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

sndbuf cleanups?
by James Hughes 23 Jun '07

23 Jun '07

I've got a chuck app that runs great for about thirty minutes and then starts acting flaky. Sound starts cutting out, gets progressively worse, and then within about five minutes it ceases to make any sound. It doesn't crash, it just stops making sound. When I fire up Activity Monitor (OS X) to see what's going on, there doesn't appear to be anything obvious. Memory usage climbs steadily over the 30 minutes but caps off at around 25-30 megabytes. CPU never really gets above 75% (out of 200% on my dual core computer). The application is kind of complicated, but essentially it starts up 6 shreds, each in control of their own SndBuf. Here's a quick overview of what each shred does: Request a file name from another app via OSC Load the wav file into a SndBuf Play it for a little while (<10 minutes), skipping around in the file (mostly rewinding to the beginning of the clip). Request a new file name (repeat) The sound files are 30-150k each. It cranks through over 100 sound files before problems start happening. It doesn't look like garbage collection ever happens, but it's not using that much memory. Here's the relevant bit of code that takes the OSC message and loads up the clip: oe.getInt() => int player; oe.getString() => string snd; // allocate a new sound buffer new SndBuf @=> newClip[player]; // stick the new sound in the new sound buffer snd => newClip[player].read; Anyone have any idea what's going on? Should I be destroying these SndBufs more explicitly rather than just dropping their references? The manual says memory management is automatic, but mentions something about garbage collection not being implemented quite yet. That's ok since I can restart the chuck process every once in a while, but I need to get at least a two hour performance out before I restart. If you're curious, here's a link to the project writeup (a little out of date these days): http://www.virtualjames.com/mechatronics/final-project.html Thanks, James

3 2

SndBuf causes hickup
by Atte André Jensen 23 Jun '07

23 Jun '07

Hi I have (and have had for as long as I can remember) problems with SndBuf. I'm pretty sure the trouble is when loading a sample, then there will most often be a small but very noticeable hickup. It's quite bad with larger samples but also there with even small samples. Generally; the larger the sample, the greater the hickup. The problem seems to be gone or almost gone when the sample has already been loaded once, but this might be due to caching in the underlying os. I've tested with both alsa and jack, and all kinds of combinations of buffersizes and what have you, but although the problem seems the vary a bit (in an unpredictable pattern) it's always there. I also tested both syntaxes ("sample" => s.read and s.read("sample")), however only with wav files. I also tried both in miniAudicle and commandline chuck. Needless to say this makes SndBuf useless on my system :-( Is this a result of me doing something wrong or a bug in chuck? In any case is there anything I can do to remedy this right now? If it's a bug, I'd be most happy to help in any way I can with getting it fixed as soon as the developers time permits. BTW: This is under linux (debian/stable) on a IBM T41, 1.6Ghz, 1G ram and 7200 RPM HD. The system is running with a realtime patched kernel and is capable of playing back 20+ tracks in ardour with effects + various soft synths running along in realtime without a single xrun or glitch. Here's a simple example of something that shows the problem: http://atte.dk/download/sndbuf_test.tgz. First run shaker.ck, then after a while add substitution.ck (a beat => big hickup) or sd.ck (a snare drum => small hickup). -- peace, love & harmony Atte http://atte.dk | http://myspace.com/attejensen http://anagrammer.dk | http://atte.dk/compositions

8 15

MIDI channel / Interfaces / Machine.Remove /
by mieslings＠i12.com 17 Jun '07

17 Jun '07

Thanks, Kassen, for your reply. >...This is asuming you meant "device" above, not "channel", the MIDI >chanel would be in the message. Yes, sorry, my mistake - I meant device. >Hope that helps at least a little, not sure what you are after here, exactly. It does and I probably wasn't clear. I'm curious to know if, when a shred terminates, there is a way for the application to be made aware of it. For example, there could be a onShredExit(Shred@ shred) callback function automatically called by ChucK just before the shred terminates whose content would be left at the discretion of the user so that for instance s/he could write some code to notify all the objects dependant on this shred. But I guess there would be potential scope issues... What about functors or some sort of pointers to functions and member functions? Now, that would be really great! Mies

2 1

MIDI channel / Interfaces / Machine.Remove / Functors (!)
by mieslings＠i12.com 17 Jun '07

17 Jun '07

Hello - this is my first post here but I've been experimenting with ChucK for quite a while, and have a few questions I can't manage to find the answers to by myself... I'm more familiar with the likes of OpenMusic or MaxMSP but enjoy the power of the ChucK language as well as the fact that all it requires is chuck.exe!!! (and the manual ;) 1. Is there a way to retrieve from MidiIn/MidiOut objects the channel originally passed to open or do I need to encapsulate open in a method of my own in a derived class? (I would override open but with super not implemented yet, I don't know how to call the implementation of open in the superclass) 2. If I'm not mistaken, ChucK only accepts single inheritance; are you (the language designers behind ChucK :) planning to implement some sort of mechanism similar to Java/.NET interfaces? I've implemented a dispatching mechanism for MidiIn (so that I don't have take care again and again of the synchronisation issues etc. and have a simple, reusable mechanism to map say a note on message with some onNoteOn callback each time I use MidiIn). I now would like to generalise it and I don't know how without multiple inheritance/interfaces... 3. Machine.remove(shred.id()) terminates a shred (great feature!) but could someone clarifies what happens behind the scene? In particular, is there a callback method automatically called when a shred exits that users could hijack to perform some cleanup? 4. Finally, does ChucK support functors (perhaps even through some clever mechanism to the same effect)? Mies

2 1

multispork
by Atte André Jensen 10 Jun '07

10 Jun '07

Hi I'd really like to be able to group shreds, making it easier to have several shreds start/stop at the same time. First I'd like to hear how others address the problem... Then I thought about some possibillities: 1) Using Machine.add/remove. This is easy if I use Machine.add to spork everything, but first replace (in miniAudicle) doesn't work with shreds sproked from Machine.add. Secondly shreds added from miniAudicle can't be removed from Machine.remove (since that function requires a function). The best solution would be if Machine.remove accepted a string, in which case it would remove all shreds with that name. To complete everything miniAudicle should be able to remove and replace shreds added from Machine.add. 2) Some more fancy sporking mechanism could be added to miniAudicle. I think this has been brought up before (correct me if i'm wront), and although it would be nice, it might be difficult to implement something flexible enough for everyone. For instance I would "need" an add-list (spork these if not running) and a remove-list (remove if running) both associated with a group. -- peace, love & harmony Atte http://atte.dk | http://myspace.com/attejensen http://anagrammer.dk | http://atte.dk/compositions

2 1

more miniAudicle things
by Atte André Jensen 10 Jun '07

10 Jun '07

Hi There are a few more things regarding miniAudicle that would be nice (I'm on linux BTW): 1) When opening a file the focus should be directed to the edit area. Now it's just at the window, which means you have to click in the text area before you can type. Almost similar: Accessing save as (or save for unsaved file) leaves focus on I-don't-know-what, whereas is should go to the file name field in the dialog. 2) The native linux cut'n'paste by select and middle-click would be appreciated. Also thinks like ctrl-k (kill line) and ctrl-y (yank from kill buffer) would be great. They might take on another form, but a copy/paste without touching the mouse or some sort would be nice. 3) I think pressing tab in the editor should always mean "indent current line to relevant intention". I looked at OnCharAdded in mAView.c, but it seems that tab presses are never showing up in event.GetKey(). I'm not really famliar with cpp/wx (I did however work a bit with wxPython) but maybe it's a instantiation flag for the window or something. Unfortunately I wasn't able to figure out how to get tab to show up in GetKey. Maybe someone here (Spencer ?) knows how? If so, I think getting tab to behave as I suggest would be possible even for me :-) -- peace, love & harmony Atte http://atte.dk | http://myspace.com/attejensen http://anagrammer.dk | http://atte.dk/compositions

2 1

two gui wishes for miniAudicle
by Atte André Jensen 10 Jun '07

10 Jun '07

Hi (I'm back after too long) I have two minor requests for the mini: 1) Make the char position avaliable somewhere, so that when you get "line(2).char(12): syntax error" you know where to look. Even better (and not so minor) would be if were possible to click on the error in the Console Monitor and the cursor in the editor would jump to the correct location (or possibly jump to the first error automatically). 2) keyboard shortcuts for switching between tabs in the editor. -- peace, love & harmony Atte http://atte.dk | http://myspace.com/attejensen http://anagrammer.dk | http://atte.dk/compositions

5 7

win32 woes
by James Hughes 07 Jun '07

07 Jun '07

I'm vainly trying to get chuck to work with my Via Epia CN10000EG in 5.1surround sound. It's running XP SP2. I'm able to play 6 channel audio files using mplayer. When I probe for audio devices I don't get any 6 channel capable devices: =========================================================== [chuck]: found 4 device(s) ... [chuck]: ------( chuck -- dac1 )--------------- [chuck]: device name = "Primary Sound Driver" [chuck]: probe [success] ... [chuck]: # output channels = 2 [chuck]: # input channels = 0 [chuck]: # duplex Channels = 0 [chuck]: default device = YES [chuck]: natively supported data formats: [chuck]: 8-bit int [chuck]: 16-bit int [chuck]: supported sample rates: [chuck]: 8000 Hz [chuck]: 96000 Hz [chuck]: [chuck]: ------( chuck -- dac2 )--------------- [chuck]: device name = "Vinyl AC'97 Audio (WAVE)" [chuck]: probe [success] ... [chuck]: # output channels = 2 [chuck]: # input channels = 2 [chuck]: # duplex Channels = 2 [chuck]: default device = NO [chuck]: natively supported data formats: [chuck]: 8-bit int [chuck]: 16-bit int [chuck]: supported sample rates: [chuck]: 11025 Hz [chuck]: 22050 Hz [chuck]: 44100 Hz [chuck]: [chuck]: ------( chuck -- dac3 )--------------- [chuck]: device name = "Primary Sound Capture Driver" [chuck]: probe [success] ... [chuck]: # output channels = 0 [chuck]: # input channels = 2 [chuck]: # duplex Channels = 0 [chuck]: default device = YES [chuck]: natively supported data formats: [chuck]: 8-bit int [chuck]: 16-bit int [chuck]: supported sample rates: [chuck]: 11025 Hz [chuck]: 22050 Hz [chuck]: 44100 Hz [chuck]: [chuck]: ------( chuck -- dac4 )--------------- [chuck]: device name = "Vinyl AC'97 Audio (WAVE)" [chuck]: probe [success] ... [chuck]: # output channels = 2 [chuck]: # input channels = 2 [chuck]: # duplex Channels = 2 [chuck]: default device = NO [chuck]: natively supported data formats: [chuck]: 8-bit int [chuck]: 16-bit int [chuck]: supported sample rates: [chuck]: 11025 Hz [chuck]: 22050 Hz [chuck]: 44100 Hz [chuck]: [chuck](via RtMidi): RtMidiIn::initialize: no MIDI input devices currently available. [chuck]: ------( chuck -- 0 MIDI inputs )------ [chuck]: [chuck]: ------( chuck -- 1 MIDI outputs )----- [chuck]: [0] : "Microsoft GS Wavetable SW Synth" [chuck]: =========================================================== Launching chuck doesn't fare much better. I tried using each different dac, in turn, but same results every time. =========================================================== $ chuck --dac0 -o6 chucksrc/inc.ck [chuck]: (via rtaudio): no devices found for given stream parameters: ... RtApiDs: requested channels (6) > than supported (2) by device (Primary Sound Driver). ... RtApiDs: requested channels (6) > than supported (2) by device (Vinyl AC'97 Audio (WAVE)). ... RtApiDs: requested channels (6) > than supported (0) by device (Primary Sound Capture Driver). ... RtApiDs: requested channels (6) > than supported (2) by device (Vinyl AC'97 Audio (WAVE)). [chuck]: cannot initialize audio device (try using --silent/-s) =========================================================== Does anyone have any insight? Thanks, James

3 3

report on my first livecoding gig
by Kassen 05 Jun '07

05 Jun '07

Well, that went well. It went well because there were lots of nice people, the competitors were friendly amongst eachother as well. It went well because nobody had seen ChucK and the mini in action before and now people were interested. In the last place it went well because I won, that's still nice, even if competition is less fun then duets, I think. I was up against two MAX-MSP users who were very talented. They should be because they won previous rounds, this was sort of the finale of the comptition but due to the format that was about three people I was thrown in as a surprise challenger. It was a bit sad for my second oponent who was clearly handicaped by having had to borrow a laptop and was thrown off by the different cursor behaviour, that's probably a huge factor in a graphical system like MAX. I was using the Mini with a external keyboard. The audience was just used to MAX which had been used by all (I think) contenders in all previous rounds and at first they gigled at my batered old keyboard and non-pro typing speed but with chucK a few lines go a long way. One fun moment was showing off "if(maybe)....." which -as I expected went over very well as it realy is quite funny. Because of the 6 minute format and my typing speed I tried to go for eficiency so I didn't dare to spend time on something like "while(float rubber_ducky) {//loop goes here}" which would've been even more funny (try it, it works). For the first round I used the Shakers as something like "Std.rand2(0,22) => shakers.preset;" seemed like the most economical way of quickly generating a nice rithmic yet changing foundation. Then I added a algorithmic bassline based partially on modulating the amount of harmonics in Blit instead of a filter. At one point, due to a misplaced "greater then" which should've been smaller sign my algorithem spun out of controll towards the high end but I was able to remidy this in time and it turned into a pleasant surprise that sounded good This round was a very narrow win, I thought my oponent Luc performed a very strong set of paterned noise and the audience agreed, aplauding at length for both of us, creating a need for a tie-breaking hand-vote-count. I the second round I improvised based on sending Impulse into ResonZ and then treating the result by sampling the dac using LiSa. This didn't go completely perfect but I think the audience picked up that I was considdering all my parameters on the fly and apreceated this. Overall it was a great experience and a huge thrill. I'd definately recomend it to those on the fense. Clearly you'd need to practice but that's only good for your ChucKing skills. I look forward to "defending my title" next season already. Yours, Kas.

7 13

Code that DOES modulate filter cutoff with ADSR
by Loscha 03 Jun '07

03 Jun '07

My answer is kind of a hack, but, it's working quite well. Use a Step waveform set to 1 at all times, have THAT 1 divided by the ADSR for the filter, then, my above code works like a charm! You could also, for example, spork off a different function that randomly triggered the filter's ADSR independently from the amplitude's ADSR, giving you all sorts of trance techno type patterns. Hope this helps you out, Josh, and other users. -Edward // Filter Wooby Wooby Woo BlitSaw bs; LPF lpf; ADSR amp; ADSR filt; Gain fg; amp.set( 1::ms, 350::ms, .2, 500::ms ); filt.set( 15::ms, 550::ms, 0, 400::ms ); Step s; s => filt; // 1 => s.value; 1 => s.next; bs => amp => lpf => dac; filt => fg => blackhole; 10 => fg.gain; 500 => lpf.freq; 2 => lpf.Q; spork ~ filteradsr (); for (int l; l < 32; l++) { <<<"L;",l,"">>>; spork ~ playanote ( Std.rand2(30,60) ); 800::ms => now; } fun void playanote (int nnum) { int mstodie; <<<"\n\n",nnum,"">>>; nnum => Std.mtof => bs.freq; <<<"On","">>>; amp.keyOn(); filt.keyOn(); Std.rand2(150,300) => mstodie; mstodie * 1::ms => now; <<<"Off","">>>; amp.keyOff(); filt.keyOff(); } fun void filteradsr () { while (true) { float currfilt; filt.last() * 800 + 250 => currfilt; currfilt => lpf.freq; //if (maybe) <<<filt.last(),currfilt>>>; 1::ms => now; } }

1 0