I have an idea for a sound that would be extremely useful to me, and I'm sure to others too, and I would like to request the assistance of the collective to help me create it. I am presently working on a sci-fi video game where the hero can be attacked by rats. These can range from just a few normal rats to perhaps a hundred giant mutated rats.
Overview::
Create a sound that is capable of playing back multiple small mono samples randomly, but with controllable density, pitch, length and timbre.
So....
I'd like to be able to have at least 10 samples (short mono files) of rat squeaks/sqeals that would play back randomly but be able to control the frequency of their occurence over time. So the control would range anywhere from a single random rat voice every 2-4 seconds up to a cacophony of constant rat voices.
Additionally:
1. the samples would need to play back with a random value of pitch shift and/or disk playback rate with a controllable range of 0 to +/-12 semitones,
2. a gentle low pass filter with a controllable cutoff, not random, (to simulate distance)
3. a random playback level determined per sample, controllable ranging from 0 to +/-10dB,
4. a random pan function per sample.
I think that's it but I'm open to ideas.
My thought is to be able to load the sound into a Timeline and be able to control all the mentioned parameters live or by automation, including of course the overall level of the sound. Then, place in the Timeline parallel with that an identical sound, but this time it has samples of scratchy feet, scurrying, etc., then another sound that has biting and chewing samples loaded up.
I'd like to end up with a sound that is not just useful for my rats, but can also be used for a herd of cattle or a swarm of insects, or even to recreate restaurant or factory ambience. In looking at existing sounds and prototypes, it appears that all the elements are there to achieve this, however I am very much trying to get my arms around how to approach it and structure it for the best result. As I am very much a relative newbie to Kyma, I would appreciate it if any responses could be presented in a "step-by-step" format if possible as I'd like to edit the results and publish them in the tutorial section of this site.
Any help is greatly appreciated, thanks.
-- RandallThomas - 04 Jan 2004
. I'd start with a SampleCloud and would set the grain enveloppe to Square, leaving
the Amp enveloppe at the sample level.
While a big rat might sound 12 nn lower than a small, I'd build a new sample for the transgenic rat. You can then parallel a few SampleCloud, each one having its Attenuator-Filter etc...
-- KarlMousseau - 05 Jan 2004
In such a case I would create a sound with the sample player, filter, panner, attenuator etc so that you can play several voices and control each voices parameters seperately. For all controllable parameters I'd use !ccOO to !ccXX. So getting one branch per voice and bundling everything in a mixer. For controlling the parameters (which is in fact your question if I understood it well) I normally use an old Archimedes computer which I program in Basic V. Now I am in the proces of transferring to a PC which I will program in VMM (virtual midi machine). VMM is a very easy and C like script language in wich you can write one algoritm for controlling all the parameters or, if you want, you can write several algoritms each controlling the parameters of one animal. All these algoritms can run at the same time because the runtime module works multitasking or multithreating if you want. I think its very complementary with Kyma. Because of the hardware only midi ports in Kyma/Capybara you'll need a midi port on your pc to connect to the capybara.
You can download VMM at http://vmm.audionetwork.be/ This is a site of former students of electroaccoustic music composition at the conservatory of Antwerp (Belgium). VMM is a free product of the professional programmer Marc Meeuws, one of them. I think It is worth giving it a try.
-- DirkVeulemans - 10 Jan 2004