As one of Vuo's developers, I work on Vuo's engine (the thing that makes compositions run), work on nodes, and write documentation. You'll see me on the forums answering people's questions about Vuo.
I enjoy using Vuo to make live music visuals. My hope for Vuo is that it will grow into a community of people of diverse backgrounds and identities making lots of different artistic, useful, unique, goofy, beautiful, crafty, wonderful compositions.
Estoy practicando para ser competente en español. Si publicas en el foro en español, trataré de responder en la misma manera.
If you look at the port popover for the output of Combine Layers, you can see that the number of descendants within the layer's hierarchical structure is continually growing. Given the Cut List node, I'm guessing this isn't what you intended.
The first time Combine Layers executes, it outputs a structure consisting of a parent and a child:
The second time, now that it has input from Hold Value, it outputs a structure like this:
Each iteration after that adds another level of depth. I think instead what you want from that second execution is this:
/ | \
x x x
The attached composition shows one way to do that — except I cheated and removed the Transform Layer that was in the original composition because it made things more complicated :) I imagine you can figure out how best to handle that within the context of your actual composition. If not, let me know.
All of that said — The composition shouldn't crash. My coworkers and I will discuss how to avoid crashing in the case of an unboundedly growing layer/scene object. (Can we change the nodes to prevent the problem from arising in the first place? If not, can we detect the problem and display an error message?)
Seems like the text layer node just doesn't like being hundreds of characters long.
Yep. We took some time profiles, and indeed most of the CPU time is spent rendering text. We already have it set up to cache rendered text so it will skip re-rendering the same text, but since the text is constantly changing here, that doesn't really help. Tentatively, it looks like most of the rendering time is spent in Apple's CoreText.
Fortunately, for this particular composition, you could avoid rendering large blocks of text by using Blend Image with Feedback. See attached example.
Chris — If you create/input the switch imagery (including any animations) using layers/images, then yes.
Based on discussion with the team, we should probably revise the proposed node to something like this, to better support your use case:
Make Toggle Button Theme (Layers) — Layer input ports:
To be clear, this is the highly customizable DIY solution. If you want a basic out-of-the-box switch where you don't have to provide the actual switch images, then I'd recommend creating a separate feature request.