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

interval keyboard
by Michael Heuer 12 Jul '11

12 Jul '11

Hello ChucKers, I wanted to emulate something I saw on youtube once (sorry, I've lost the link), an interval keyboard where keys on the right hand play notes increasing by various intervals and keys on the left hand play notes decreasing by various intervals. I haven't any music theory and the wikipedia pages for muscial intervals are over my head. http://en.wikipedia.org/wiki/Interval_(music) http://en.wikipedia.org/wiki/List_of_musical_intervals My question is this, what should the method on Interval be called that "inverts" it (from low-to-high pitch to high-to-low pitch)? https://github.com/heuermh/lick/blob/f49ddd8285ee412f9b6c05c2ef5ba6c6cdb543… I have learned that is not the same as an inversion, which does something else to ratios https://github.com/heuermh/lick/blob/f49ddd8285ee412f9b6c05c2ef5ba6c6cdb543… The example is here if you want to give it a try https://github.com/heuermh/lick/blob/f49ddd8285ee412f9b6c05c2ef5ba6c6cdb543… Cheers, michael

3 4

OSC string concatenation bug
by Kyle Hipke 07 Jul '11

07 Jul '11

I'm writing a program that talks with Wekinator, but I'm having a problem with sending the feature names over OSC that I'm fairly certain is a problem with chuck. At the bottom of the code (see bottom of email) is the relevant method, sendOscFeatureNames(). When I run wekinator and get to the training screen, I run this code. The instant it sends the feature names to Wekinator (i.e., the instant it is run), an exception gets raised: " Exception in thread "Thread-4" java.lang.ArrayIndexOutOfBoundsException: 1536 at com.illposed.osc.utility.OSCByteArrayToJavaConverter.lengthOfCurrentS tring(Unknown Source) at com.illposed.osc.utility.OSCByteArrayToJavaConverter.readString(Unkno wn Source) at com.illposed.osc.utility.OSCByteArrayToJavaConverter.readArgument(Unk nown Source) at com.illposed.osc.utility.OSCByteArrayToJavaConverter.convertMessage(U nknown Source) at com.illposed.osc.utility.OSCByteArrayToJavaConverter.convert(Unknown Source) at com.illposed.osc.OSCPortIn.run(Unknown Source) at java.lang.Thread.run(Unknown Source) " This exception is not thrown when I replace numDataPoints[i] with a constant (like 5). Here is the entire .ck file. You can use it just like you would any OSC feature extractor with Wekinator. //Extracts the following for different numDataPoints //0 - centroidAvg //1 - centroidStdDev //2 - centroidMin //3 - centroidMax //4 - rmsAvg //5 - rmsStdDev //6 - rmsMin //7 - rmsMax //8 - fft bin with highest fval (highest over all windows) //... //whenever an amplitude > threshold is detected from input, computed over //numDataPoints 128-sample ffts that overlap by 64-samples // "127.0.0.1" => string hostname; OscSend xmit; xmit.setHost( hostname, 6448 ); OscSend xmit2; xmit2.setHost( hostname, 6448 ); //Custom objects adc => FFT f =^ RMS rms => blackhole; f =^ Centroid centroid => blackhole; f =^ Flux flux => blackhole; f =^ RollOff rolloff => blackhole; UAnaBlob b; //for storing the results from each window float rmsArr[50]; float cent[50]; float highestFFTBin[50]; //bin with highest fft fval float highestFFTVal[50]; //the actual value of the bin //Set up bin stuff 128 => int FFT_SIZE; FFT_SIZE => f.size; Windowing.hamming(64) => f.window; 1::second / 1::samp => float SR; SR/FFT_SIZE => float bin_width; //constants .5 => float threshold; 0 => int peakDetected; //flag if peak has been detected now => time lastPeakTime; 100::samp => dur peakWindow; 10::samp => dur peakPollRate; [1,5,10,15,20,25,30,35,40,45,50] @=> int numDataPoints[]; //list of numDataPoints values to try 1 => float rmsMultiplier; //so it isn't out of wek's range (this actually doesn't matter, leave it at 1) 0 => int currentlyAnalyzing; //flag if currently analyzing i.e. don't detect peak //Run the peak detector in parallel to set the peakDetected flag spork ~peakDetector(); spork ~sendOscFeatureNames(); //Extract features and send via osc when peak detected. while (true) { if (peakDetected) { <<<"Peak detected! Analyzing">>>; analyzeAndSend(); } .1::second => now; } //When peak detected, compute min, max, avg, and std. dev for centroid and rms //and send fun void analyzeAndSend() { 1 => currentlyAnalyzing; xmit.startMsg( "/oscCustomFeatures", "fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff"); //Get all float and rms values for the maximum numDataPoints getRmsCentroidAndFFT(); for (0 => int i; i < 11; i++) { analyzeNumDataPoints(numDataPoints[i]) @=> float result[]; for (0 => int j; j < 9; j++) { result[j] => xmit.addFloat; } } <<<"done">>>; 0 => currentlyAnalyzing; } //populates rms, cent, and fft arrays fun void getRmsCentroidAndFFT() { 50 => int maxDataPoints; 128::samp => now; rms.upchuck(); for (0 => int i; i < maxDataPoints; 1 +=> i) { rms.fval(0) => rmsArr[i]; centroid.upchuck(); centroid.fval(0) => cent[i]; 0 => float highestBinMagnitude; 0 => int highestBin; f.upchuck(); //find highest bin for (1 => int j; j < FFT_SIZE; j++) { if (f.fval(j) > highestBinMagnitude) { f.fval(j) => highestBinMagnitude; j => highestBin; } } highestBin => highestFFTBin[i]; highestBinMagnitude => highestFFTVal[i]; } } //Analyze data using the specified number of data points, return the float array corresponding to //the osc message format described at top //requires that the arrays have been populated fun float[] analyzeNumDataPoints(int numDataPoints) { float centroidTotal; float centroidMin; float centroidMax; float centroidStdDev; float rmsTotal; float rmsMin; float rmsMax; float rmsStdDev; float highestBin; float highestBinMagnitude; new float[numDataPoints] @=> float centroidData[]; new float[numDataPoints] @=> float rmsData[]; //get 1st centroid and rms rmsArr[0] +=> rmsTotal; rmsArr[0] => rmsMin => rmsMax => rmsData[0]; cent[0] +=> centroidTotal; cent[0] => centroidMin => centroidMax => centroidData[0]; //do rest for (1 => int i; i < numDataPoints; i++) { rmsArr[i] +=> rmsTotal; Math.max( rmsArr[i], rmsMax ) => rmsMax; Math.min( rmsArr[i], rmsMin ) => rmsMin; rmsArr[i] => rmsData[i]; cent[i] +=> centroidTotal; Math.max( cent[i], centroidMax ) => centroidMax; Math.min( cent[i], centroidMin ) => centroidMin; cent[i] => centroidData[i]; //compute current highestBin if (highestFFTVal[i] > highestBinMagnitude) { highestFFTVal[i] => highestBinMagnitude; highestFFTBin[i] => highestBin; } } //calculate rmsTotal / numDataPoints => float rmsAvg; centroidTotal / numDataPoints => float centroidAvg; //get std dev float rmsDevSquaredTotal; float centroidDevSquaredTotal; for (0 => int i; i < numDataPoints; i++) { (rmsData[i] - rmsAvg)*(rmsData[i] - rmsAvg) +=> rmsDevSquaredTotal; (centroidData[i] - centroidAvg)*(centroidData[i] - centroidAvg) +=> centroidDevSquaredTotal; } Math.sqrt(rmsDevSquaredTotal / numDataPoints) => rmsStdDev; Math.sqrt(centroidDevSquaredTotal / numDataPoints) => centroidStdDev; float result[9]; centroidAvg => result[0]; centroidStdDev => result[1]; centroidMin => result[2]; centroidMax => result[3]; rmsAvg * rmsMultiplier => result[4]; rmsStdDev * rmsMultiplier => result[5]; rmsMin * rmsMultiplier => result[6]; rmsMax * rmsMultiplier => result[7]; highestBin => result[8]; return result; } //Pretty inefficient but does the trick //Looks for a peak in the last peakWindow fun void peakDetector() { while (true){ if (adc.last() > threshold || adc.last() < -1 * threshold) { 1 => peakDetected; // analyze now => lastPeakTime; } else if (now > lastPeakTime + peakWindow) { 0 => peakDetected; } peakPollRate => now; //This is silly; would be better do do a low-pass filter envelope and look for peaks there. But don't worry about it right now. } } //run in parallel to send osc feature names fun void sendOscFeatureNames() { while (true) { xmit2.startMsg( "/oscCustomFeaturesNames sssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss"); for (0 => int i; i < 11; i++) { numDataPoints[i] => int n; xmit2.addString(n + "centroidAvg"); xmit2.addString(n + "centroidStdDev"); xmit2.addString(n + "centroidMin"); xmit2.addString(n + "centroidMax"); xmit2.addString(n + "rmsAvg"); xmit2.addString(n + "rmsStdDev"); xmit2.addString(n + "rmsMin"); xmit2.addString(n + "rmsMax"); xmit2.addString(n + "highestFFT"); } .1::second => now; } }

1 0

ICMC-unConf
by Scott Hewitt 06 Jul '11

06 Jul '11

Hello, Please except my apologies about cross posting, however I have submitted an idea to the unconf concerning the use of ChucK and Git version control http://icmc-unconf.inclusiveimprov.co.uk/doku.php/unconf/proposals/chuck-an… This message is being cross posted to further publicise the Un-Conference events running along side the 2011 ICMC . . . http://icmc-unconf.inclusiveimprov.co.uk/ This year’s International Computer Music Conference is being hosted by the University of Huddersfield, Sunday July 31 -- Friday August 5 ( http://icmc2011.org.uk ) The unConf features presentations, discussions and conversation on Monday 1st August and Tuesday 2nd and Gigs on Sunday 31st July and Wednesday 3rd. What’s an unConf? Its a conference without the formality and fixed proceedings. If you have something you want to discuss or present then head online and submit an idea otherwise turn up on the day and start a conversation, best thing about attending the unConf is it is free of charge. While it is not formally part of the conference the unConf does give a great opportunity to meet with international practitioners. Hope to see some of you there, ii

1 0

urgent: documentation problem i've not gotten an answer to
by tempjayren＠gmail.com 03 Jul '11

03 Jul '11

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 i can't read the docs properly. reading the pdf file, straight or converting it to text and reading that shows this. %>chuck [chuck]: no input les... (try {help) it is supposed to be files, not pagebreakles and reading code fragments done in pdf does the same. i can't learn chuck this way. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJODsPmAAoJEIaKvi7tbzRHpLoH/1x1sPKA/1j20FZ+H067iGKg PUwlSOhNiOFJ+ikSYxuP3spNQAqlEDVegvrnGoMXWM0o5rQ+AQjfLOkjF3ZxcTYB E3mk9c+F0WvblHzLsth55BkW3ln2WuVwF1mc6gcVFrJVQoEJL4S/lxEoMp+el7Yy rwcB4q3qYT1aPRKs8kp0nPAH1+jhf5Yj+Fp6Ta4aEMgepUnJLke2f7BFNcGMca9/ ZgolstsW2k0t8A6k+h4P6zOs2LAmImAonlJvBtelnIF7EiApYDFAVjPG21zflOOm VGf7TbN6bq36afmQX3AUKKZ362h6URthpUdMjyuT33U6tbIHUA5JvFvAUGB2iPA= =arIG -----END PGP SIGNATURE-----

2 3

parallel processing
by Kassen 03 Jul '11

03 Jul '11

http://t.co/4lUPedY Paper on designing a new engine for SC that makes use of parallel processing. Kas.

1 0

link in wiki broken
by tempjayren＠gmail.com 02 Jul '11

02 Jul '11

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 it would appear that the link for the ChucK text editor (XP) : Jon Fisher in the wiki is broken. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJOD0evAAoJEIaKvi7tbzRHRgkIAJeHHLyGWxigA4jiu8BgP/TW jE0aA5Vmg+ZdkawVw30QBFPSUKGRGocyXOCgwLD1ymS+25mAqwGybGbrmufXGGTH t3k2I5Shlt5UJ0r60V6zdf8hvTUApKARGHCw5ON3e8KYTn4ED2e6yfJj0wI3RecJ GvnAo8nskv0efXdpnRv5DgyV6b7RIvNPDG6zAdsCSda77GKlfDdwG+kuRfRKhmPw ValX3o9o9PyGs8afFtj2+MCz4xnX9VamdtiTz2yFikf1ou8TZPkfzk/NB/2vPQfl 3yQvptsxdlGnQRHtOY0DRdOfi/Xc537xKs7ZHnA7sPJmaJX/7mdxqt/pofqqwI8= =6doV -----END PGP SIGNATURE-----

1 0

avoiding lag
by Timothy Leonido 27 Jun '11

27 Jun '11

Hey All, Is there a way to ensure that two patches will synch properly? Sometimes the simultaneous chucking of two heavy patches will cause a lag: one will be set back .5::T behind the other. I tried machine add for this, but ran into the same problem. Thanks, as always, & nice weekend, Tim

4 3

documentation: pdf totextplain and speech synthesizer
by tempjayren＠gmail.com 26 Jun '11

26 Jun '11

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 i have the pdf version of the chuck manual, very nice that, though because i'm using a speech synthesizer, the easier thing for me is to convert to text and to read that way, either that or old habits die hard, i haven't played with pdf and text access with speech lately, it might have gotten better since early 2000 when i toyed with it last. anyway, enough of that, i want to make the pdf manual text, however when i do so it doesn't appear to convert properly for some reason. it is hard to follow directions and learn from tutorials when the text isn't exactly what was intended. am curious if anyone else has encountered this or a solution for such. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJOButJAAoJEIaKvi7tbzRHPl8H/AtbTuUgc7cv3McTdjrg6U+c HkM9olDQ5TdFLTCcVBThg4xQ2Gd5vKaCRnea0rc5b4TQ2qLbOWlxdLvzSCGYxOBr SH+7QEUnfr6Q7Ko7dHpZDfXBP2bjK/dBw+MgKukIZ1uFN7D8WtUHH4uMDGeRFcxT bwU5lFadDP6rCrU1zPFu5iwXLM7Fc6x1QXXpA8ZdM5Vf4bUDll+JoVweVj/VtH4P 7KN3PCkV/ZzzjaeWcPhYNJfHPKgbMKOutKfsC/P/manoevi1XwltsnUKfVB1aE7u EUM9mHur7fBRwgyJ/m6SIV3a1vNYRs1/4BXjIStT6N2SWpJH+Z8E4Rl1w4oheso= =A6dJ -----END PGP SIGNATURE-----

1 0

OT: Prague - Electronic Music, etc.
by mike clemow 20 Jun '11

20 Jun '11

Hey Chuckists, So, I find myself in Prague for work and I would love to see some electronic music / meet Chuck people / even get a gig (?) while I'm here. Any advice / ideas / contacts? I've been here for a week and a half, and I'm here until July 5th. If any of you will be in the Czech Republic, please holler. Best, Mike -- http://michaelclemow.com

2 1

Virtual Audio World (VAW)
by Les Hall 17 Jun '11

17 Jun '11

I have decided to create a ChucK-based distributed virtual audio world. In this world we will have places which are physically located on each other's computers and virtually located throughout cyberspace. The first place in this world will be on my Mac mini. I have received permission to use five of the public streaming ports on radio.electro-music.com for this world temporarily. In the event that it grows, we will address the issue of expansion with electro-music server resources, however this world will not be restricted in any way other than morally, ethically, and legally as appropriate. This means that if you have streaming resources you may use them as well. The focal point for discussion of this virtual audio world will be a thread that I just started on the ChucK forum of electro-music.com. The thread is here: http://electro-music.com/forum/viewtopic.php?t=48042 Tonight I will set up a two-channel version of this world so that I and up to two guests can participate as soon as possible. It will use public ports public_1, public_2, and public_3, the latter two of which are for guests of my place in this world. I will post full information of the admittedly somewhat complex task of setting up for streaming into this world in the thread linked above. I openly invite your participation. Les (Inventor)

2 5