From bbattey at dmu.ac.uk Wed Sep 1 02:33:23 2004 From: bbattey at dmu.ac.uk (Bret Battey) Date: Wed, 1 Sep 2004 10:33:23 +0100 Subject: [CM] with-mix & fully specified path Message-ID: Sorry if this is an issue resolved in a later version of CLM (?) -- this issue appears in CLM 2, 27 Jan 2004, Mac OSX, OpenMCL. My first attempts at using with-mix. When I attempt to use it as follows, it works fine, writing the with-mix files granibird26-wmix01.aif and .clm to my home directory. (my-final-with-sound "/Users/bbattey/04/granifun/granibird26.aif" (sound-let ((filtered () (with-mix () "granibird26-wmix01.aif" 0 (loop for i from 1 to 8 do (butterbandpass 0 "/Users/bbattey/04/samples/india/DelhiChantFiltered.aif" :freq (* i 100); :bwq .001 :file-start 17 :dur 10) ) ; end with-mix )) (progn (move-sound (:path (make-path :polar T :path '((1 -45)))) (grani 0 20 1 filtered ETC... HOWEVER, If I designate a full path for the with-mix file I get an error. It appears that it is looking for an snd temp file in the given directory rather than the file name I specified: Changing the above to use instead (with-mix () "/Users/bbattey/04/granifun/granibird26-wmix01.aif" 0 ==> mus_header_read: can't open /Users/bbattey/04/granifun/snd19.aiff: No such file or directory> Error in process listener(2): can't reopen /Users/bbattey/04/granifun/snd19.aiff: Bret Battey http://www.BatHatMedia.com/ ------------------------------------------------------------------------ --------------------------- Senior Lecturer: Music, Technology and Innovation De Montfort University, Leicester, UK http://www.mti.dmu.ac.uk/ From bil at ccrma.Stanford.EDU Wed Sep 1 10:23:11 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 01 Sep 2004 10:23:11 -0700 Subject: [CM] env-interp In-Reply-To: References: Message-ID: <4136057F.1010509@ccrma> > There seems to be a bug in the code connected with the 'base-slot of > env's when given to #'env-interp (only tested outside run-loops in > cmucl): It was assuming the original make-env didn't set the base. I think it's ok now -- thanks for the bug report! (defun env-interp (x e &optional base) (if (eq (seg-type e) :seg) (+ (seg-offset e) (* (seg-scaler e) (envelope-interp x (seg-original-data e) (or base 1.0)))) (envelope-interp x (seg-original-data e) (or base (seg-base e))))) Also, I made the config change to all.lisp. From andersvi at extern.uio.no Thu Sep 2 05:44:53 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Thu, 02 Sep 2004 14:44:53 +0200 Subject: [CM] (set! (dac-combines-channels) #f) stops C-q In-Reply-To: (Anders Vinjar's message of "Tue, 31 Aug 2004 15:18:31 +0200") References: <04D5B73D-FB43-11D8-9FB2-000A95674CE4@uiuc.edu> Message-ID: If i (set! (dac-combines-channels) #f) - C-q refuses to play anything but chan-1 of a sound, ie. theres no reaction to it. However, (play-channel) plays the selected channel as expected. Its been like this for a while, not sure how long. (Snd version 7.6, only tested with alsa.) From rsegnini at ccrma.Stanford.EDU Thu Sep 2 11:31:31 2004 From: rsegnini at ccrma.Stanford.EDU (Rodrigo Segnini) Date: Thu, 2 Sep 2004 11:31:31 -0700 (PDT) Subject: [CM] extract track regions In-Reply-To: Message-ID: Hi gurus. I have some audio files several minutes long. Each file has instrument samples separated by silence. Both, the samples and the silences have variable duration. I want to extract each sample as an independent file automatically, so I want to keep in the new file name some reference to the original long file. I kind of see this possible in snd, so I wonder if someone has done this already. Thanks, Rodrigo From taube at uiuc.edu Thu Sep 2 06:42:18 2004 From: taube at uiuc.edu (Rick Taube) Date: Thu, 2 Sep 2004 08:42:18 -0500 Subject: [CM] OT: Notes from the Metalevel available? In-Reply-To: <042DC062-FAF3-11D8-823F-000393D984EA@rcn.com> References: <042DC062-FAF3-11D8-823F-000393D984EA@rcn.com> Message-ID: > I'm just wondering what the best way is to order the paperback for US > delivery. Im sorry for the delay in my response to your quetion -- its taken a few days to get the information about how to order Notes from the Metalevel. The process has been confusing because the publisher (Swets) was purchased by another publishing house just before the book came out! Here is the "official" reply I received, please see the link below for ordering the paperback version of Notes from the Metalevel on line. I will call Thomson and try to get them to list the paperback version on Amazon... -Rick > Following the transfer of publishing activities from T&F NL (former > Swets & Zeitlinger Publishers) to the offices of T&F in the USA and > UK, as of 1 May all order handling has been transferred to Thomson > Publishing Services. > > Orders can be placed at: > Taylor & Francis Customer Services > c/o Thomson Publishing Services > Cheriton House > North Way > ANDOVER - Hants SP10 5BE > UNITED KINGDOM > Tel: +44 1264 343 071 > Fax: +44 1264 343 005 > > E-mail: > book.orders at tandf.co.uk (individual customers) > salesorder.tandf at thomsonpublishingservices.co.uk (retail customers) > international.tandf at thomsonpublishingservices.co.uk (international > trade) > > Please refer those people who want to order the paperback to: > > http://www.thomsonpublishingservices.co.uk/catalog/info.asp? > isbn=9026519753 > > This is the link were the paperback can be ordered directly. I trust > this is of some help, in particular for sales of the paperback > version! From dlphilp at bright.net Fri Sep 3 05:45:18 2004 From: dlphilp at bright.net (Dave Phillips) Date: Fri, 03 Sep 2004 08:45:18 -0400 Subject: [CM] OT: Notes from the Metalevel available? In-Reply-To: References: <042DC062-FAF3-11D8-823F-000393D984EA@rcn.com> Message-ID: <4138675E.4030506@bright.net> Hi Rick: The link below leads to a no-show (at least it did yesterday). I'd like to purchase the paperback too, but it's not listed on Amazon (their ad says the book isn't released yet). Best, dp Rick Taube wrote: >> I'm just wondering what the best way is to order the paperback for >> US delivery. > > > Im sorry for the delay in my response to your quetion -- its taken a > few days to get the information about how to order Notes from the > Metalevel. The process has been confusing because the publisher > (Swets) was purchased by another publishing house just before the > book came out! Here is the "official" reply I received, please see > the link below for ordering the paperback version of Notes from the > Metalevel on line. I will call Thomson and try to get them to list > the paperback version on Amazon... > > -Rick > > >> Following the transfer of publishing activities from T&F NL (former >> Swets & Zeitlinger Publishers) to the offices of T&F in the USA and >> UK, as of 1 May all order handling has been transferred to Thomson >> Publishing Services. >> >> Orders can be placed at: >> Taylor & Francis Customer Services >> c/o Thomson Publishing Services >> Cheriton House >> North Way >> ANDOVER - Hants SP10 5BE >> UNITED KINGDOM >> Tel: +44 1264 343 071 >> Fax: +44 1264 343 005 >> >> E-mail: >> book.orders at tandf.co.uk (individual customers) >> salesorder.tandf at thomsonpublishingservices.co.uk (retail customers) >> international.tandf at thomsonpublishingservices.co.uk (international >> trade) >> >> Please refer those people who want to order the paperback to: >> >> http://www.thomsonpublishingservices.co.uk/catalog/info.asp? >> isbn=9026519753 >> >> This is the link were the paperback can be ordered directly. I trust >> this is of some help, in particular for sales of the paperback version! > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From taube at uiuc.edu Fri Sep 3 05:05:51 2004 From: taube at uiuc.edu (Rick Taube) Date: Fri, 3 Sep 2004 07:05:51 -0500 Subject: [CM] OT: Notes from the Metalevel available? In-Reply-To: <4138675E.4030506@bright.net> References: <042DC062-FAF3-11D8-823F-000393D984EA@rcn.com> <4138675E.4030506@bright.net> Message-ID: <994FC89A-FDA1-11D8-8FFD-000A95674CE4@uiuc.edu> Hi Dave, The link is working for me, please let me know if its still a problem. http://www.thomsonpublishingservices.co.uk/catalog/info.asp? isbn=9026519753 I will try to get them to list the paperback on amazon. On Sep 3, 2004, at 7:45 AM, Dave Phillips wrote: > Hi Rick: > > The link below leads to a no-show (at least it did yesterday). I'd > like to purchase the paperback too, but it's not listed on Amazon > (their ad says the book isn't released yet). > > Best, > > dp From taube at uiuc.edu Fri Sep 3 05:11:19 2004 From: taube at uiuc.edu (Rick Taube) Date: Fri, 3 Sep 2004 07:11:19 -0500 Subject: [CM] extract track regions In-Reply-To: References: Message-ID: <5D0BCFAA-FDA2-11D8-8FFD-000A95674CE4@uiuc.edu> I have a function that I use to do just this and it works quite well. It fins the befginnings and endings of samples by comparing a windowed average to thresholds that you specify. I've used it to cut all the piano samples from the McGill CD and much of the U Iowa sample data base. THe only downside is that it desnt use the 'run' loop, ie clm calls are made in a lisp loop so its not the speediest thing. Ill post it if you want to use it. On Sep 2, 2004, at 1:31 PM, Rodrigo Segnini wrote: > Hi gurus. > I have some audio files several minutes long. Each file has instrument > samples separated by silence. Both, the samples and the silences have > variable duration. > I want to extract each sample as an independent file automatically, so > I > want to keep in the new file name some reference to the original long > file. > I kind of see this possible in snd, so I wonder if someone has done > this > already. > Thanks, > Rodrigo > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From dlphilp at bright.net Fri Sep 3 06:58:07 2004 From: dlphilp at bright.net (Dave Phillips) Date: Fri, 03 Sep 2004 09:58:07 -0400 Subject: [CM] OT: Notes from the Metalevel available? In-Reply-To: <994FC89A-FDA1-11D8-8FFD-000A95674CE4@uiuc.edu> References: <042DC062-FAF3-11D8-823F-000393D984EA@rcn.com> <4138675E.4030506@bright.net> <994FC89A-FDA1-11D8-8FFD-000A95674CE4@uiuc.edu> Message-ID: <4138786F.1040305@bright.net> Hi Rick: Okay, I got the link, thanks. Alas, I won't order the book from England. The last book I ordered from an English seller took nearly two months to get here, and I'm loathe to pay for faster shipping (it would be nearly as much as the book itself). I guess I'll wait for Amazon to get hip. ;) Best, dp Rick Taube wrote: > Hi Dave, The link is working for me, please let me know if its still > a problem. > > http://www.thomsonpublishingservices.co.uk/catalog/info.asp? > isbn=9026519753 > > I will try to get them to list the paperback on amazon. > > On Sep 3, 2004, at 7:45 AM, Dave Phillips wrote: > >> Hi Rick: >> >> The link below leads to a no-show (at least it did yesterday). I'd >> like to purchase the paperback too, but it's not listed on Amazon >> (their ad says the book isn't released yet). >> >> Best, >> >> dp > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist > From rsegnini at ccrma.Stanford.EDU Fri Sep 3 09:49:46 2004 From: rsegnini at ccrma.Stanford.EDU (Rodrigo Segnini) Date: Fri, 3 Sep 2004 09:49:46 -0700 (PDT) Subject: [CM] extract track regions In-Reply-To: <5D0BCFAA-FDA2-11D8-8FFD-000A95674CE4@uiuc.edu> Message-ID: Thank you Rick. Please let me know how to get it. Rodrigo On Fri, 3 Sep 2004, Rick Taube wrote: > I have a function that I use to do just this and it works quite well. > It fins the befginnings and endings of samples by comparing a windowed > average to thresholds that you specify. I've used it to cut all the > piano samples from the McGill CD and much of the U Iowa sample data > base. THe only downside is that it desnt use the 'run' loop, ie clm > calls are made in a lisp loop so its not the speediest thing. Ill post > it if you want to use it. > > On Sep 2, 2004, at 1:31 PM, Rodrigo Segnini wrote: > > > Hi gurus. > > I have some audio files several minutes long. Each file has instrument > > samples separated by silence. Both, the samples and the silences have > > variable duration. > > I want to extract each sample as an independent file automatically, so > > I > > want to keep in the new file name some reference to the original long > > file. > > I kind of see this possible in snd, so I wonder if someone has done > > this > > already. > > Thanks, > > Rodrigo > > > > _______________________________________________ > > 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 Fri Sep 3 10:51:16 2004 From: taube at uiuc.edu (Rick Taube) Date: Fri, 3 Sep 2004 12:51:16 -0500 Subject: [CM] extract track regions In-Reply-To: References: Message-ID: Ok I put it on pinhead http://pinhead.music.uiuc.edu/~hkt/sampling.tar.gz there are two files 1 samples.lisp contains the CLM code to find/preview/cut the samples from an audio file. 2 flute.lisp is an example file that parses/cuts flute samples from the Universtiy of Iowa's sample database. the two main functions in sample.lisp are parse-samples and cut-samples. i looked at parse-samples and it seems that I use the lisp loop only so that I can return a Lisp list as the result of the call to the instrument :/ and there are probably better ways to do this (i wrote this code several years ago blah blah blah...) also, you may have to fiddle with wait-dac if you are not in openmcl -- i needed some way to tell sndplay to preview one sample file at a time, ie wait for one to finish playng before moving on to the next sample file. I _think_ snddplay now does this via a switch... i also have some related files: vkey.ins is a "virtual key" object that you can use to do concrete with a sample database that you cut that does basic mixing, filtering and timestreatching. it contains some replacemnt CLM instrumnts that Michael Klilngbeil wrote so that the instruments could work with more than stero files. Im pretty sure that he wont mind if this code is posted, if hes ok i can also make this stuff available if anyone is interested. From bil at ccrma.Stanford.EDU Sun Sep 5 05:01:31 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sun, 05 Sep 2004 05:01:31 -0700 Subject: [CM] with-mix & fully specified path In-Reply-To: References: Message-ID: <413B001B.7040905@ccrma> > If I designate a full path for the with-mix file I get an error. I can't get this to happen in OpenMCL/OSX/CLM-3 or ACL. From andersvi at extern.uio.no Mon Sep 6 08:06:43 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Mon, 06 Sep 2004 17:06:43 +0200 Subject: [CM] extract track regions In-Reply-To: (Rick Taube's message of "Fri, 3 Sep 2004 12:51:16 -0500") References: Message-ID: Heres a version of Ricks parse-samples to do similar things with Snd. It works in the same manner: placing marks at the onsets, ready to edit if needed, leaving to mark-explode or some such to actually cut out the samples to sound-files. ; might be useful to do some normalizing or something first (compand-channel) ; from examp.scm or somewhere (map (lambda (x) (add-mark (seconds->samples (car x)))) (parse-attacks #:len 1000 #:on 0.1 :off 0.1)) ;; edit placement if needed, and do the splicing out (mark-explode) ; ive used a modified version of mark-explode which splices from ; first mark onwards, and takes last cut from last mark to ; end-of-file, as well as small enveloping on start and end. This leaves file-0.snd, file-1.snd... etc -------------- next part -------------- A non-text attachment was scrubbed... Name: parse-attacks.scm Type: application/octet-stream Size: 1702 bytes Desc: parse-attacks for Snd URL: -------------- next part -------------- I need the functions 'seconds->samples and 'samples->seconds included here because the default ones in snd seem to be hardwired at srate=22050 or something. Heres the version of mark-explode ive used here: -------------- next part -------------- A non-text attachment was scrubbed... Name: mark-explode.scm Type: application/octet-stream Size: 1681 bytes Desc: tilted mark-explode URL: From andersvi at extern.uio.no Mon Sep 6 07:07:51 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Mon, 06 Sep 2004 16:07:51 +0200 Subject: [CM] [snd] save-sound-as Message-ID: Im getting some weird behaviour using 'save-sound-as in a script. Basically its the same as the example 'extract-channel in snd.html: (define (extract-channel filename snd chn) (save-sound-as filename snd :channel chn)) (extract-channel "tmp1.snd" 0 0) This works fine if i call it from snd's listener. However, running snd as an emacs subjob makes snd pop up a blank window, and snd hangs foreverafter (extract-channel "tmp4.snd" 0 0) ... From bil at ccrma.Stanford.EDU Wed Sep 8 07:13:25 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Wed, 08 Sep 2004 07:13:25 -0700 Subject: [CM] re: samples->seconds Message-ID: <413F1385.1000609@ccrma> > I need the functions 'seconds->samples and 'samples->seconds > included here because the default ones in snd seem to be > hardwired at srate=22050 or something. They're considered a part of the clm module, so they use mus-srate, rather than srate -- there are a bazillion srates floating around. From bbattey at dmu.ac.uk Wed Sep 8 09:00:18 2004 From: bbattey at dmu.ac.uk (Bret Battey) Date: Wed, 8 Sep 2004 17:00:18 +0100 Subject: [CM] re: samples->seconds In-Reply-To: <413F1385.1000609@ccrma> References: <413F1385.1000609@ccrma> Message-ID: <2E4FD1B2-01B0-11D9-A6E9-000D9369AD4E@dmu.ac.uk> Actually, this brings up an ongooing question for me (though perhaps the question is resolved/clarified in CLM 3) I've taken to setting both *clm-srate* and *srate* before starting any work -- and below we have reference to mus-srate. It has never been clear to me which one, if any, was "official." Often setting only one of them I would get unpredictable results, depending on what instrument I was using, etc. Rather than tracking down the problem, I just started using the shotgun approach. Particularly when one is designing one's own instruments: which one "should" one reference to make sample-rate based calculations? Also, seems like these days default should now be 44.1k? Bret Battey http://www.BatHatMedia.com/ ------------------------------------------------------------------------ --------------------------- Senior Lecturer: Music, Technology and Innovation De Montfort University, Leicester, UK http://www.mti.dmu.ac.uk/ On 8 Sep 2004, at 15:13, Bill Schottstaedt wrote: > > I need the functions 'seconds->samples and 'samples->seconds > > included here because the default ones in snd seem to be > > hardwired at srate=22050 or something. > > They're considered a part of the clm module, so they use > mus-srate, rather than srate -- there are a bazillion > srates floating around. > > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From andersvi at extern.uio.no Thu Sep 9 04:36:58 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Thu, 09 Sep 2004 13:36:58 +0200 Subject: [CM] re: samples->seconds In-Reply-To: <2E4FD1B2-01B0-11D9-A6E9-000D9369AD4E@dmu.ac.uk> (Bret Battey's message of "Wed, 8 Sep 2004 17:00:18 +0100") References: <413F1385.1000609@ccrma> <2E4FD1B2-01B0-11D9-A6E9-000D9369AD4E@dmu.ac.uk> Message-ID: You can set the default by something like (setf *clm-srate* 44100) somewhere (you find these in "defaults.lisp") However, its easy to get confused, as (mus-srate) ignores the initial value of *clm-srate*, and seems to stick with its last value (from call on with-sound), * *clm-srate* 44100 * (mus-srate) 22050.0 * (with-sound (:srate 48000 :play nil) (format t "(mus-srate) = ~A, *srate* = ~A, *clm-srate* = ~A" (clm::mus-srate) clm::*srate* clm::*clm-srate*)) (mus-srate) = 48000.0, *srate* = 48000, *clm-srate* = 44100 "test.snd" * clm::*srate* 44100 * (clm::mus-srate) 48000.0 * Maybe a switch to configure: --with-srate=48000 or some such? From rsegnini at ccrma.Stanford.EDU Fri Sep 10 00:42:48 2004 From: rsegnini at ccrma.Stanford.EDU (Rodrigo Segnini) Date: Fri, 10 Sep 2004 00:42:48 -0700 Subject: [CM] cm/clm/cmn & snd on windows Message-ID: <00a801c49709$c6397290$6400a8c0@ccswin> Hi all. What's the status on running a full suite of cm/clm/cmn & snd on windows xp, either natively or through emulators (i.e. cygwin), and what lisp is used? I have spent some time checking various threads on the list archives, and will continue doing so, but thought (and hoped) that maybe someone already has the whole enchilada. So far I have found the following thread-ends: * CM + CLISP + Win Native -> Ok $ clisp > (load "c:/path/to/cm-2.4.2/src/cm.lisp") when done building it will save a cm.bat file $ chdir c:\path\to\cm-2.4.2 $ bin\cm.bat * CM + CLISP + Cygwin -> error message *** - The value of TOK must be one of DARWIN, LINUX, CYWIN_NT-5.0, FREEBSD Thanks, Rodrigo From bil at ccrma.Stanford.EDU Fri Sep 10 03:34:54 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 10 Sep 2004 03:34:54 -0700 Subject: [CM] re: samples->seconds In-Reply-To: <2E4FD1B2-01B0-11D9-A6E9-000D9369AD4E@dmu.ac.uk> References: <413F1385.1000609@ccrma> <2E4FD1B2-01B0-11D9-A6E9-000D9369AD4E@dmu.ac.uk> Message-ID: <4141834E.1050004@ccrma> > I've taken to setting both *clm-srate* and *srate* before starting any work -- > and below we have reference to mus-srate. It has never been clear to me which one, > if any, was "official." Often setting only one of them I would get unpredictable > results, depending on what instrument I was using, etc. Rather than tracking down > the problem, I just started using the shotgun approach. > > Particularly when one is designing one's own instruments: which one "should" one > reference to make sample-rate based calculations? *srate*. Here's a paragraph I added to clm.html:

In both Scheme and CL, the *clm-* variables (like *clm-srate*) set the default values. In CL the corresponding un-clm'd versions (*srate*) hold the current values. So, if with-sound doesn't include the :srate argument, *srate* is the same as *clm-srate*; otherwise it reflects the :srate value for the duration of the with-sound call. The local variables (in CL) that are currently exported are: *srate*, *safety*, and *debug*. Unexported, but available in the clm package are *channels*, *data-format*, *header-type*, *notehook*, *clipped*, *verbose*, and *statistics*.

mus-srate is used in Snd (and clm.c). In that context, *clm-srate* holds the default with-sound setting, and mus-srate holds the current value (rather than try to deal with dynamic variables in Scheme). (with-sound's can be nested, so *srate* is (intended to be) a dynamic variable in CL, but I'm not sure that was ever a good idea). In Snd/C/CLM, we have mus-sound-srate (a sound file's srate given the filename), mus-srate (the current with-sound or CLM output srate), srate (the currently selected sound's srate), and *clm-srate* (in ws.scm -- the default with-sound :srate argument value). You could yoke mus-srate to srate by adding a couple hook functions: (add-hook! select-sound-hook (lambda (snd) (set! (mus-srate) (srate snd)))) (add-hook! after-open-hook (lambda (snd) (set! (mus-srate) (srate snd)))) You can set the default srate to be 44100 in CL by (setf *clm-srate* 44100). In Snd it would be (set! *clm-srate* 44100) after loading ws.scm. Also, in CL there was a sound-srate that was the same as mus-sound-srate; it's still in CLM-3, but I wish it weren't. And mus-audio-srate which refers to the DAC srate. From bil at ccrma.Stanford.EDU Fri Sep 10 03:40:15 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 10 Sep 2004 03:40:15 -0700 Subject: [CM] buffer gen? Message-ID: <4141848F.7090100@ccrma> Would it greatly inconvenience anyone if I remove the CLM buffer generator? It didn't turn out to be as useful as I originally thought it would be -- too complicated. From bil at ccrma.Stanford.EDU Fri Sep 10 04:04:28 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 10 Sep 2004 04:04:28 -0700 Subject: [CM] cm/clm/cmn & snd on windows In-Reply-To: <00a801c49709$c6397290$6400a8c0@ccswin> References: <00a801c49709$c6397290$6400a8c0@ccswin> Message-ID: <41418A3C.2060809@ccrma> I don't know about clm -- I think I tried clm-3 in acl 5.0 running in windoze 95. I'm told that snd runs in cygwin. From taube at uiuc.edu Fri Sep 10 04:44:26 2004 From: taube at uiuc.edu (Rick Taube) Date: Fri, 10 Sep 2004 06:44:26 -0500 Subject: [CM] cm/clm/cmn & snd on windows In-Reply-To: <00a801c49709$c6397290$6400a8c0@ccswin> References: <00a801c49709$c6397290$6400a8c0@ccswin> Message-ID: If you want to run under cygwin I think you should try using the CM from CVS because I think this problem is solved there. I will make another 2.4 bug release tarball but Im currently too busy with 2.5.0 to get to it for several weeks, sorry! You might be able to hack around the problem by adding the FREEBSD token in cm.lisp... On Sep 10, 2004, at 2:42 AM, Rodrigo Segnini wrote: > Hi all. > > What's the status on running a full suite of cm/clm/cmn & snd on > windows > xp, either natively or through emulators (i.e. cygwin), and what lisp > is > used? > I have spent some time checking various threads on the list archives, > and will continue doing so, but thought (and hoped) that maybe someone > already has the whole enchilada. > > So far I have found the following thread-ends: > > * CM + CLISP + Win Native -> Ok > > $ clisp >> (load "c:/path/to/cm-2.4.2/src/cm.lisp") > > when done building it will save a cm.bat file > > $ chdir c:\path\to\cm-2.4.2 > $ bin\cm.bat > > * CM + CLISP + Cygwin -> error message > > *** - The value of TOK must be one of DARWIN, LINUX, CYWIN_NT-5.0, > FREEBSD > > Thanks, > > Rodrigo > > _______________________________________________ > Cmdist mailing list > Cmdist at ccrma.stanford.edu > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist From philthom at freeshell.org Fri Sep 10 12:51:42 2004 From: philthom at freeshell.org (Phil Thomson) Date: Fri, 10 Sep 2004 19:51:42 +0000 (UTC) Subject: [CM] Re: Cmdist digest, Vol 1 #599 - 5 msgs In-Reply-To: <20040910190003.11685.99119.Mailman@cm-mail.stanford.edu> References: <20040910190003.11685.99119.Mailman@cm-mail.stanford.edu> Message-ID: Sorry if this is a repost, but I haven't yet seen this posting. I am trying to compile snd on Darwin/Mac OS X, but I'm having trouble. Is this an appropriate place to ask this question? Or can someone point me to an FAQ that might help? P "Computers in the future may weigh no more than 1.5 tons." Popular Mechanics, 1949 ========= Phil Thomson home: http://www.sfu.ca/~pthomson label: http://centibel.org/ blog: http://pthomson.blogspot.com/ group: http://groups.yahoo.com/group/databenders/ ========= SDF Public Access UNIX System http://www.freeshell.org/ Geekier than you since 1987. From Meino.Cramer at gmx.de Fri Sep 10 19:22:56 2004 From: Meino.Cramer at gmx.de (Meino Christian Cramer) Date: Sat, 11 Sep 2004 04:22:56 +0200 (CEST) Subject: [CM] CVS access failed... Message-ID: <20040911.042256.74750543.Meino.Cramer@gmx.de> Hi, I tried to access the cvs of snd and got this back: Logging in to :pserver:anonymous at cvs.sourceforge.net:2401/cvsroot/snd CVS password: cvs [login aborted]: unrecognized auth response from cvs.sourceforge.net: cvs pserver: cannot read /cvsroot/snd/CVSROOT/config: Input/output error cvs [update aborted]: unrecognized auth response from cvs.sourceforge.net: cvs pserver: cannot read /cvsroot/snd/CVSROOT/config: Input/output error Has somethinh changed I didn't notice ? What did I wrong ? Kind regards, Meino From bil at ccrma.Stanford.EDU Sat Sep 11 04:51:14 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 11 Sep 2004 04:51:14 -0700 Subject: [CM] CVS access failed... In-Reply-To: <20040911.042256.74750543.Meino.Cramer@gmx.de> References: <20040911.042256.74750543.Meino.Cramer@gmx.de> Message-ID: <4142E6B2.2080308@ccrma> > Has somethinh changed I didn't notice ? What did I wrong ? Has it been about 3 months since you accessed this account? If so, I think I'd just delete it and start over from scratch -- the project name changed slightly at sourceforge (you can also fix it by changing the CVS/Root file, I think). Otherwise, it's probably just some maintenance downtime at sourceforge. The currently correct form is in README.Snd. From bil at ccrma.Stanford.EDU Sat Sep 11 04:52:34 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Sat, 11 Sep 2004 04:52:34 -0700 Subject: [CM] Re: Cmdist digest, Vol 1 #599 - 5 msgs In-Reply-To: References: <20040910190003.11685.99119.Mailman@cm-mail.stanford.edu> Message-ID: <4142E702.4050608@ccrma> > I am trying to compile snd on Darwin/Mac OS X, but I'm having trouble. > Is this an appropriate place to ask this question? > Or can someone point me to an FAQ that might help? Did you check the relevant portion of README.Snd? It has a long section on OSX. From Meino.Cramer at gmx.de Sat Sep 11 21:27:28 2004 From: Meino.Cramer at gmx.de (Meino Christian Cramer) Date: Sun, 12 Sep 2004 06:27:28 +0200 (CEST) Subject: [CM] CVS access failed... In-Reply-To: <4142E6B2.2080308@ccrma> References: <20040911.042256.74750543.Meino.Cramer@gmx.de> <4142E6B2.2080308@ccrma> Message-ID: <20040912.062728.75186150.Meino.Cramer@gmx.de> From: Bill Schottstaedt Subject: Re: [CM] CVS access failed... Date: Sat, 11 Sep 2004 04:51:14 -0700 Hi Bill, Thanks for your reply ! :) ...I am updateing my snd cvs snapshot nearly every day. The "digital hickup" was a temporary limited effekt, it seems. Everything is working fine again, now :O) Thank you for makeing software like HiFi -- not only AM !:) Have a nice weekend ! Meino > > Has somethinh changed I didn't notice ? What did I wrong ? > > Has it been about 3 months since you accessed this account? If so, > I think I'd just delete it and start over from scratch -- the project > name changed slightly at sourceforge (you can also fix it by changing > the CVS/Root file, I think). Otherwise, it's probably just some > maintenance downtime at sourceforge. > > The currently correct form is in README.Snd. > > From andersvi at extern.uio.no Mon Sep 13 02:14:26 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Mon, 13 Sep 2004 11:14:26 +0200 Subject: [CM] wildcards in rewrite-pattern In-Reply-To: (Johannes Quint's message of "Tue, 30 Sep 2003 14:38:18 +0200") References: Message-ID: Has the rewrite pattern changed in some way? Or am i misunderstanding how it works. I cant get the wildcards to work in the rewrite: * (new rewrite :of '(a b c d) :rules '(((a) b c -> b b) (a (b) c -> a b) (b (b) a -> c) (* (c) * -> a))) => Error in function (LABELS GETNODE PARSE-RULES): No node for id * in rule (* (C) * -> A). [Condition of type SIMPLE-ERROR] Restarts: 0: [ABORT] Return to Top-Level. Debug (type H for help) ((LABELS GETNODE PARSE-RULES) * # (* (C) * -> A)) Source: ; File has been modified since compilation: ; /div/notam02/site/cm-sys/cm-2.5.0/src/patterns.lisp ; Using form offset instead of character position. (ERROR "No node for id ~s in rule ~s." ID RULE) 0] From kouhia at nic.funet.fi Thu Sep 16 03:09:01 2004 From: kouhia at nic.funet.fi (Juhana Sadeharju) Date: Thu, 16 Sep 2004 13:09:01 +0300 Subject: [CM] Re: buffer gen? Message-ID: >From: Bill Schottstaedt > >Would it greatly inconvenience anyone if I remove the >CLM buffer generator? It didn't turn out to be as useful >as I originally thought it would be -- too complicated. What it does? What you wanted it to do? What ever you do, please keep the code in CLM package so that it stays in history books. Perhaps to some attic directory. Juhana -- http://music.columbia.edu/mailman/listinfo/linux-graphics-dev for developers of open source graphics software From rbastian at club-internet.fr Thu Sep 16 00:23:47 2004 From: rbastian at club-internet.fr (rbastian) Date: Thu, 16 Sep 2004 09:23:47 +0200 Subject: [CM] cmn, grace notes Message-ID: <04091609234701.00763@rbastian> Hello, i need the following figure and i dont know how to write it : trill------------------------ grace-note half half half grace-note grace note bar eighth ... the prolongatione of th trill should be like a vibrato thanks for your help -- Ren? Bastian From rbastian at club-internet.fr Thu Sep 16 01:54:53 2004 From: rbastian at club-internet.fr (=?iso-8859-15?q?Ren=E9=20Bastian?=) Date: Thu, 16 Sep 2004 10:54:53 +0200 Subject: [CM] Fwd: cmn, grace notes Message-ID: <04091610345902.00763@rbastian> Hm, not "vibrato" wavy-line & wavy-time ! ---------- Message transmis ---------- Subject: cmn, grace notes Date: Thu, 16 Sep 2004 09:23:47 +0200 From: rbastian To: cmdist at ccrma.Stanford.edu Hello, i need the following figure and i dont know how to write it : trill------------------------ grace-note half half half grace-note grace note bar eighth ... the prolongatione of th trill should be like a vibrato thanks for your help -- Ren? Bastian ------------------------------------------------------- -- Ren? Bastian http://www.musiques-rb.org : Musique en Python From gnus at chello.at Fri Sep 17 13:52:30 2004 From: gnus at chello.at (Josef Oswald) Date: Fri, 17 Sep 2004 22:52:30 +0200 Subject: [CM] Question to SND Message-ID: Hi, This question has buzzeld me quite a bit, on my PC I have _two_ Soundcards one is a Soundplaster Live 5.1 the other a VIA chip ( on the motherboard) How can I tell SND which Card it should use? Josef Oswald -- registered-linux-user # 134.818 at http://counter.li.org The box said Windows, NT or better, so I installed Linux :-) From andersvi at extern.uio.no Mon Sep 20 07:48:58 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Mon, 20 Sep 2004 16:48:58 +0200 Subject: [CM] bug with #'signum in run Message-ID: (I have only checked this with clm-3 & cmucl) (signum samp) returns 0.0 for values between -1.0 and 1.0 inside run-loop (definstrument test (amp) (run (loop for i from 0 to 10 do (clm-print "~A" (signum (* amp (- (random 1.0) 1.0))))))) (with-sound () (test 1.0)) * 0,0,0,0,0,0,0,0,0,0,0, "test.snd" (with-sound () (test 2.0)) * 0,-1,0,-1,-1,-1,0,0,-1,0,-1, "test.snd" * From andersvi at extern.uio.no Wed Sep 22 04:39:04 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Wed, 22 Sep 2004 13:39:04 +0200 Subject: [CM] snd record panel & Jack freezes box In-Reply-To: (Anders Vinjar's message of "Mon, 20 Sep 2004 16:48:58 +0200") References: Message-ID: Trying to use snd as a recorder in a "jack-pipe" Having jack running, and firing up a jack-configured snd makes hitting the "Record" from menu freeze the machines here. The error message is something about "overflow.." and "vertical slide axis.." Snd version 7.6, 19-Aug. I'll write the message down and send along if it isn't reproducible at your place. If jack is not running i cant open any record-panel, just getting an error about : "[13:18:17] no audio input devices available" From andersvi at extern.uio.no Wed Sep 22 08:14:12 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Wed, 22 Sep 2004 17:14:12 +0200 Subject: [CM] replacement for def-clm-struct? Message-ID: What would be a wise way to define data-structures accessible in run now that def-clm-struct is gone? Im looking at dlocsig.lisp for clm-3. It functions allright (putting in 4th argument *output* to some calls on #'out-any), however the need to carry all the previous struct-fields as variables or some such is quite messy and errorprone. Ive tried replacing make-dlocs with a macro which places the various fields in an array, making them accessible by straight array-referencing in run. For some reason i cant get it to work, getting loads of errors when compiling. (defmacro make-dlocs (&key start end out-channels rev-channels out-delays out-map gains rev-gains delay path rev) ;; 11 fields to keep with the generator `(let ((dl-arr (make-array 11))) (setf (aref dl-arr 0) ,start) (setf (aref dl-arr 1) ,end) (setf (aref dl-arr 2) ,out-channels) .... and then defining "accessors" along the lines of: (defmacro dlocs-start (dloc) `(aref ,dloc 0)) (defmacro dlocs-end (dloc) `(aref ,dloc 1)) (defmacro dlocs-out-channels (dloc) `(aref ,dloc 2)) (defmacro dlocs-rev-channels (dloc) `(aref ,dloc 3)) Suggestions? From bil at ccrma.Stanford.EDU Thu Sep 23 03:54:54 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Thu, 23 Sep 2004 03:54:54 -0700 Subject: [CM] Fwd: cmn, grace notes In-Reply-To: <04091610345902.00763@rbastian> References: <04091610345902.00763@rbastian> Message-ID: <4152AB7E.3070200@ccrma> I think you can use the "wavy-line" function: (cmn staff treble c4 q (c4 q (trill (wavy-line (list 0 0 3.0 0)))) d4 h e4 h f4 h g4 h c5 h) From bil at ccrma.Stanford.EDU Fri Sep 24 03:29:21 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 24 Sep 2004 03:29:21 -0700 Subject: [CM] bug with #'signum in run In-Reply-To: References: Message-ID: <4153F701.6090408@ccrma> > (signum samp) returns 0.0 for values between -1.0 and 1.0 inside > run-loop > > (clm-print "~A" (signum (* amp (- (random 1.0) 1.0))))))) I forgot to support float args in signum in the CLM run macro. (Also, your code will always return a negative number -- perhaps you want (- (random 1.0) 0.5)?) From bil at ccrma.Stanford.EDU Fri Sep 24 03:31:41 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 24 Sep 2004 03:31:41 -0700 Subject: [CM] Question to SND In-Reply-To: References: Message-ID: <4153F78D.3090001@ccrma> > This question has buzzeld me quite a bit, on my PC I have _two_ > Soundcards one is a Soundplaster Live 5.1 the other a VIA chip ( on the > motherboard) > > How can I tell SND which Card it should use? It's been years since I tried it, but it used to be the case that you could choose the card (in Linux/OSS or the SGI) with the "system" portion of the "device" argument to the various mus-audio-* functions. I think I was using SGI terminology -- "system" means "sound card" on a PC. So the full device argument is (+ device (ash card-number 16)). See rtio.scm which uses the function card+device. There's also the variable audio-output-device to set the default. From bil at ccrma.Stanford.EDU Fri Sep 24 03:32:53 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Fri, 24 Sep 2004 03:32:53 -0700 Subject: [CM] replacement for def-clm-struct? In-Reply-To: References: Message-ID: <4153F7D5.4080804@ccrma> > Ive tried replacing make-dlocs with a macro which places the > various fields in an array, making them accessible by straight > array-referencing in run. For some reason i cant get it to work, > getting loads of errors when compiling. def-clm-struct still exists (see prc-toolkit95.lisp for example), but its fields have to be real or integer, not generators. The run macro assumes array elements are all of one type, so you can't mix things there either. I actually thought the dlocsig code was easier to read without all the repetitious (and un-informative) struct field names. From andersvi at extern.uio.no Fri Sep 24 04:49:40 2004 From: andersvi at extern.uio.no (Anders Vinjar) Date: Fri, 24 Sep 2004 13:49:40 +0200 Subject: [CM] bug with #'signum in run In-Reply-To: <4153F701.6090408@ccrma> (Bill Schottstaedt's message of "Fri, 24 Sep 2004 03:29:21 -0700") References: <4153F701.6090408@ccrma> Message-ID: The example was made up a bit to quick im afraid :-) - just for illustration. >>> "BS" == Bill Schottstaedt writes: (signum samp) returns 0.0 for values between -1.0 and 1.0 inside run-loop (clm-print "~A" (signum (* amp (- (random 1.0) 1.0))))))) I forgot to support float args in signum in the CLM run macro. (Also, your code will always return a negative number -- perhaps you want (- (random 1.0) 0.5)?) From taube at uiuc.edu Sun Sep 26 08:29:15 2004 From: taube at uiuc.edu (Rick Taube) Date: Sun, 26 Sep 2004 10:29:15 -0500 Subject: [CM] Plotter in GTK2 available Message-ID: There was an old program named Plotter With source code so hacked it did totter. But in GTK2 Its now spanking brand new, And Zoom even works like it oughter. --- Screenshots, documentation, examples, install instructions, and bug list are available at: http://pinhead.music.uiuc.edu/~hkt/plotter/doc/dict/plotter-topic.html Tarball download at: http://pinhead.music.uiuc.edu/~hkt/plotter.tar.gz Perversly enough (and despite the fact that I'm using X/GTK2) this first version is for OSX/OpenMCL only because it turned out that OpenMCL is the only CLtL that makes it easly to interface to such a huge library. My goal is to have Plotter (and whatever other GUI stuff is developed for CM) working in three environments: Darwin/OpenMCL, Linux/SBCL Linux/Guile. The next step is to try to get it working in SBCL (partially done), then Guile, and finally to integrate the code into cm's 2.5.0 cvs tree. If anyone wants to help out with this gui coding please let me know. From bbattey at dmu.ac.uk Sun Sep 26 09:31:05 2004 From: bbattey at dmu.ac.uk (Bret Battey) Date: Sun, 26 Sep 2004 17:31:05 +0100 Subject: [CM] PSOLA in CLM? In-Reply-To: Message-ID: Not highly hopeful, but sure would save me some time: has anyone out there already implemented PSOLA (Pitch-Synchronous Overlap Add) pitch/time shifting for CLM? -=Bret Bret Battey http://www.BatHatMedia.com/ ---------------------------------------------------------------------------- Senior Lecturer: Music, Technology and Innovation De Montfort University, Leicester, UK http://www.mti.dmu.ac.uk/ From carl.boingie at verizon.net Sun Sep 26 12:27:18 2004 From: carl.boingie at verizon.net (Carl Edwards) Date: Sun, 26 Sep 2004 15:27:18 -0400 Subject: [CM] Plotter in GTK2 available In-Reply-To: Message-ID: > There was an old program named Plotter > With source code so hacked it did totter. > But in GTK2 > Its now spanking brand new, > And Zoom even works like it oughter. LOL... That's pretty darn exciting. I think I am going to have to get a modern Mac! Carl Edwards From bil at ccrma.Stanford.EDU Mon Sep 27 07:08:33 2004 From: bil at ccrma.Stanford.EDU (Bill Schottstaedt) Date: Mon, 27 Sep 2004 07:08:33 -0700 Subject: [CM] PSOLA in CLM? In-Reply-To: References: Message-ID: <41581EE1.4050004@ccrma> Perhaps rubber.scm (in Snd)?