- chuck-users - lists.cs.princeton.edu

noob: machine.add but can't find function
by George Locke 26 Aug '12

26 Aug '12

Hi, I've got a chuck file called rand.ck that defines functions, and another that machine.add's rand.ck, but can't find the functions rand.ck " fun float rand_norm() { // Marsaglia polar method, wastes the second ind var float s, x, y; 0 => s; while (s == 0 || s >= 1) { Std.randf() => x; Std.randf() => y; x*x + y*y => s; } return x * Math.sqrt(-2 * Math.log(s) / s); } " test.ck "Machine.add("/Users/George/Documents/chucko/util/rand.ck"); for (0 => int i; i<10; i++) { rand_norm() => float x; <<< x >>>; } " I add test.ck to the miniAudicle VM and the following error error: line(3): undefined variable 'rand_norm'... Am noob. plz help. kthx. - George

2 2

Re: [chuck-users] chuck-users Digest, Vol 85, Issue 26
by Joel Matthys 26 Aug '12

26 Aug '12

You need parens after closeFile: w.closeFile(); -Joel On 08/26/2012 12:00 PM, chuck-users-request(a)lists.cs.princeton.edu wrote: > Send chuck-users mailing list submissions to > chuck-users(a)lists.cs.princeton.edu > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users > or, via email, send a message with subject or body 'help' to > chuck-users-request(a)lists.cs.princeton.edu > > You can reach the person managing the list at > chuck-users-owner(a)lists.cs.princeton.edu > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of chuck-users digest..." > > > Today's Topics: > > 1. WvOut: writes soundfile, but always silent (Click Nilson) > > > _______________________________________________ > chuck-users mailing list > chuck-users(a)lists.cs.princeton.edu > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users

2 1

WvOut: writes soundfile, but always silent
by Click Nilson 26 Aug '12

26 Aug '12

Hi all, Just trying chimera to see what is up in ChucK world. I've found that I can't get WvOut to work; I've tried innumerable permutations, from: chuck curly.ck rec.ck to the code examples below. I can get an output sound file, but it's always silent (though the right length). I've tried different orders of WvOut and dac, and many variations (including wav and aif output). Is this a known problem, or is there some other official way to do this that supersedes the manual included with the distribution? I couldn't find any resolution hunting on old posts from this mailing list. (my system: 10.6.8 OS X) I also find attempting --silent or -s that I just get the system locking up, and a crash on ctrl+c: chuck_oo.cpp:125: failed assertion `m_ref_count > 0' Assertion failed: (libcompiler_rt abort), function __eprintf, file /SourceCache/clang/clang-421.0.60/src/projects/compiler-rt/lib/eprintf.c, line 31. Abort trap best, Nick simple test, from an old chuck-users posting: Noise whitenoise => dac; whitenoise => WvOut waveOut => blackhole; 0.5 => whitenoise.gain; "testchuckoutput"=>waveOut.aifFilename; 2::second => now; waveOut.closeFile; ------------------------------------------------------------------ more complicated (I want to do sample by sample synthesis and write out a sound file): //sample by sample writing of audio in ChucK //for maximal speed in NRT mode, run chuck with -s flag or --silent //e.g. ./chuck /data/sussex/courses/advancedCM/lectures/samplelevel/writesoundfile.ck --silent Impulse i => dac; //=> Gain g dac => WvOut w => blackhole; "chuckoutput.wav" => w.wavFilename; 1 => w.record; //.9 => g.gain; now + 2::second => time later; //workaround to save file on end of synthesis: see examples/basic/rec.ck //null @=> w; while( now < later ) { //set current value of Impulse UGen Std.randf() => i.next; 1::samp => now; } 0 => w.record; w.closeFile;

1 0

Re: [chuck-users] build fails on LinuxMint 64 (was: 1.3.0.0 (chimera) unleashed!)
by Spencer Salazar 25 Aug '12

25 Aug '12

Hey Michael, I would guess that is a 64-bit issue based on that error message, and perhaps its causing problems now because of substantial internal changes between 1.2.1.3 and 1.3.0.0. I will investigate; the makefile should probably specify 32-bit compilation explicitly to avoid this, as it does on Mac OS. Ultimately we would like to have proper 64-bit support, of course! spencer On Sat, Aug 25, 2012 at 10:01 AM, Michael Heuer <heuermh(a)gmail.com> wrote: > Hello, > > The 1.3.0.0 build fails for me on LinuxMint 64 > > $ uname -a > Linux xxx 3.2.0-23-generic #36-Ubuntu SMP Tue Apr 10 20:39:51 UTC 2012 > x86_64 x86_64 x86_64 GNU/Linux > > $ make linux-alsa > ... > gcc -O3 -D__LINUX_ALSA__ -O3 -fno-strict-aliasing > -D__CK_SNDFILE_NATIVE__ -c ugen_xxx.cpp -o ugen_xxx.o > ugen_xxx.cpp: In function ‘void foogen_ctor(Chuck_Object*, void*, > Chuck_VM_Shred*, CK_DL_API)’: > ugen_xxx.cpp:1200:64: error: cast from ‘FooGen_Data*’ to ‘unsigned > int’ loses precision [-fpermissive] > ugen_xxx.cpp:1229:70: error: cast from ‘Chuck_Object*’ to ‘unsigned > int’ loses precision [-fpermissive] > ugen_xxx.cpp:1241:77: error: cast from ‘double*’ to ‘unsigned int’ > loses precision [-fpermissive] > make: *** [ugen_xxx.o] Error 1 > > > I don't think this is related to 64-bitness, or at least it is > something different than I get with 1.2.1.3, which builds fine but > core dumps when run. > > michael > > > Spencer Salazar <spencer(a)ccrma.stanford.edu> wrote: >> chuck-1.3.0.0 (chimera) is available: >> >> http://chuck.cs.princeton.edu/ >> http://chuck.stanford.edu/ >> >> This release incorporates ChuGins, a new way to extend ChucK. ChuGins >> are a continual work in progress, and we encourage feedback in >> anticipation of rapid updates, tweaks, and fixes. Chugens and >> Chubgraphs provide additional mechanisms for extending ChucK's >> built-in audio capabilities. Furthermore, this release variety of >> smaller fixes and additions. >> >> To start developing ChuGins, please see several examples here, in >> addition to the chuginate utility, which generates the skeleton >> framework for a new ChuGin. >> https://github.com/spencersalazar/chugins >> and: >> https://ccrma.stanford.edu/~spencer/publications/CCC2012.pdf >> >> (Please see the release notes at the end of this email.) >> >> Many many thanks once again to Kassen, Casper Schipper, Jorge Herrera, >> Hongchan Choi, and everyone on chuck-users, as well as PLOrk, SLOrk, >> the Food Services Division at Smule, Fernando Lopez-Lezcano, Carr >> Wilkerson, Chris Chafe, and the rest of the chuck community for making >> this release happen! >> >> Thanks and Happy ChucKing, UpChucKing! >> >> Best, >> Ge + Spencer, on behalf of entire chuck team >> >> --- >> 1.3.0.0 >> - (added) Chugins: dynamically loaded compiled classes/ugens >> see http://chuck.stanford.edu/extend/ for examples. >> - (added) new command line options >> --chugin-load:{auto|off} disable/enable chugin loading >> -gFILE/--chugin:FILE load chugin at FILE >> -GPATH/--chugin-path:PATH load all chugins in directory PATH >> --dac:NAME use dac with name matching NAME >> --adc:NAME use adc with name matching NAME >> - (added) Chubgraphs: create ugens by compositing existing UGens >> see examples/extend/chubgraph.ck >> - (added) ChuGens: create ugen by implementing audio-rate processing in ChucK >> see examples/extend/chugen.ck >> - (added) new ugens: >> - WvOut2: stereo uncompressed audio file output >> - SndBuf2: stereo uncompressed audio file input >> - (added) new functions: >> - me.sourcePath() >> returns file path of source file corresponding to this shred >> - me.sourceDir() >> returns directory of source file corresponding to this shred >> - MidiIn.open(string name) >> open MIDI input matching name >> - MidiOut.open(string name) >> open MIDI output matching name >> - Hid.open(string name) >> open HID matching name >> - (added) experimental OS X multitouch HID >> - (fixed) real-time audio on Mac OS X 10.7 (Lion), 10.8 (Mountain Lion) >> - (fixed) IO.newline() now flushes "chout" >> - (fixed) "chout" and "cherr" now invoke member functions correctly >> - (fixed) no for statement conditional handled correctly >> - (fixed) STK WvOut correctly handles case when file open fails >> - (fixed) << operator for arrays now correctly references counts >> - (fixed) crashing bug when receiving external events at a high-rate >> ("the OSC bug") >> - (fixed) scope closing correctly dereferences local objects >> - (fixed) crash when exiting shreds with connected ugens >> - (fixed) deallocation of "empty" Shred object no longer crashes >> - (fixed) crash when HID events are sent to an exited shred >> - (fixed) destructors for built-in objects are executed > _______________________________________________ > chuck-users mailing list > chuck-users(a)lists.cs.princeton.edu > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users >

1 0

chugins
by Kassen 25 Aug '12

25 Aug '12

Hey Spencer! I thought I'd send this to you through the list because it potentially affects others. I'll explain things you already know for them. You'll remember I (like Caspar) did a ChuGin for testing and fun. I wrote it, which wasn't very hard; nearly everything in the C++ that is unlike writing your own DSP in a ChucK Shred could be copy-pasted. After verifying that this worked and was fun I wrote some documentation for it so that others could enjoy it and I seem to remember I slapped the GPLv2 on it so that they'd be allowed to enjoy it in any way I'd call reasonable. All in all a nice package... But let's be honest; I'm no C++ wizard, English isn't my first language and I can generally be a bit distracted; there are quite possibly a issue or two with it. Back when I send it I asked for opinions but while we are being realistic; you probably have a lot of other things to review too, things that probably do have deadlines. Currently my "KasFilter", which is supposed to be a fun "non-clean" filter especially suited for kinds of music with a low-aesthetic that need a prominent filter, doesn't seem to be centrally distributed. I'd like it to be so I'd like to talk about how we can address the above in a way that makes everyone happy. For that I'd be perfectly willing to run tests, change stuff, etc and generally be a guinea-pig for such a process. If we don't I can host it too, but then I foresee having to maintain a list of links to ChuGins that people made, with all of the inconveniences that might bring. What I could for example do is make a GitHub account and request privileges to push my work into your repository? That'd centralise stuff though not everyone who'd like a few new UGens might be comfortable with Git and make. Yours, Kas.

2 1

miniAudicle 0.2.1 (gidora) released
by Spencer Salazar 24 Aug '12

24 Aug '12

Hello, miniAudicle 0.2.1 (gidora) is now available. http://audicle.cs.princeton.edu/mini/ This release integrates the latest version of ChucK (1.3.0.0, chimera). In addition to many new ChucK features, this fixes real-time audio on Lion and Mountain Lion. Please let us know about any problems, bugs, feature requests, etc. thanks, miniAudicle team --- 0.2.1 - (added)(all) chuck-1.3.0.0 - (added)(all) GUI chugin manager - (added)(Mac) default VM stall timeout changed to 2 seconds - (fixed)(Linux) build in Ubuntu 11.10 (thanks Ricardo Fabbri) - (fixed)(Windows) miniAudicle now exits properly

1 0

Comprehensive ChucK performance source code [was soft kill]
by Robert Poor 20 Aug '12

20 Aug '12

By the way, I posted my entire live ChucK setup on github. This includes the ramping down a patch when switching to another: https://github.com/rdpoor/chuck_performance_setup It's sparsely documented but very cleanly structured, and it's been performance-tested and debugged. I'm happy to document additional bits upon request. Perhaps the most important starting point to know is that it uses an Oxygen8 keyboard and a Game Trak "6 axis" controller as inputs. - Rob

2 2

soft kill
by Michael Heuer 19 Aug '12

19 Aug '12

Hello, I was having some fun with delays in ChucK when I should have been paying attention in a meeting and ran into something I may have asked about before -- is there (or should there be) any way any way to soft kill a shred, either by ramping down its gain or by waiting until the shred is no longer sending output to dac? If I $ chuck - 1 something that is running it kills it immediately, and if I don't advance time long enough explicitly at the end of a delay example, the nice feedback tail gets abruptly cut off. E.g. https://github.com/heuermh/lick/blob/master/examples/mfm2.ck (doesn't require LiCK to run) michael

7 12

Coin in Chuck
by ronni montoya 19 Aug '12

19 Aug '12

Hi , in supercollider there is a number method called coin. It just gives you back true or false based on probabilities. For example: 0.2.coin; -------------> this means 20 % chance for true. So if i do : if(0.2.coin,{ doSomething(); }); The function is gonna have 20% change to be called each time i run that code. Is there is something like this in chuck? cheers. R.

3 2

Insteresting sound when playing 2 or more instances at the same time.
by ronni montoya 18 Aug '12

18 Aug '12

Hi list, when i play 2 or instances of the same code it generate wierd results, it adds new sounds to the existing sounds. Do anybody have any idea whats the reason of this. Is this because some kind of lack in the memory? Any idea of how to replicate this sound from inside chuck Thanks Try this code 3 times and your are gonna listen rythmic patterns , for example: chuck chuqui.ck chuqui.ck chuqui.ck here is the code: Step s => dac; // modulator SinOsc m => blackhole; 0.00052 => m.freq; SinOsc g => blackhole; 0.0082 => g.freq; SinOsc t => blackhole; 0.00055 => t.freq; 6 => float v; 6 => float k; 6 => float n; fun void sample_manipulation( ) { 0.002 + k => k; 0.0002 + n => n; 210 * (now/ samp) => Math.sin => Math.tan => v; if (v > Math.sin(k / 5) + 0.95) { v + Math.sin(k / 10) => k; 0.008 * 10 / m.last()=> k; } if (v < Math.sin( k) + 0.3) { -0.2 * 2 / g.last() * t.last() => v; } if (v < 0.2 ||v > 0.8 ) { v + Math.sin(n / 25) => v; } v + ( v * 10 ) / m.last() * Math.atan( m.last() ) => v; v * ( v * 100 ) * m.last() * Math.asin( m.last() ) => v; } while (true) { sample_manipulation ( ) ; v => s.next; 1::samp => now; }

3 2