Code blocks

korhanerel

I am not literate in C++ and therefore am stuck with using Pure Data (which is not necessarily a bad thing). I was wondering if there will be a repository where users will be able to get code blocks which they can combine to create larger projects, assuming this idea makes any sense in C++.

Thanks

bela_robert

Hi korhanerel, mmm interesting idea of code blocks. I suppose the example projects that we have provided can act in a way like code blocks, but to piece them together you do need a bit of c++ experience. They're more like demonstrations of different techniques than pieces of code that could just click together, and this is kind of how you have to work in c++. We are really keen to grow the number of examples and build Bela into a great resource for learning. If you think of any specific examples that are missing, the forum is a good place to say - we want this to be a place of sharing projects and learning from other users.

The PureData examples should also show you how to do all the simple stuff and move on to introduce you to more complete and complex projects. With PureData you can do almost everything that you can do in c++ (just with a slightly different logic) so it shouldn't limit at all what you can build.

korhanerel

bela_robert Hi Robert. Thank you for the quick reply. I am a musician who tries to strike a careful balance between engineering and music-making (in my own personal view of things), and therefore do not want to get involved in coding that much. It's good to know that Pure Data is as efficient as C++ with Bela, at least in terms of CPU capacity - I guess it's because it is compiled as C++ code, right?

What I would like to see is code snippets (or pure data subpatches) optimized for use with Bela. My personal interest lies in sampling, creative sample-playback, granular synthesis and MIDI control of all that, the latter being very important for me as I am a computer performer who needs to react fast (free improvisation) and therefore uses a lot of controllers.

Thank you!
Korhan

andrew

Using Pd will be about as efficient as C++ if you compile it using Heavy. If you run it using libpd (which is what the browser IDE supports) then the CPU performance won't be as efficient. However libpd supports more Pd objects than Heavy at this point.

For Bela-specific Pd, have a look at how the analog and digital I/O is handled in the examples. In many cases with Bela you can treat everything, including sensor data, as an extra audio channel.

bela_robert

Hi Korhan, I've added an example to the repo for sample manipulation with Pd that should help get you started dealing with samples - it's called sample-scrub in the 08-PureData examples. You'll need to update the board to see the example, by following the instructions here: https://github.com/BelaPlatform/Bela/wiki/Updating%20Bela

This is just a simple example of loading samples and then scrubbing through them using potentiometers. Could be a good basis for granular manipulation.

korhanerel

Perfect! Thank you Robert! I'll upload it in the next few days.

I do know how to playback samples in Pure Data, as a matter of fact, but was not quite sure if my PD Vanilla knowledge was compatible with what Bela expects.

giuliomoro

late reply: if you are using libpd, then you can freely user [readsf~] and [soundfiler]