys very much to throw out clients that doesn't > supply data fast enough. > > Running jack in "soft mode" might > help the situation a little bit. At least earlier "soft mode" > made alsa "handle" xruns much much faster (by ignoring them). > > Another thing could be not to lock down memory. > Some programs requiring large amounts > of memory, like vmware, and sometimes also Snd, > fails completely if memory is locked, but I don't > know if that is the reason here. > > But frankly, since Sbcl's garbage collector most > likely was never designed to work very well in > hard realtime, the best solution is probably > to send data from sbcl to jack via a lockless > ringbuffer. I emailed Vytautas separately and he said that the callback is being called 43 times per second. I have never used Jack, but that seems a little excessive to me. I can imagine SBCL having a fit if the callback is asynchronous; if SBCL is still in the callback when Jack fires the callback again. - Luke From bil at ccrma.Stanford.EDU Thu Oct 16 05:57:12 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 16 Oct 2008 05:57:12 -0700 Subject: [CM] gauche gone Message-ID: <20081016125440.M94449@ccrma.Stanford.EDU> Gauche support has been removed. If you were using gauche-optargs.scm or gauche-format.scm, they are now named s7-optargs.scm and s7-format.scm. 100 hours to write, 1 hour to erase. From errordeveloper at gmail.com Sat Oct 18 18:39:18 2008 From: errordeveloper at gmail.com (errordeveloper at gmail.com) Date: Sun, 19 Oct 2008 01:39:18 +0000 Subject: [CM] grid and y-axis Message-ID: <20081019013918.GA16088@00110101.errordevlopment.0> hello Bill and the list, i found show-grid option potentially usefull.. i mean potentially, cause really cannot quite see what amplitude values are on there .. i reckon y-axis needs more labling .. prolly it could be done in scheme with xm.so ..but i haven't touched that bit yet ..(( so ..may be say having .5 at least be very useful! From maw5 at dcs.qmul.ac.uk Wed Oct 22 07:46:23 2008 From: maw5 at dcs.qmul.ac.uk (Max Worgan) Date: Wed, 22 Oct 2008 15:46:23 +0100 Subject: [CM] Receiving OSC Message-ID: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> Hi, I've been trying to get CM to receive osc messages from processing and have got a little stuck. I'm running CM 2.10.0 on OS X. It seems straight forward sending messages from CM: CM> (defvar *osc-stream* (new osc-stream :remote-host "127.0.0.1" :remote-port 12000)) *OSC-STREAM* CM> (open-io *osc-stream* t) # CM> (send-msg '(test) *osc-stream*) 12 Processing is listening on 12000 and can successfully receive the messages. I've tried calling osc-stream with the :local-port args like so: CM> (defvar *listen* (new osc-stream :local-port 12001)) *LISTEN* CM> (open-io *listen* t) # And I can clearly see the dppccl process listening on port 12001 I've tried to define *osc-recv-callback* to something useful as well as using set-receiver! but it complains about osc-stream not supporting pthreads. I'm kind of stabbing in the dark here, as I'm no lisp expert. Any suggestions/hints? Thanks, Max. From errordeveloper at gmail.com Wed Oct 22 10:55:09 2008 From: errordeveloper at gmail.com (errordeveloper at gmail.com) Date: Wed, 22 Oct 2008 17:55:09 +0000 Subject: [CM] Receiving OSC In-Reply-To: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> Message-ID: <20081022175509.GA14565@00110101.errordevlopment.0> i'd really like to have OSC in snd as well! i have got an idea of using puredata as a processing host, so snd would send it a message and it would open a file (which represent a selection in snd) and process it then it'd send a msg back to snd that it's done and snd would put that back into the sound. there different ways of how you can go about doing it in pd.. now i'm thinking of shell script which will send it all to pd via pdsend and then wait for pd to return something and then when the script has exited snd will take the file back .. From taube at uiuc.edu Wed Oct 22 11:28:28 2008 From: taube at uiuc.edu (Heinrich Taube) Date: Wed, 22 Oct 2008 13:28:28 -0500 Subject: [CM] Receiving OSC In-Reply-To: <20081022175509.GA14565@00110101.errordevlopment.0> References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: I thnk the right way to do this is to link to the OscPack library. This would not be hard to do in scheme. Im trying mightily to get cm3 running in s7 scheme together with sndlib inside a juce framework, ive only taken some baby steps but its looking good so far. On Oct 22, 2008, at 12:55 PM, errordeveloper at gmail.com wrote: > i'd really like to have OSC in snd as well! > i have got an idea of using puredata as a processing host, > so snd would send it a message and it would open a file (which > represent > a selection in snd) and process it then it'd send a msg back to snd > that > it's done and snd would put that back into the sound. > there different ways of how you can go about doing it in pd.. > now i'm thinking of shell script which will send it all to pd via > pdsend > and then wait for pd to return something and then when the script has > exited snd will take the file back .. > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From bil at ccrma.Stanford.EDU Wed Oct 22 13:30:32 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 22 Oct 2008 13:30:32 -0700 Subject: [CM] snd infinite recursion Message-ID: <20081022202611.M95866@ccrma.Stanford.EDU> Meino Cramer noticed that the gtk version of Snd 10.1 can get into an infinite loop if you move the "zoom" slider on a somewhat large file. This bug was a side-effect of other changes that were trying to prepare for Gtk 3 (they are hiding the gtk structs, which is a good idea, but the replacement functions are subtly different in at least one case, causing an unanticipated callback causing an infinite loop). My fix seems to make the slider jerky, but at least it isn't catatonic. From maw5 at dcs.qmul.ac.uk Thu Oct 23 04:22:30 2008 From: maw5 at dcs.qmul.ac.uk (Max Worgan) Date: Thu, 23 Oct 2008 12:22:30 +0100 Subject: [CM] Receiving OSC In-Reply-To: References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: <3FE603BC-62A8-4BD7-8177-0D74C4EB51B8@dcs.qmul.ac.uk> So, am I to assume that CM can not receive OSC messages at this time? There don't appear to be any examples of anyone doing so, nor any documentation. I wonder if anyone could let me know where things are at, concerning the OSC library. On 22 Oct 2008, at 19:28, Heinrich Taube wrote: > I thnk the right way to do this is to link to the OscPack library. > This would not be hard to do in scheme. > Im trying mightily to get cm3 running in s7 scheme together with > sndlib inside a juce framework, ive only taken some baby steps but > its looking good so far. > > On Oct 22, 2008, at 12:55 PM, errordeveloper at gmail.com wrote: > >> i'd really like to have OSC in snd as well! >> i have got an idea of using puredata as a processing host, >> so snd would send it a message and it would open a file (which >> represent >> a selection in snd) and process it then it'd send a msg back to snd >> that >> it's done and snd would put that back into the sound. >> there different ways of how you can go about doing it in pd.. >> now i'm thinking of shell script which will send it all to pd via >> pdsend >> and then wait for pd to return something and then when the script has >> exited snd will take the file back .. >> >> _______________________________________________ >> 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 errordeveloper at gmail.com Thu Oct 23 06:28:40 2008 From: errordeveloper at gmail.com (errordeveloper at gmail.com) Date: Thu, 23 Oct 2008 13:28:40 +0000 Subject: [CM] snd infinite recursion In-Reply-To: <20081022202611.M95866@ccrma.Stanford.EDU> References: <20081022202611.M95866@ccrma.Stanford.EDU> Message-ID: <20081023132840.GB14565@00110101.errordevlopment.0> i also have a big trouble with filter drawing in the controls panel, when you put one point right under another one so the line goes sreight vertical ..it goes into infinite loop too ! On Wed, Oct 22, 2008 at 01:30:32PM -0700, Bill Schottstaedt wrote: > Meino Cramer noticed that the gtk version of Snd 10.1 can get into an > infinite loop if you move the "zoom" slider on a somewhat large > file. This bug was a side-effect of other changes that were trying to > prepare for Gtk 3 (they are hiding the gtk structs, which is a good idea, > but the replacement functions are subtly different in at least one case, > causing an unanticipated callback causing an infinite loop). My fix > seems to make the slider jerky, but at least it isn't catatonic. > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From k.s.matheussen at notam02.no Thu Oct 23 08:32:34 2008 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Thu, 23 Oct 2008 17:32:34 +0200 (CEST) Subject: [CM] Receiving OSC In-Reply-To: References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: I'm no expert in OSC, but I think using a C library, such as liblo would be easier to use in Snd than OscPack, which is a C++ library. Here's an example client: http://liblo.svn.sourceforge.net/viewvc/liblo/trunk/examples/example_client.c?revision=121&view=markup Using eval-c, the example (up to the first "send" command only and without the "blob"), can be written like this: (eval-c-add-int-type "int32_t") (eval-c "-llo" "#include " (proto->public "lo_address lo_address_new(const char *host, const char *port); int lo_send_message (lo_address targ, const char *path, lo_message msg); lo_message lo_message_new (void); void lo_message_free (lo_message m); void lo_message_add_int32 (lo_message m, int32_t a); void lo_message_add_float (lo_message m, float a); void lo_message_add_string (lo_message m, const char *a);")) (define t (lo_address_new #f "7770")) (define message (lo_message_new)) (lo_message_add_float message 0.12345678) (lo_message_add_float message 23.0) (define result (lo_send_message t "/foo/bar" message)) I don't know how easy it is to receive OSC though. That's probably a lot harder, but by using "in" to poll a ringbuffer, it isn't impossible to do without having to modify snd itself. On Wed, 22 Oct 2008, Heinrich Taube wrote: > I thnk the right way to do this is to link to the OscPack library. > This would not be hard to do in scheme. > Im trying mightily to get cm3 running in s7 scheme together with > sndlib inside a juce framework, ive only taken some baby steps but > its looking good so far. > > On Oct 22, 2008, at 12:55 PM, errordeveloper at gmail.com wrote: > >> i'd really like to have OSC in snd as well! >> i have got an idea of using puredata as a processing host, >> so snd would send it a message and it would open a file (which >> represent >> a selection in snd) and process it then it'd send a msg back to snd >> that >> it's done and snd would put that back into the sound. >> there different ways of how you can go about doing it in pd.. >> now i'm thinking of shell script which will send it all to pd via >> pdsend >> and then wait for pd to return something and then when the script has >> exited snd will take the file back .. >> >> _______________________________________________ >> 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 k.s.matheussen at notam02.no Thu Oct 23 09:33:02 2008 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Thu, 23 Oct 2008 18:33:02 +0200 (CEST) Subject: [CM] Receiving OSC In-Reply-To: References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: On Thu, 23 Oct 2008, Kjetil S. Matheussen wrote: > > I'm no expert in OSC, but I think using a C library, such as > liblo would be easier to use in Snd than OscPack, > which is a C++ library. > > Here's an example client: > http://liblo.svn.sourceforge.net/viewvc/liblo/trunk/examples/example_client.c?revision=121&view=markup > > Using eval-c, the example (up to the first "send" command only > and without the "blob"), can be written like this: > > (eval-c-add-int-type "int32_t") > (eval-c "-llo" > "#include " > (proto->public > "lo_address lo_address_new(const char *host, const char *port); > int lo_send_message (lo_address targ, const char *path, lo_message msg); > lo_message lo_message_new (void); > void lo_message_free (lo_message m); > void lo_message_add_int32 (lo_message m, int32_t a); > void lo_message_add_float (lo_message m, float a); > void lo_message_add_string (lo_message m, const char *a);")) > > (define t (lo_address_new #f "7770")) > > (define message (lo_message_new)) > > (lo_message_add_float message 0.12345678) > (lo_message_add_float message 23.0) > > (define result (lo_send_message t "/foo/bar" message)) > A convenient function: (define (osc_send host port path . args) (define address (lo_address_new host (if (string? port) port (number->string port)))) (define message (lo_message_new)) (for-each (lambda (arg) (cond ((and (number? arg) (integer? arg)) (lo_message_add_int32 message arg))) ((number? arg) (lo_message_add_float message arg)) ((string? arg) (lo_message_add_string message arg)) (else (error "unkown arg" arg)))) args) (let ((res (lo_send_message address path message))) (lo_message_free message) res)) (osc_send t #f 7770 "/foo/bar" 5.2 9.4 "hello" 9234/234) ;; ( #f means local host ) From meino.cramer at gmx.de Thu Oct 23 09:44:21 2008 From: meino.cramer at gmx.de (meino.cramer at gmx.de) Date: Thu, 23 Oct 2008 18:44:21 +0200 Subject: [CM] [meino.cramer@gmx.de: Re: snd infinite recursion] Message-ID: <20081023164421.GA6343@solfire> Hi Bill, I did a cvs-updtae this morning and recompiled snd. The bug is still there. sorry for the bad news... ;-/ Do you want the soud file ? Can I help in any other way ? Kind regards, Meino Cramer Bill Schottstaedt [08-10-23 03:58]: > Meino Cramer noticed that the gtk version of Snd 10.1 can get into an > infinite loop if you move the "zoom" slider on a somewhat large > file. This bug was a side-effect of other changes that were trying to > prepare for Gtk 3 (they are hiding the gtk structs, which is a good idea, > but the replacement functions are subtly different in at least one case, > causing an unanticipated callback causing an infinite loop). My fix > seems to make the slider jerky, but at least it isn't catatonic. > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > -- Please don't send me any Word- or Powerpoint-Attachments unless it's absolutely neccessary. - Send simply Text. See http://www.gnu.org/philosophy/no-word-attachments.html In a world without fences and walls nobody needs gates and windows. From bil at ccrma.Stanford.EDU Thu Oct 23 11:51:53 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 23 Oct 2008 11:51:53 -0700 Subject: [CM] snd infinite recursion In-Reply-To: <20081023164421.GA6343@solfire> References: <20081023164421.GA6343@solfire> Message-ID: <20081023184558.M26585@ccrma.Stanford.EDU> > The bug is still there. sorry for the bad news... ;-/ Hmm -- I wonder if sox wrote a messed up "wave" file -- you mentioned that the mp3 original was 570 MBytes, and you translated it with sox. I tried a similar sequence and got a wave file with bogus data sizes -- the truncation sent my chunk seeker into some random place in the file, it found a negative number for the next chunk's size, and we're in another loop (sigh). I added more error checking to the header reader -- the current Snd is worth another try. (wave file headers only accommodate 32 bits of bytes, so it's easy to overflow). From bil at ccrma.Stanford.EDU Thu Oct 23 13:56:46 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 23 Oct 2008 13:56:46 -0700 Subject: [CM] snd infinite recursion In-Reply-To: <20081023132840.GB14565@00110101.errordevlopment.0> References: <20081022202611.M95866@ccrma.Stanford.EDU> <20081023132840.GB14565@00110101.errordevlopment.0> Message-ID: <20081023205517.M96541@ccrma.Stanford.EDU> > i also have a big trouble with filter drawing in the controls panel, > when you put one point right under another one so the line goes sreight > vertical ..it goes into infinite loop too ! I can't get this to happen. Could you run Snd in gdb, get it to hang, type C-C, then "where" to gdb, and send me the stacktrace? Also, send whatever ./snd --version prints out, so I have some better idea where to look. From meino.cramer at gmx.de Thu Oct 23 19:43:44 2008 From: meino.cramer at gmx.de (meino.cramer at gmx.de) Date: Fri, 24 Oct 2008 04:43:44 +0200 Subject: [CM] snd infinite recursion In-Reply-To: <20081023184558.M26585@ccrma.Stanford.EDU> References: <20081023164421.GA6343@solfire> <20081023184558.M26585@ccrma.Stanford.EDU> Message-ID: <20081024024344.GA6329@solfire> Hi Bill, this time: GOOD NEWS! No scrolling til the end of time! Thank you very much for the fix!!! Kind regards and have a nice weekend Meino Cramer Bill Schottstaedt [08-10-24 03:49]: > > The bug is still there. sorry for the bad news... ;-/ > > Hmm -- I wonder if sox wrote a messed up "wave" file -- you > mentioned that the mp3 original was 570 MBytes, and you translated > it with sox. I tried a similar sequence and got a wave file with > bogus data sizes -- the truncation sent my chunk seeker into > some random place in the file, it found a negative number for > the next chunk's size, and we're in another loop (sigh). I added > more error checking to the header reader -- the current Snd > is worth another try. > > (wave file headers only accommodate 32 bits of bytes, so it's > easy to overflow). > > -- Please don't send me any Word- or Powerpoint-Attachments unless it's absolutely neccessary. - Send simply Text. See http://www.gnu.org/philosophy/no-word-attachments.html In a world without fences and walls nobody needs gates and windows. From k.s.matheussen at notam02.no Tue Oct 28 11:55:20 2008 From: k.s.matheussen at notam02.no (Kjetil S. Matheussen) Date: Tue, 28 Oct 2008 19:55:20 +0100 (CET) Subject: [CM] Receiving OSC In-Reply-To: References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: On Thu, 23 Oct 2008, Kjetil S. Matheussen wrote: > > I'm no expert in OSC, but I think using a C library, such as > liblo would be easier to use in Snd than OscPack, > which is a C++ library. > > Here's an example client: > http://liblo.svn.sourceforge.net/viewvc/liblo/trunk/examples/example_client.c?revision=121&view=markup > > Using eval-c, the example (up to the first "send" command only > and without the "blob"), can be written like this: > > (eval-c-add-int-type "int32_t") > (eval-c "-llo" > "#include " > (proto->public > "lo_address lo_address_new(const char *host, const char *port); > int lo_send_message (lo_address targ, const char *path, lo_message > msg); > lo_message lo_message_new (void); > void lo_message_free (lo_message m); > void lo_message_add_int32 (lo_message m, int32_t a); > void lo_message_add_float (lo_message m, float a); > void lo_message_add_string (lo_message m, const char *a);")) > > (define t (lo_address_new #f "7770")) > > (define message (lo_message_new)) > > (lo_message_add_float message 0.12345678) > (lo_message_add_float message 23.0) > > (define result (lo_send_message t "/foo/bar" message)) > > > I don't know how easy it is to receive OSC though. That's probably > a lot harder, but by using "in" to poll a ringbuffer, it isn't > impossible to do without having to modify snd itself. > I had a strong urge to procrastrinate a little bit from my normal work, so I wrote an osc receiver as well: http://heim.ifi.uio.no/~ksvalast/osc.scm " (load-from-path "osc.scm") (define port 6012) (osc_start_server port) (osc_add_callback (lambda (path . args) (c-display "path: " path ", args:" args))) (osc_send #f port "/foo/bar" 5.2 9.4 "hello" 5) " will print: path: /foo/bar , args: (5.19999980926514 9.39999961853027 "hello" 5) Bill, please include osc.scm in Snd if you want to. From bil at ccrma.Stanford.EDU Tue Oct 28 12:29:22 2008 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Tue, 28 Oct 2008 12:29:22 -0700 Subject: [CM] Receiving OSC In-Reply-To: References: <273F24C1-1048-44B3-9AE9-24B637EFC392@dcs.qmul.ac.uk> <20081022175509.GA14565@00110101.errordevlopment.0> Message-ID: <20081028192901.M16948@ccrma.Stanford.EDU> > Bill, please include osc.scm in Snd if you want to. ok -- thanks! From gregorybonar at gmail.com Tue Oct 28 10:50:12 2008 From: gregorybonar at gmail.com (Gregory Bonar) Date: Tue, 28 Oct 2008 10:50:12 -0700 Subject: [CM] CM in Cygwin problem Message-ID: <65fdecfa0810281050n759bcbf4p6bd2de1b7423b957@mail.gmail.com> I'm just starting to learn CM. I have it running in Linux but I would like to also run it under cygwin. When I start it in cygwin I get the following errors: /usr/local/bin/cm: line 14: $'\r': command not found /usr/local/bin/cm: line 16: $'\r': command not found /usr/local/bin/cm: line 20: $'\r': command not found /usr/local/bin/cm: line 27: $'\r': command not found /usr/local/bin/cm: line 113: $'\r': command not found /usr/local/bin/cm: line 115: $'\r': command not found /usr/local/bin/cm: line 116: $'\r': command not found /usr/local/bin/cm: line 120: $'\r': command not found /usr/local/bin/cm: line 121: syntax error near unexpected token `$'{\r'' 'usr/local/bin/cm: line 121: `imatch_head_token () { Could someone point me in the right direction please? I just downloaded cm 2.10 from Sourceforge and unzipped it . Thanks, Greg -------------- next part -------------- An HTML attachment was scrubbed... URL: From taube at uiuc.edu Tue Oct 28 16:01:05 2008 From: taube at uiuc.edu (Heinrich Taube) Date: Tue, 28 Oct 2008 18:01:05 -0500 Subject: [CM] CM in Cygwin problem In-Reply-To: <65fdecfa0810281050n759bcbf4p6bd2de1b7423b957@mail.gmail.com> References: <65fdecfa0810281050n759bcbf4p6bd2de1b7423b957@mail.gmail.com> Message-ID: <368D66CE-676E-4597-8D8E-2656B7F506DE@uiuc.edu> hmm . did you compile it? in not sure what lisp you are using. you might consider just downloading the latest Grace for windows, its drag and drop with cm3 built in. http://camil.music.uiuc.edu/Software/grace/index.html On Oct 28, 2008, at 12:50 PM, Gregory Bonar wrote: > I'm just starting to learn CM. I have it running in Linux but I > would like to also run it under cygwin. When I start it in cygwin I > get the following errors: > > /usr/local/bin/cm: line 14: $'\r': command not found > /usr/local/bin/cm: line 16: $'\r': command not found > /usr/local/bin/cm: line 20: $'\r': command not found > /usr/local/bin/cm: line 27: $'\r': command not found > /usr/local/bin/cm: line 113: $'\r': command not found > /usr/local/bin/cm: line 115: $'\r': command not found > /usr/local/bin/cm: line 116: $'\r': command not found > /usr/local/bin/cm: line 120: $'\r': command not found > /usr/local/bin/cm: line 121: syntax error near unexpected token ` > $'{\r'' > 'usr/local/bin/cm: line 121: `imatch_head_token () { > > Could someone point me in the right direction please? > I just downloaded cm 2.10 from Sourceforge and unzipped it . > > Thanks, > > Greg > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From ahcnz at ihug.co.nz Tue Oct 28 22:35:42 2008 From: ahcnz at ihug.co.nz (Adam) Date: Wed, 29 Oct 2008 18:35:42 +1300 Subject: [CM] CM in Cygwin problem In-Reply-To: <368D66CE-676E-4597-8D8E-2656B7F506DE@uiuc.edu> References: <65fdecfa0810281050n759bcbf4p6bd2de1b7423b957@mail.gmail.com> <368D66CE-676E-4597-8D8E-2656B7F506DE@uiuc.edu> Message-ID: <200810291835.43062.ahcnz@ihug.co.nz> I can't help, but just mention that I've installed the svn CM, and CLisp 2.45 and Emacs 22.3.1 with Slime on an XP machine. Actually I wanted to do the Cygwin setup thing, but left it out this time. Maybe soon. CM compiled under CLisp with no such problems. OT: the new setup for this arrangement seems not use use a save/load -initmem anymore, rather now an (in-package :cm). And it seems to work. On Wednesday 29 October 2008 12:01 pm, Heinrich Taube wrote: > hmm . did you compile it? in not sure what lisp you are using. > you might consider just downloading the latest Grace for windows, its > drag and drop with cm3 built in. > > http://camil.music.uiuc.edu/Software/grace/index.html > > On Oct 28, 2008, at 12:50 PM, Gregory Bonar wrote: > > I'm just starting to learn CM. I have it running in Linux but I > > would like to also run it under cygwin. When I start it in cygwin I > > get the following errors: > > > > /usr/local/bin/cm: line 14: $'\r': command not found > > /usr/local/bin/cm: line 16: $'\r': command not found > > /usr/local/bin/cm: line 20: $'\r': command not found > > /usr/local/bin/cm: line 27: $'\r': command not found > > /usr/local/bin/cm: line 113: $'\r': command not found > > /usr/local/bin/cm: line 115: $'\r': command not found > > /usr/local/bin/cm: line 116: $'\r': command not found > > /usr/local/bin/cm: line 120: $'\r': command not found > > /usr/local/bin/cm: line 121: syntax error near unexpected token ` > > $'{\r'' > > 'usr/local/bin/cm: line 121: `imatch_head_token () { > > > > Could someone point me in the right direction please? > > I just downloaded cm 2.10 from Sourceforge and unzipped it . > > > > Thanks, > > > > Greg > > _______________________________________________ > > Cmdist mailing list > > Cmdist at ccrma.stanford.edu > > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From taube at uiuc.edu Thu Oct 30 03:41:28 2008 From: taube at uiuc.edu (Heinrich Taube) Date: Thu, 30 Oct 2008 05:41:28 -0500 Subject: [CM] clm/cm3 in s7 scheme Message-ID: <8D558D76-916D-4ECC-BFE8-BA69A1F7327D@uiuc.edu> thanks to heavy lifting by bill -- who got sndlib/s7 building in g++ -- i have cm3 and clm running in s7 scheme. the executable is a juce console app statically linked against sndlib/s7. its fully relocatable and contains all the clm and cm scheme sources inside it, sources are autoloaded when you start the app so you dont have to do anything except load snd instruments! the repl is really crappy right now but you can load fm-violin and make toots, and the entire cm3 toolkit (patterns, loop etc) is working modulo unfound bugs. if you want to play with it ive saved a source tarball that should build on linux and osx: http://camil.music.uiuc.edu/Software/grace/downloads/cm3.tar.gz its a snap to build -- see the cm3/readme.text for how to do it. I dont have midi or grace ports or realtime yet, the next step is to get ports going, ill see if i cant add ketjils osc port while im at it. I dont know if s7 will run in realtime, but regardless getting clm going in Grace should be very easy now. dhcp204a:cm3 hkt$ bin/cm S7 Scheme (c) 2008, William Schottstaedt /\\\ ---\\\--------- ----\\\-------- ----/\\\------- Common Music 3.2.0 ---/--\\\------ --/----\\\----- / \\\/ cm> (load "/Users/hkt/Software/snd-10/v.scm") fm-violin cm> (with-sound () (fm-violin 0 1 440 .1)) "test.aiff" cm> (define pat (make-heap (hz '(c4 d ef f g af bf c5)))) pat cm> (with-sound () (loop repeat 20 for t from 0 by .1 for f = (next pat) do (fm-violin t .2 f .1))) "test.aiff" cm> (quit) Bye! dhcp204a:cm3 hkt$ -------------- next part -------------- An HTML attachment was scrubbed... URL: From taube at uiuc.edu Thu Oct 30 04:31:36 2008 From: taube at uiuc.edu (Heinrich Taube) Date: Thu, 30 Oct 2008 06:31:36 -0500 Subject: [CM] CM in Cygwin problem In-Reply-To: <3BC96F59-4F81-43AC-8450-39F83FD6852B@verizon.net> References: <65fdecfa0810281050n759bcbf4p6bd2de1b7423b957@mail.gmail.com> <368D66CE-676E-4597-8D8E-2656B7F506DE@uiuc.edu> <3BC96F59-4F81-43AC-8450-39F83FD6852B@verizon.net> Message-ID: <3AF34DEC-0623-47CB-8282-4B3C836F302F@uiuc.edu> The link to the windows binary should be working now, sorry for the delay http://camil.music.uiuc.edu/Software/grace/downloads/Grace-1.1.2-win32.zip On Oct 28, 2008, at 10:03 PM, Carl Edwards wrote: > > Hi Rick, > > It appears there's a broken link under the "Grace for Windows" link: > http://camil.music.uiuc.edu/Software/grace/downloads/Grace-1.1.2-win32.zip > > > Regards, > Carl Edwards > > > > > On Oct 28, 2008, at 7:01 PM, Heinrich Taube wrote: > >> hmm . did you compile it? in not sure what lisp you are using. >> you might consider just downloading the latest Grace for windows, its >> drag and drop with cm3 built in. >> >> http://camil.music.uiuc.edu/Software/grace/index.html >> From bingxin.shen at gmail.com Fri Oct 31 11:32:34 2008 From: bingxin.shen at gmail.com (Bingxin Shen) Date: Fri, 31 Oct 2008 14:32:34 -0400 Subject: [CM] trouble with snd using X11 on mac tiger Message-ID: <559c00e00810311132m1323f419oaa24513ddc87ec2c@mail.gmail.com> Hi all, I am new to snd. My laptop is Mac OS 10.4.11, and the X11 version is 1.1.3. I am trying to remote access the snd on the sever using ssh -X connection. BTW the server is Debian. Whenever I remote logged in with ssh, if I run snd, I could only get the graphic interference, but when I clicked on the menu it just died... with such error info: "The program 'snd' received an X Window System error. This probably reflects a bug in the program. The error was 'BadWindow (invalid Window parameter)'. (Details: serial 256 error_code 3 request_code 38 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.)" The serial number changes... I could not figure out what is wrong. Could any one help me out? I know the other guy using Mac OS 10.5.1 has no problem with the connection. I think all the other settings/hardwares for me are the same as his. Does snd only work with Leopad(10.5) but not with tiger(10.4)? Any comments? Thanks a lot! -- Best, Bing -------------- next part -------------- An HTML attachment was scrubbed... URL: