November 2005 - chuck-users - lists.cs.princeton.edu

performance of chuck
by Atte André Jensen 30 Nov '05

30 Nov '05

Hi I've made a working version of the midi setup I've been rambling about. Now that I make som reak music with the setup I've made a few new observations. Most problematic is the fact that chuck often (well not often, but sometimes) click and pops, esp at noteons, but also other places. I guess I have to see if I can reduce the code run at noteon time, right? But besides that is there any general approaches for making chuck run more smoothly? Just as anoying is the fact that if I punish chuck with too many notes she breaks up the audio. That's just the way it is and I can live with that. But the problem is that after such an incident timing seems to be very sloppy. I could post some examples if someone care to look at them. I'm sure my code can be optimized quite a bit. But I guess the above is still valid since optimized code will behave similar if pressed hard enough. As a final remark at least the following code is totally un-elegant (I have a feeling Chorus is not the cheapest UGen around), but I can't figure out how to envelopes on a stereo signal (and Chorus is mono anyways, or?): sinosc s1 => ADSR env => pan2 pan; sinosc s2 => Chorus chor_l => ADSR env_l => pan.left; sinosc s3 => Chorus chor_r => ADSR env_r => pan.right; pan => dac NB: Still under linux/debian/unstable with realtime kernel from demudi. NB2: With csound I was able to enhance performance by running in a virtual terminal without X. But with chuck, that doesn't semm to make much difference. Neither does running as root. Neither does running with nice -19 and killing a few daemons.... -- peace, love & harmony Atte http://www.atte.dk

5 10

stereo suggestion
by Atte André Jensen 30 Nov '05

30 Nov '05

Hi I mentioned this already in the middle of something else, but recieved no reply, so I thought I'd start a new thread. It would make sense to have all UGens stereo aware. This might include supplying all with a .pan, .left and .right parameter, but at least it should be possibe to route a stereo signal through them and get a stereo signal out. Would that be expensive, both in execution time and in coding time for the developers? -- peace, love & harmony Atte http://www.atte.dk

2 1

Power operator in chuck
by Tennessee Leeuwenburg 30 Nov '05

30 Nov '05

Hi all, I just downloaded chuck. Perhaps I'm being a bit blind/dense, but there doesn't seem to be a simple power operator in chuck. Given that power-laws appear in many places, it would be a very useful operator. Hopefully I'm just missing it? Cheers, -T

3 3

Re: [chuck-users] pitching samples
by Ollie Glass 29 Nov '05

29 Nov '05

Brilliant, this is exactly what I'm looking for. Thank you Atte. Ollie > Ollie Glaskovik wrote: > >> I'd like to make a pitched sample playing function which, given a >> sample at a known pitch (say C-2) and a MIDI pitch value (0-127), >> plays the sample at that pitch. I suppose this could be done with >> sndbuf's .rate function and some math but I'm not sure where to >> start. > > Given you have the midi pitch in the variable "note", the following > will work. "transpose" is just a handy way of transposing the > instrument, "root" is the known pitch of the sample. Note also that > mtof will work with float, so you can have 12.01 => transpose, to > transpose an octave + a cent. > > 12 => transpose; > 60 => root; > > sndbuf s => dac; > > std.mtof(note + transpose)/ std.mtof(root) => s.rate; > > -- > peace, love & harmony > Atte > > http://www.atte.dk

1 0

which language resembles chuck the most
by Atte André Jensen 28 Nov '05

28 Nov '05

Hi I normally use ctags for finding my way through my own code. For those un-aware of ctags, its a utility that generates a TAGS file containing all functions and variables in a project. Armed with that emacs (or vi?) can quickly jump to the place where a certain definition is made, across files and directories. Now, ctags clearly needs to understand the structure of the code, and unfortunately (but understandably) chuck is not on the list of supported languages. But parsing in a syntactically similar language is often enough. My question is: which of the following languages resembles chuck the most, syntacticlly speaking (my guess is Java): [atte@aarhus midi_input_07]$ etags --list-languages Asm Asp Awk BETA C C++ C# Cobol Eiffel Erlang Fortran HTML Java JavaScript Lisp Lua Make Pascal Perl PHP Python REXX Ruby Scheme Sh SLang SML SQL Tcl Vera Verilog Vim YACC -- peace, love & harmony Atte http://www.atte.dk

2 2

pitching samples
by Ollie Glaskovik 28 Nov '05

28 Nov '05

I'd like to make a pitched sample playing function which, given a sample at a known pitch (say C-2) and a MIDI pitch value (0-127), plays the sample at that pitch. I suppose this could be done with sndbuf's .rate function and some math but I'm not sure where to start. Any suggestions would be appreciated. Thanks, Ollie

2 1

sndbuf vs WaveLoop
by Atte André Jensen 28 Nov '05

28 Nov '05

Hi I tried both sndbuf and WaveLoop for playing and there seems to be a problem with both: sndbuf sounds like 8 bit, maybe the interpolating is broken? WaveLoop sounds fine, but is out of tune, something like 250 cents (from memory). This is all under debian/linux/unstable with the latest chuck... -- peace, love & harmony Atte http://www.atte.dk

2 1

Re: [chuck-users] Inter-shred communication
by Ollie Glass 28 Nov '05

28 Nov '05

> Hi Ollie, > > All is good except that you probably want to make a static TheEvent > somewhere and use that one instance among the shreds. slightly > modified code below, with a new class 'The' to hold the global/ > static object(s). Perfect, thank you Ge. Ollie

1 0

stereo patching
by Atte André Jensen 28 Nov '05

28 Nov '05

Hi Suppose we have a stereo generating UGen, like Chorus. If I do the following: Chorus c => ADSR e => dac; everything is back to mono, since ADSR (like almost all UGens) is mono. Wouldn't it be more practical if all UGens were stereo? They should then also have .left, .right and .pan like pan2. Or am I missing something here? What's the nicest way to do the above mentioned patching while maintaining stereo? -- peace, love & harmony Atte http://www.atte.dk

1 0

Re: [chuck-users] The ChucK Manual prerelease
by Felix Meinhold 28 Nov '05

28 Nov '05

Hi Eduard, i would like to translate your fine manual to german. are you okay with releasing the latex code to the wild? -felix On 26.11.05 17:35, Adam Tindale wrote: > Hi Eduard, > > Thanks for your comments. I agree about having more thorough documentation. I > will add the changes you suggest. I don't have time to add in an example for > everything in the docs. I also don't know when things break or change. > Sometimes I hear about it and sometimes I don't. I am more than happy to make > small changes and keep the manual as up to date as possible. If you have any > small patches that demonstrate ugen parameters I would love to add them to the > docs. That goes for all you ChucKers out there. This is a community project. > Let's all help each other! > > Please let me know anything that you would like to see in the manual. I am the > documentation editor. I develop some of the docs, but most of it has been > developed by Ge, Ananya and Phil. They deserve a big round of applause. > > The docs should help you learn ChucK. If they don't then let me know. > > Thanks for your input. > > --art > > > > On 26-Nov-05, at 4:34 AM, eduard wrote: > > >Hello all, > > > >I think you have done a great job on writing the documentation, however I > >have found that it is not "completely" up to date. > >For instance, when I use impulse.value() I get an error like class 'impulse' > >has no member 'value'. > > Right, I haven't put that in yet. Will do. Thanks for pointing it out. > > >The same for things like math.pi, which now seems it's only pi. > This I am not aware of, but thanks for pointing it out. Maybe this should be > fixed so that old patches aren't broken. Ge? > > >But both of them ( impulse.value() and math.pi() ) appear in the > >documentation. These two examples are the only ones I can remember now, but I > >think I found other ones in the past, but not sure now. > > > >I would find very useful to have more explicit examples for each UGen, such > >as writing an example in which you use all (or most of) the members. > >Also, I would give an interval, when possible, for the values they can take > >and give a little explanation for each member depending on the values they > >take. i.e: > >.rate [0, inf) a rate of 0 would mean blablabla and a rate of 2 would mean > >blablablabla > > > >Explaining every member can be a bit redundant, but sometimes it is not that > >clear how you can use them and what values make sense. > > > > > >I hope this feedback is useful, > > > Extremely, thank you. > > >Thanks > > > >Eduard > > > > > > > >On 17 Nov 2005, at 17:01, Ge Wang wrote: > > > >>Hi Adam and all, > >> > >>This is great!! (It's about time we retire the poor doc/GOTO file in the > >>current distribution and include actual documentation.) > >> > >>Spectacular job to Adam and many others for creating the manual! > >> > >>We still have much work to do on documentation: tutorial series are coming > >>soon, and we need to document the emerging features/bugs and class > >>libraries. It would also be nice to aggregate existing chuck documentation > >>nodes (including Gary Williams's Beginning ChucK blog and others). Please > >>pipe up if you want to join this effort. > >> > >>Great job once again to Adam and others on the manual. Adam, when do you > >>think we can start shipping it with the distribution? > >> > >>Best, > >>Ge! > >> > >>On Mon, 14 Nov 2005, Adam Tindale wrote: > >> > >> > >>>Hi All, > >>> > >>>I have been working on the manual for almost a year. It is nearing > >>>"completion". Some of you may have seen it in cvs. You can get your copy > >>>here: > >>> > >>>http://web.uvic.ca/~art/chuck/ChucK_manual.pdf > >>> > >>>I am making a public plea for help. The manual needs your help, fellow > >>>ChucKer. What needs to be documented? Do you have ideas? Do you know what > >>>all the values of the UGens are? The manual should help you use ChucK > >>>better. I hope to put it out in the wild "soon", in the truest ChucK sense > >>>of soon... > >>> > >>>I will take patches in plain text or .diff files. I will take criticism or > >>>abuse or ideas. I am more than happy to typeset text, but I only have so > >>>much time to write up tutorials, etc. If you have an idea for a tutorial, > >>>get it started and then I can finish it off when I merge it to the docs. > >>> > >>>The manual is done with LaTeX. If you or someone you know wants to help > >>>with that part of the manual it would be greatly appreciated. > >>> > >>>Enjoy. I look forward to you comments. > >>> > >>>--art > >>> > >>>_______________________________________________ > >>>chuck-users mailing list > >>>chuck-users(a)lists.cs.princeton.edu > >>>https://lists.cs.princeton.edu/mailman/listinfo/chuck-users > >>> > >>> > >>_______________________________________________ > >>chuck-users mailing list > >>chuck-users(a)lists.cs.princeton.edu > >>https://lists.cs.princeton.edu/mailman/listinfo/chuck-users > >> > > > >_______________________________________________ > >chuck-users mailing list > >chuck-users(a)lists.cs.princeton.edu > >https://lists.cs.princeton.edu/mailman/listinfo/chuck-users > > > _______________________________________________ > chuck-users mailing list > chuck-users(a)lists.cs.princeton.edu > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users >

2 2