Overlapping grains (buffers?) in polybuffer granulator

James Kogane's icon

Hi--Have a basic polybuffer granulator here and controlling which samples are chosen by a multi-slider but haven't been able to implement a successful method for overlapping the grains, which in this case I guess are buffers? Plan to implement volume control per grain/buffer from the multi-slider next but the overlapping is the thing that's getting me.

Thanks

Max Patch
Copy patch and select New From Clipboard in Max.

Sébastien Gay's icon

Hi James,

I am not sure to understand what the multi-slider is there for : randomly choose a subset of the sound files stored in the poly buffer ? But why a range of values for each individual slider in this case ?

As far as "overlapping" is concerned, if you mean playing several grains at the same time so that triggering a new grain does not stop playing the previous one, then you should use poly~ or even better mc.groove to get each grain assigned a specific voice/channel. If my reading is correct, and if not already done, it may be a good idea for you to watch FEDERICO-AMAZINGMAXSTUFF's wonderful video series on this topic (https://www.youtube.com/watch?v=YnROTQAQW_I). It made a lot of concepts clear to me when I was trying to understand better granular synthesis and mc.

James Kogane's icon

yeah planning on making it on or off instead of a range for the multisliders but know how to do that. goal is to be manipulating folders rather than a single buffer without using flucoma. There's a well known patch in supercollider called "mulitgranulator" which granulates a folder of samples and controls them based on weights in a multislider, so trying to recreate that here. I guess mc would be better.

Sébastien Gay's icon

So each sample/sound file is granulated individually with it's own parameters and the "relative density" is controlled by the slider, right ? But why randomization, then ? Sounds like ruining the efforts made to select the weights individually ?
Interesting project, though. Thanks for sharing the idea. And yes, from what I understand, you definitely need to go mc.

James Kogane's icon

i may be doing it wrong but idea is to pull out certain samples and then have them "granulated" in a random series rather than linearly

James Kogane's icon

figured it out will post soon

Arne Herrmann's icon

Hey James!

Is there any update on your process yet?
I'm thinking of a similiar process with recording into a polybuffer and using mc to move the samples through different outputs individually - but I'm still hanging at how to design the polybuffer - mc.groove connection - so I'd be interested in your approach!

Best!