k y m a • t w e a k y — the kyma collective || view the current website || February 2015 Archive

/ WebHome / CAGAsyRelease5 / Share.CAGAsyRelease5PreModuleTableDiscussions


Share Section



All Sections


Login / Register 
Change password
Forgot password?

Symbolic Sound

Kyma Forum
Eighth Nerve

TWiki Links

TWiki Formatting FAQ
What is TWiki?
%SESSION_ENDIF% twiki.org

Early reponses for CAG Asy Release 5

i'm going to download your version 4 files and give them some time...

but i can tell you now, i'm always up for more ways to chop up or split up sounds (ie: granular things, with various controls, or band-specific processes)....

-- TaylorDeupree - 26 Nov 2006

"Chop up or split up" what does that mean to you? More Granular then is already possible? Because granular for me is more of a highlevel isue. I think you can do already a lot with the prototypes that are available in Kyma. I need some more material or more extensive description to get an idea what it is your looking for. Oke...sometimes you do not know what it is your looking for.

-- ChristiaanGelauff - 28 Nov 2006

well, in a more abstract way.. i like to create rhythms out of melodies... so, for example, take a pad, or some texture and scan it/chop it/loop it/rearrange it..locked to a tempo.. and spit out some choppy, syncopated rhythm on the other side...

of course, i do this now with kyma (or reaktor, etc)...

i'm just brainstorming...

CAG: I was still thinking about what your saying, but it would need some new microsounds protos with 48 bit sample index pointers. Non of my current oscillator can handle bigger wavetables then 32768 samples. They only use a 24 bit sample index pointer. I still believe making BPM synced sample operation can very easily be done with an AnalogSequencer and just the Sample prototypes. Can you make a small example with the current prototypes of this choppy syncopated pad and zip everything and post it here somewhere. I still have no clue what it is you exactly want.

-- TaylorDeupree - 30 Nov 2006

Thanks a lot for your interesting contributions. I appreciate your efficient and straightforward basic building blocks. Therefore: Yes, please continue with it. I'm curious about a Release 5 (especially filters and envelope generators, as you suggested). By the way: does Share --> Sounds --> Continuum --> "Experimental Continuum sound using CAG Microsounds" (EV4Continuum0.9.kym, DiscussEV4Continuum) not be a sort of feedback and acknowledgement? (Further wishes from me: some basic things dealing with MIDI input and output manipulations on byte level, e.g. control streams derived from MIDI input in bytes, NRPNs, triggers derived from pattern matches of MIDI inputs/outputs ...)

-- EckardVossas - 03 Dec 2006

Oke, sorry Eckard I did not know you used my asy-code in one of your posted sounds. I will have a look at it as soon as possible. I do not own a continuum, but I have seen/touched it once.

I looked at your posted EV4Continuum0.9 sound. See Discussion. Can I conclude from this example, you would be interested in highly flexibel (stereo) LFO's (=+/-multi-segment envelopes!) and maybe another more extensive (stereo) HarmonicResonator?

I was already thinking of making a microsounds for this. For an new LFO thing I do not need to write any DSP-code I think, just another wrap-around class over my CAGConstant3. It does need some funky Small/CapyTalk programming for making the waves. For another HarmonicResonator, I could re-use the assembler code from my CAGFBDelay1 (The HarmonicResonator is a delay with negative feedback). I need to add sample interpolation and remove the non-lineair stage (could well be interesting...), the LP filters I will keep. I will need to pay some more attention to tuning the delayline, but that will be do-able I hope.

-- ChristiaanGelauff - 06 Dec 2006

Concerning highly flexibel (stereo) LFOs and multi-segment envelopes (independent of LFOs): yes, you can conclude this. I've never thought about another implementation of HarmonicResonator, because the existing one yields pleasant results to me. Generally spoken I'm an improvising musician who often tries things only with ears: therefore any interesting new (and inconvenient) sound is welcome and will be investigated in the experimentation labor of my ears and senses. (Sorry about going in-betweeen, leaving the time sequence because of the logic of answering)

-- EckardVossas - 07 Dec 2006

As far as MIDI stuff, I think Kyma could use some extension, specially some simple basic sequencing stuff, but who am I. Although it is a very interesting area I will not look into this, since MIDI stuff is not handled in DSP microsounds in Kyma. All MIDI information flow is handled more on the Small-CapyTalk level and in ISR's (I guess). These ISR's are not accessable for us. Can you explain a bit more what you exactly want to do that is not possible now. I think you can do a lot in writing scripts, but Small-CapyTalk is just not my thing. A lot of the time I spent on finding out how things work in Small-Capytalk. I am more a "C" programmer.

-- ChristiaanGelauff - 03 Dec 2006

Hi Christiaan, I have been keeping an eye on your work, sorry I havent given any feedback till now but I don't get much time to use my Capybara these days. However after I saw your comments I decided this weekend to make it a mission to play with some of your distortion modules. I have been wanting to make my Kyma sound dirtier for a while and the standrad waveshapers just didnt fit into my sound. But your distotion 5 module is nice, my kick drums sound suitably trashy now :) thanks!

I particularly like the bit reduction sounds, I plan to use them to create a horrible distortion on my next synth patch. Next time I have a Kyma session I also plan to try and use some of your oscillator stuff to create some new synth modules. I have also created the demo 4 song, it is a good illustration of your programming work. I have been using the demo timelines to see how you use the modules yourself.

As for suggestions I will have to spend more time with the modules, but after this first extensive play I would like to see a Stereo Dynamic Range module to connect into your stereo distortions.

I would like to know if you would be willing to integrate your code into Peter and Davids User Microsounds? Really I'm hoping that symbolicsound will allow us to merge .asy files on the fly soon (i.e. just load as many .asy files as you like) but in the meantime I think there has already been one "merge" of .asy code. Maybe we could send them a message via the wiki? CAG: Kyma X.42 (6.42) can just do that!

In the meantime please keep up the good work, it's great to have some elements for Kyma which trash things a little!

-- DavidMoss - 04 Dec 2006

If you like the distortion 5, I suggest you play a little with AddHardLimit from Release1. In fact it can dynamically change the softclipping range at the top (Perfect 909 bdrum limiting). You should input "your sound to be distorted" two or more times! If you put Distortion 1(or 2 or 4) before the bitreduction and the same distortion after with the negated setting you can make non-lineair bitreduction. Also adding the Aliaser at the end give dirtiness.

Stereo Dynamic module = as an stereo compressor(=difficult to make) or a way to get the same loudness output when changing distortion (I was thinking about this today)?

-- ChristiaanGelauff - 05 Dec 2006

Hi, thanks for the distortion tips, I will try them out over the next few days. If I come up with anything interesting I will post some examples (this will take time unfortunately).

I finally realised some modules I think it would be great to see in Kyma. I'm not much of a programmer but I understand how difficult some of these ideas may be. However I want to describe them just incase you can think of a way to implement them (and because you asked for suggestions!).

1) I would really like some switch modules, the simplest would be an on/off that I can patch a signal into and control from the VCS. I sometimes try to achieve this with expressions but I more of a visual programmer, I like to patch things together rather than write small/capytalk code. I imagine this wouldnt be so difficult to code, but would it be better as a sound rather than ASY code?

CAG: okay, your have another level of difficulties. Being a teacher I do not mind this at all, but I think this kind of questions are more for Carla or Kurt to resolve. From this I get the believe, you do not really grasp the functionality of Kyma because of the expressions. That is a pitty, because hard ON/OFF is pinuts. Although a real smooth ON/OFF is more difficult. I can help you if you like, but many more can!

CAG2: I slept on it for two nights...I will make you a nice UnMuteMixer. It will be able to do nice smooth mutes for every input and I will make an example for you with buttons in the VCS. Since we must make it a bit more interesting I thought of syncing the ON/OFF or mutes to a BPM related grid. This could be usefull for live performances and well be appealling to TaylorDeupree for the simple way of cutting thing in sync with BPM.

2) As well as a just switching a sound on or off I would really like an easy way to select between a few signal sources. For example recently I wanted to have 3 lfos feeding into an oscillator and to be able to use a radio button to select which lfo was connected to frequency, for example.

CAG: A solution could be => copy a mixer from the prototype bar into the frequency field. You should write some extra capytalk in there: ((Mixer L * 12)+ !KeyPitch)nn. The Mixer is what you just pasted into the frequency field. Now put all LFO sounds into that mixer. Inside the blue Scale field of your LFOs you write: !LFOSelect truncated eq: 1 for LFO 1 and !LFOSelect truncated eq: 2 for LFO 2 and ... 3 for LFO 3. Start the sound and change the range for !LFOSelect to min 1 max 3 step 1. You will not have a radio button or list box, but a slider with the numbers 1,2 or 3 that selects the LFO......I think C&K should make some simpler LFO expressions available for you. Did you already try the "!Trigger repeatingTraingle: periode ms" for LFO?

3) I have been dreaming of something beyond this simple switching......... a patch matrix similar to that found in the EMS VCS 3 or the Vostok synthesisers. The way this works is that the signals go in to one axis of the matrix and the outs go along the other axis. Then by putting in a patch pin you can connect an in to an out. What I really like about this is that you can connect multiple pins to route a signal to multiple destinations. I'm pretty sure you think I'm crazy now, but imagine being able to take even 4 control inputs/outputs and using the VCS quickly experiment with different interconnections in realtime.

CAG: Your are not crazy at all! That would be fantastic to have, but...there are 2 buts: 1) You are looking for a better graphical user interface, I can not provide this. 2) The whole basis of the sounds in Kyma is: you have multiple inputs and only one stereo output(exception quad output mixer).

CAG continue: Since I also consider this a restriction I was thinking of a way to send more audio signals from one sound to another or even multiple others. It could work as follows: I have a new sound called CAGBroadcast. This sound can broadcast his input signal on a frequency (example: 1,2,3,4,...,16 like MIDI channels). All sounds that I like to broadcast I end with such a CAGBroadcast. The CAGBroadcast sounds are input to a CAGBroadcastCollector. This collector is input into a mixer. The connection is just a dummy connection (no audio). The other inputs to this mixer at the end have a CAGReceiver which is tuned to one or more frequencies. Well the Idea is there, but I have no Idea if this is processor hungry or even possible to make. I forsee some major pitfalls in multivoices and heavy duty processing after an CAGReceiver. These prototypes should work a little as the FeedbackLoopInput/Output prototypes. The purpose for my would not be matrixing what you want, but making multiple feedbacks in a simple way. Let's keep this idea for a release in the future.

CAG continue2: When reading your comments again I wonder if you are not mixing up the meaning of event values (=control values running at a max rate of 1kHz) and real audio. All blocks that are connected in a sound, pass an audiostream, unless you pasted the sound into a blue field (=event value field). All things that are defined with a "!" before a name in a blue field can be used as many times as needed. Also you can add as many as you like.

Well I'm just throwing in some ideas to make sure you don't think that no one is interested in your work :) CAG: THANKS I needed this! I also realised in one of your posts that you wanted someone to check your English. I'm an English teacher so if have any doubts let me say I haven't noticed any big mistakes. If you want me to check something specific related to your code documentation please send me a message....

-- DavidMoss - 06 Dec 2006

-- ChristiaanGelauff copied the discussion here on 12 Mar 2007

© 2003-2014 by the contributing authors. / You are TWikiGuest