at last I had more time to play around with KAIVO and I fell in love with it like I did with Aalto. Mainly for its experimental sound design architecture. It is a great tool with maaaany possibilities I much appreciate.
I am coming from the piano and I am used to play fast and sometimes many notes in a row - regarding short percussive or Key-style sounds. And here I noticed a big problem when it comes to a somehow authentic reproduction of e.g. a Koto kind of instrument. Even 8 Voices seem not to be enough because I regognized when releasing the key on your keyboard the release-phases of each before triggered voice (ADSR) seems to be still an active voice until the release phase fades out completely. Same is while playing with a foot pedal.
This always results in cut off voices which sounds very unnatural. Am I doing something wrong or is KAIVO constructed like that (in that case maybe there is a workaround?).
Of course there is no problem when it comes down to crazy experimantal one-knob hit and run-wonder-patches :)
It sounds to me like having only 8 voices is the problem here. At some point a voice must be stolen for a new note to sound, and with a long sustain there's no way to do this without hearing it.
At some point, maybe when Kaivo gains multicore support, then it will probably make sense to allow more voices.
Thanks for the feedback.
You are welcome :)
Yes, it is running out of voices simply speaking and therefore a wild natural sounding instruments performance is not quite possible. But luckily that is not the main reason I will buy and use Kaivo. I simply love sound design and new ways to shape sounds over time.
But maybe you will find a solution otherwise than just adding more voices (we would need really powerful CPUs even in multichore-operation)...the problem I see is that the release phase after releasing a key on the master-keyboard is regarded by Kaivo as an active voice. Perhaps there is a way to trick Kaivo at this point? I dont´t know I am not a programmer. One could make a workaround while programming the sounds short without a long decay/release and than use a very good third-party reverberation plug-in to emulate this phase. That would result in loss of some character of Kaivo but a well programmed reverberation plug-in could hide this fact a bit.
In the release phase the granulator and resonator of the voice still need to be working to make sound, just as hard as in any phase. If another voice tries to use the resonator at a different pitch, you will hear this stealing.
The body is a bit different because it is shared by all voices. It is very much like the setup you are talking about with a reverberation plug-in.
Thank you for your support and sharing your knowledge.
I think it´s good to know how an instrument works. And I think that knowledge links the musician more deeply to it - what influences the creative output, whether it is a real instrument or a virtual one.
I am looking sooooo much forward for further improvements of Kaivo :)