I’m using 3.0. I’m not sure if the switch is avaliable in 2.93. With the switch you could easily just plug in two of the cameracull nodes into it, each with a different camera, and then you can switch between them quickly. Are you able to keyframe the camera, because that would be a pretty simple way to change between them. Not sure if you can though.
A blend node with multiple geo inputs would be great
What do you mean by blend?
Yes, that will be great. There’s a patch for it: https://developer.blender.org/D12199
I mean a Blendshape node
With fields we should be able to just “set position” using another geometry’s frozen position attribute
(Just two months ago this message would have been gibberish lol)
Of course you want to… well, blend from one position to the next, maybe by plugging both attributes into a lerp node. Do we have a vector lerp node ? or is it necessary to separate vector first ?
yeah probably it’s only in 3.0
unfortunately it looks like there’s no way to add a keyframe on the camera menu… need to think of something else
Hmm. Maybe there’s a way to use markers? I know that’s how you animate to change the active camera, but I can’t quite figure out how to use them to change the object so maybe it isn’t possiible. I’ll keep messing around with it and see if I can figure something out.
I was desperately looking for something like that a few weeks ago. I think it’s called a case-switch in some programming languages.
How does the position node know which geometry it’s getting the positions of?
AFAIK it “knows” the moment you plug the field into one of the “data” nodes (the ones with geometry sockets going in and out).
unfortunately there’s no way to do that in 2.93 apparently… this is the answer I got from Erindale on twitter:
No way to animate it in 2.93 unfortunately! The best thing I can suggest is to animate the camera values directly so it jumps between specific frames
This is “evolution” at it’s best… GN requires that the ‘artist’ can conceptualize mentally the ‘object’ it needs or want to create… For many artists, creation starts with a common object and then follows with modifications of its geometry, but without ‘thinking’ about the mathematical aspects of the object, only its physical shape or form… GN will be difficult for artists as the hidden parameters are not specified in term of action and position in a workflow… One can produce a displacement with many different GN nodes organized in different ways and using different parameters… This is counter intuitive to artists… Also when working in a production within a group, artists get directions and references that are visuals, not symbolic. So if one is asked to ‘make’ a landscape with scattered objects on it, it is based on story board, drawings and even physical models. Using GN to fulfil such specific goals is not evident at first… Maybe in the future, once GN will have been fully integrated, it will become common sense to work that way…
It takes a vector field as input, and that field comes from the “freeze attribute” node set to freeze the position. That’s my understanding
@yvesbodson2 this is more common than you might think, even though most modeling jobs won’t require geometry nodes (definitely not for characters for instance, or even for “hero” props), a procedural asset has great value to populate a scene in that there should be no need to create variations by hand. Procedural asset creation is just another way to go about the same task, with a different mindset indeed, but it’s good to have in your toolbelt really.
The arch-vis crowd will love this (as a lot of artificial environments make use of lots and lots of trim, which curve nodes would be perfect for).
Though we are talking about geometry nodes here, I would not be surprised if people use the new functionality for everything from pipes to trees.
What a step in development… Great, I am looking forward to work more with curves… Thanks to the devs