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

chuck-1.4.1.0 (numchucks) released!
by Ge Wang 26 Jun '21

26 Jun '21

Dear all, chuck-1.4.1.0 (numchucks) is available: http://chuck.stanford.edu/ http://chuck.cs.princeton.edu/ This is a significant update to ChucK that addresses many issues while adding new features, UGens, examples, and internal groundwork for future directions. We heartily recommend updating to this latest version. The release notes are reproduced below and can be found at this URL: http://chuck.stanford.edu/release/VERSIONS Thank you and keep on chuckin'... Perry, Spencer, Ge and ChucK Team ------- 1.4.1.0 (NOTE) chuck-1.4.1.0 is the first significant release resulting from a coordinated team effort to update and improve ChucK for teaching and research. In April 2021, Perry and Ge led a ChucK "Global Repair-a-thon", which identified issues and put forth a new Development Roadmap. The ChucK dev team will follow this roadmap for updates, new features, and new ChucK mediums (e.g., Chunity, WebChucK, FaucK, ChucKTrip, and more). Many people were involved in this significant release, including Jack Atherton, Mike Mulshine, Matt Wright, Rebecca Fiebrink, Dan Trueman, Lloyd May, Mark Cerqueira, Mario Buoninfante, Perry Cook, Spencer Salazar, and Ge Wang. Many thanks to everyone who contributed. Happy ChucKing! --- - (added) static Object.help(); All ChucK objects (i.e., all non-primitive types) can now invoke .help(), which will dynamically generate and print to console the description and inheritance hierarchy of the underlying type, as well as the available functions at runtime. This can be called on any object instance or on any class (without an instance): // any object instance SinOsc foo; // print to console info about its underlying type foo.help(); (OR) // call help() from any class (identical output as above) SinOsc.help(); Intended to be a quick runtime reference / learning tool! - (added) examples/help.ck -- show .help() in action - (added) examples/deep/chant.ck (by Perry Cook, 2006) -- a chant singing synthesizer that makes use of ChucK's strongly-time and concurrent programming features. - (added) new Multi-channel LiSa UGens! (LiSa10 was already present) LiSa2 (stereo) LiSa4 (quad), LiSa6 (6-channel; laptop orchestra edition!) LiSa8 (8-channel), LiSa16 (16-channel) - (fixed) LiSa playback bug; voice panning now handles the mono case; previously was silencing voice playback - (added) new STK set/get access functions in FM: float op4Feedback( float value ); Set operator 4 feedback. float op4Feedback(); Get operator 4 feedback. float opADSR( int opNum, float a, float d, float s, float r ); Set operator ADSR: attack, decay, sustain, release. float opAM( int opNum, float value ); Set operator amplitude modulation. float opAM( int opNum ); Get operator amplitude modulation. float opGain( int opNum, float value ); Set operator gain. float opGain( int opNum ); Get gperator gain. float opRatio( int opNum, float ratio ); Set operator frequency ratio. float opRatio( int opNum ); Get operator frequency ratio. float opWave( int opNum, int wave ); Set operator waveform [1-8]. - (added) new Synthesis Toolkit (STK) FM instruments from Perry! HnkyTonk: STK-style Honkey Tonk Piano FM synthesis instrument. (aka, Algorithm 1 of the Yamaha TX81Z DX11 FM synthesizer) FrencHrn: STK-style French Horn FM synthesis instrument. (aka, Algorithm 2 of the Yamaha TX81Z DX11 FM synthesizer) KrstlChr: STK-style "Crystal Choir" FM synthesis instrument. (aka, Algorithm 7 of the Yamaha TX81Z DX11 FM synthesizer) *** for each of these, try using the new .help() to examine its documentation and interface; e.g., HnkyTonk.help(); *** - (added) new Moog function: float filterStartFreq( float freq ); Set filter starting frequency. float filterStartFreq(); Get filter starting frequency. - (added) new STK examples: examples/stk/frenchrn-algo2.ck -- FM French Horn demo examples/stk/honkeytonk-algo1.ck -- fun with FM Honkey Tonk examples/stk/krstlchr-algo7.ck -- heavenly FM Kristal Choir examples/stk/hevymetl-acoustic-algo3.ck -- FM HevyMetl => Nylon Guitar examples/stk/hevymetl-trumpet-algo3.ck -- FM HevyMetl => Trumpet examples/stk/rhodey-song.ck -- a lovely little number - (updated) tweaks to a number of STK UGens, which may result in audio changes; the affected UGens are: BandedWG, BeeThree, FMVoices, HevyMetl, PercFlut, Rhodey, TubeBell, Wurley - (fixed) issue in PluckTwo in computing delay length; this affects the tuning of various plucked string models, including Mandolin. - (added) .frames() for SndBuf and SndBuf2; returns the number of sample frames in the loaded sound file. This is identical in behavior to .samples() -- but for stereo+, .frames() is semantically more accurate than .samples(). - (added) "global" declarations for int, float, string, Object, UGen, Event; this can be used in-language across ChucK files (each file must first "declare" the global value before using it). This also support usage from external hosts such as Chunity. (thanks to Jack Atherton for this work!) - (added) constant power panning (as controlled by the .pan parameter) for all UGen_Stereo--including Pan2, stereo dac, stereo adc; this is controllable using the new .panType parameter. By default, constant power panning is NOT enabled for dac and adc, whose panning behavior is unchanged (panType==0) and favors a kind of unity-gain-preserving but perceptually suspect panning scheme. By contrast, Pan2 will default to use constant power panning (panType==1). This change will alter the effect of any existing code that uses Pan2. Specifically, it will introduce a ~3 dB reduction in the output Pan2 (it is possible to restore the previous behavior by explicitly setting .panType to 0). This ~3 dB reduction corresponds to the equal-powered amplitude value when pan is at center (pan=0). Mathematically, the reduction amount can be calculated using Std.lintodb(Math.cos(pi/4)) (To hear this new behavior, check out): examples/stereo/stereo_noise.ck - (BEHAVIOR CHANGED) clearly separated array.size() and array.capacity(); for historical/compatibility reasons, array.cap() is the same as array.size() and generally should be avoided in favor of either .size() or .capacity() - (BEHAVIOR CHANGED): internally, all ChucK static functions are now passed an extra parameter (Chuck_Type * TYPE) as a pointer to the base type/class. This will affect Chugins, which should be updated/ rebuilt with the latest headers. - (BEHAVIOR CHANGED): when connecting two multi-channel UGen's (both are at least stereo), say LHS => RHS, if LHS channels < RHS channels, the channels will be connected one-by-one up to the LHS channel count; previously, the LHS channels will be automatically repeated to connect to the RHS channels that have no corresponding LHS channels. This behavior can still be achieved by manually connecting the individual channels of LHS and and RHS through LHS.chan(n) and RHS.chan(n). - (fixed) resolved a bug with SndBuf not playing when .rate < 0 and while .pos beyond the end - (fixed) resolved a bug with SndBuf where it could output DC when .rate < 0 after it reaches the beginning of the file (e.g., if the first sample is non-zero) - (fixed) resolved a bug in ADSR when sustain is set to 0 and audio stops, .releaseTime return NaN and basically the world ends (thanks to mariobuoninfante for help in addressing this) - (fixed) reset Gen10 table to allow add/remove partials; thanks to mariobuoninfant - (fixed) updated MidiOut and removed check in order to allow messages of size less than or greater than 3 bytes; thanks to mariobuoninfant for this issue report on Ubuntu Studio 64-bit; this fix addresses the sending of MIDI clock, stop, continue, and SysEx -- also thank you mariobuoninfante for the fix. - (fixed) ChucK shell can now be exited either by the "exit" command or through ctrl-c. Previously. ctrl-c sometimes failed to end the the program due to logic in the global cleanup function; this has been addressed. - (fixed) addressed an issue where Chugins were being loaded twice. - (fixed) the --chugin-load:off option is now correclty taken into account and will appropriately result in no Chugins being loaded. - (deprecated) Chubgraph => (new) Chugraph; Chubgraph will continue to function, with a deprecation warning. - (documentation) updated embedded documentation text for all statically linked UGens and libraries. - (documentation) brought examples in alignment between distribution and the online examples collection; many examples updated for clarity and utility http://chuck.stanford.edu/doc/examples http://chuck.cs.princeton.edu/doc/examples - (internal/refactor) the "Mega Merge 2021" integrates a long-standing branch since REFACTOR-2017 that includes support for globals and was the basis for Chunity. Future Chunity releases will align with mainline ChucK language features. (thanks to Jack Atherton for several years of laying the groundwork and for ChucK in Chunity!) - (internal/developement) added m_started flag to the ChucK class to avoid a strange (but mostly harmless) condition when the virtual machine is shutting down and ChucK::run() attempt to call ChucK::start() - (internal/development) Thanks to Mark Cerqueira -- migrated from old Travis CI (continuous integration testing) to GHA (GitHub Actions).

2 1

SndBuf | accessing samples from stereo file
by mario buoninfante 17 Jun '21

17 Jun '21

Hi, I'm working on an algorithm that requires accessing samples from a stereo file and am using SndBuf to achieve that. I was expecting that to work so that samples were returned in "frames", for example: ''' buffer.valueAt(0) = sample 0 on left channel buffer.valueAt(1) = sample 0 on right channel buffer.valueAt(2) = sample 1 on left channel buffer.valueAt(3) = sample 1 on right channel ... ''' but that doesn't seem to be the case. Also SndBuf.samples() always returns the same amount of samples, regardless the file being mono or stereo (the size matches the time in msec, so it seems that reports the size of a single channel). I also played around with .channel(n), but I get the same set of samples regardless the channel number used. Then I just now found out that basically the channels seem to be stored (or at least accessed) sequentially. Let's say you have a stereo file containing 1024 samples (512 per channel), you get: samples 0-511 = left channel samples 512-1023 = right channel I was wondering if my assumption is correct, but also if this is intentional. Cheers, Mario -- musician, QA engineer http://mbuoninfante.tumblr.com/ https://vimeo.com/creativecodingsalerno https://github.com/mariobuoninfante

2 4

ResonZ Documentation
by Matt Comeione 17 Jun '21

17 Jun '21

I was reviewing some code from the “Physics-Based Sound Syntheses for Games and Interactive Systems”. Some of the code examples utilize .gain on the ResonZ filters which is not outlined in the documentation. I was able to extract some information on the documentation of the vec3 and vec4 by looking at the code (additionally have submitted this documentation to GitHub), however, the ResonZ code is a little more elusive. Is there someone that might have some insight into the code that could point me in the direction to where I might find the appropriate relevant code?

3 5

Accessibility request.
by Gabriele Battaglia 16 Jun '21

16 Jun '21

Hi All. I'm new to Chcuk and blind since born. I'd like to contact the Audicle's DEV in order to request accessibility to this IDE which is totally inaccessible under Windows. May some of you can help me with this? Many thanks in advance. Gabe. -- Gabriele Battaglia, IZ4APU (GMail) Sent from TB on Windows 10, Genus Bononiae's computer.

1 0

Chugin code signing issue on macOS
by David Braun 04 Jun '21

04 Jun '21

I've had success with a Chugin on Windows but have an issue on macOS. I build the Chugin in terminal: cmake -Bbuild cd build make Then I rename the libWarpBuf.dylib to a WarpBuf.chug and place it in /usr/local/lib/chuck/ This is the console log in miniAudicle when trying to use the chugin. [chuck]:(3:SEVERE): | | | error from chuck_dl: 'dlopen(/usr/local/lib/chuck/WarpBuf.chug, 2): no suitable image found. Did find: /usr/local/lib/chuck/WarpBuf.chug: code signature in (/usr/local/lib/chuck/WarpBuf.chug) not valid for use in process using Library Validation: mapped file has no cdhash, completely unsigned? Code has to be at least ad-hoc signed.' How can I fix this? I also tried cmake -Bbuild -G Xcode and then compiling using the xcode project, with various experiments with provisioning profiles / code signing, but it didn't work. Thank you!

1 0

Building MiniAudicle on RaspberryPi 4
by Matt Comeione 20 May '21

20 May '21

I’m attempting to build MiniAudicle on a RaspberryPi 4, however, I am getting the following error in the build process: /usr/bin/ld: cannot find -lqscintilla2 The dependency required for apt based systems is indicated as libqscintilla2-dev, however, on Raspbian the closest library is libqscintilla2-qt4-dev which appears not to be correct. Any suggestions on getting this to build?

1 0

New chugin that warps audio like Ableton Live
by David Braun 18 May '21

18 May '21

I created a chugin that warps and transposes audio like Ableton Live does with clips. It actually parses Ableton ".asd" files in a kind of hilariously hacky way. https://github.com/ccrma/chugins/pull/51 Control parameters: - .read - ( string , WRITE only ) - loads file for reading - .playhead - ( float , READ/WRITE ) - set/get playhead position in quarter notes relative to 1.1.1 - .loop - ( int , READ/WRITE ) - toggle looping - .play - ( int , READ/WRITE ) - toggle playing - .bpm - ( float , READ/WRITE ) - set/get BPM rate ( Audio files without Ableton warp files are assumed to be 120 BPM ) - .transpose - ( float , READ/WRITE ) - set/get pitch transpose in semitones - .startMarker ( float , READ/WRITE ) - set/get start marker of the clip - .endMarker ( float , READ/WRITE ) - set/get end marker of the clip - .loopStart ( float , READ/WRITE ) - set/get loop start marker of the clip - .loopEnd ( float , READ/WRITE ) - set/get loop end marker of the clip - .reset ( float , WRITE ) - reset the internal process buffer of the Rubberband stretcher It's working well in my tests on Windows, so please give it a try, especially if you're on macOS which I haven't been able to test. Cheers, David

1 0

building miniAudicla on debian "cannot find -lqscintilla2"
by Atte 12 Apr '21

12 Apr '21

Hi! I'm trying to build miniAudicle under debian but the build fails like this: /usr/bin/ld: cannot find -lqscintilla2 collect2: error: ld returned 1 exit status make[1]: *** [makefile.qt-alsa:270: miniAudicle] Error 1 make[1]: Leaving directory '/mnt/data/atte/medium/software/chuck/miniAudicle-1.3.5.2/src' make: [makefile:38: linux-alsa] Error 2 (ignored) If I try to install libqscintilla2-dev I get this: atte@pandrup:~/software/chuck/miniAudicle-1.3.5.2/src$ sudo apt-get install libqscintilla2-dev Reading package lists... Done Building dependency tree Reading state information... Done Note, selecting 'libqscintilla2-qt4-dev' instead of 'libqscintilla2-dev' libqscintilla2-qt4-dev is already the newest version (2.10.4+dfsg-2.1). Any pointers appreciated. -- Atte http://latestyoutube.a773.dk | http://a773.dk

2 1

PhD postion on the ERC-funded Intelligent Instruments project
by thor list 07 Apr '21

07 Apr '21

Hello ChucKers, and apologies for x-posting! We are seeking applications for the first PhD position on the ERC-funded Intelligent Instruments research project hosted at the Iceland University of the Arts. This is a three-year fully-funded scholarship. The candidate will propose their own project under the rubric of Intelligent Instruments (see advert) and work in an exciting lab in the heart of Reykjavik, Iceland. Deadline is May 17th and the start date for the project is Sept 1st, 2021. Further information here: https://www.lhi.is/en/intent <https://www.lhi.is/en/intent> Please share and forward this email or the above URL! Looking forward to hear from you - happy to answer any questions at thor.magnusson(a)lhi.is <mailto:thor.magnusson@lhi.is> All best Thor Magnusson https://thormagnusson.github.io <https://thormagnusson.github.io/>

1 0

Newbie troubleshooting Mac/Windows
by Tim Burgess 24 Mar '21

24 Mar '21

Hi, I've installed ChucK 1.4.0.1 using the provided installers on both the Mac Catalina and Windows 10 partitions of my BootCamped MacBook Pro. Booted to the Mac side of things, I created and successfully ran the first tutorial program, as far as randomizing the sine wave frequencies, so that's great. I then booted to Windows and used the same file to validate my installation but, unfortunately, I got no audio output and no error messages. Pressing Ctrl+C generates a message saying "cleaning up", but the command prompt never returns. Can anybody help out a newbie with troubleshooting tips, please? Cheers. Tim

2 13