Pandoras Box User Forum

Before I started to work with Coolux I used different other video tools like Max/Msp/Jitter, vvvv and Processing. I started to use Pandoras Box because it's time saving and easier for workflows like mapping, edge blending, live video, HD video playback, multichannel audio routing, etc... It's much more powerful for playback and it's much more stable and it offers nearly endless possibilities. With the effects and Widget Designer a lot of interactive and some generative stuff is possible. But I'm missing a bit the possibility of generative video design because in my projects (theatre, media art) I'm the content designer as well and I like experimenting "on stage" or within a build up setting. So my question would be if and how it could be possible to implement more features for generative video design? Or would it in general be possible to implement e.g. a feature that allows to drop a Processing patch on the timeline and get a patching tab where the code can be modified? Or something similar/ something else? I know that it is something special and of course nothing for the next revision but I'm interested in how Coolux thinks about generative features or if there is a certain demand generative possibilities?

I would love to see more generative effects as well, but I am sure they are difficult to write and it is a long process.
In the meantime! You can bring in a live input with another computer using MaxMSP, Processing or whatever (I have seen this done before), or you could play with web assets and see what you can load in there. Often Processing can be loaded as webpages with a bit of cleverness. You will need to test the webpage as the browser system in PB is not the same as having a full version of Chrome or Firefox at your disposal, but I imagine it is possible.

Hi,
We are all for generativ Content and FX. We are also working on making more FX available.
In the mean time,we also already have a pretty powerfull System in place, where you can write your own shaders that generate Content.
Shader SDK Training is available on Request and gives everybody the option to write their own Effects.
The Shader Language is HLSL. There are a lot of Shaders available online for example at https://www.shadertoy.com/ and even at youtube there are Shader Tutorials.
Most of those Shaders are written in GLSL which for the most part can be translated to HLSL fairly easy.

If you are interested in Shader Development feel free to contact us at support(at)coolux.de

Daniel

I think that the HLSL plug-in architechture is the cleanest way to integrate generative media at this point. You can already combine the shape generators with U/V scroll and distort effects to produce animated generative media. If you don't feel like the FX/Tools in the current release provide many options for generative design, then the current, massive, nested folder structure of the FX library is a likely culprit. There's a demo project somewhere on here for example where I put together a tunnel demo using some basic shape gens, and fractals. But you can't do stuff like that if you don't have a clear handle on all the bits that are there, and how they work. Maybe coolux could do a generative media web series, or perhaps we can workshop some of this stuff in January.

What's nice about coding stuff for PB's HLSL compiler is that you automatically get access to the FX/UI-Control wrapper. So that you can create keyframes in the PB timeline to animate your effect parameters. Developing something similar for Processing would be difficult, because external parameter control there is provided by external libraries. Also, while I like Processing rapid development workflow, it's Java underpinnings mean that for the most part, you don't get to benefit as much from the main yummy part of working on a platform like PB, which is GPU based processing.


EDIT: Link to the tunnel, demo.