Best Approach for rendering similar characters?
Hi there! We're using Rive to build emotes for our app. It's such a delight! :)
I have four emotes, Happy, Love, Mad, Sad. They're all the same base character with different things popping out of their heads.
My initial approach was to create a single art board and draw the character once. Then, I would make a timeline for each of the four characters to initialize their colors and face shapes. These four timelines would be the base layer of the state machine, and we would create facial and movement animations on other layers.
A few issues with this approach: Each of the "initialization timelines" have a ton of keys that touch pretty much each part of the character (colors, vertices for mouth shape, various opacities to show and hide gfx, etc) It feels like it gets hard to manage. On top of that, when creating timelines to animate the characters' movements, I think you'll be working with the general design that was build in Design mode, rather than the specific character that you're animating for.
So I've decided to go with a different approach. I'm making four individual art boards, one for each emote. I think this will make it easier for the animator to reason about the timelines, and it will simplify the state machine logic. However, it does increase the file size, and in some ways I have to four times the work. Like, if I need to change the base shape of the character, then I have to do it on all four art boards, and that makes the programmer in me sad.
Does that all make sense? Am I heading in the right direction? Are there any workflows that I'm missing?
Hmm, the more I think about it, I think a lot depends on how we plan to animate these characters. I realized today that I don't have a lot of clarity on that, yet. So really I guess it's a matter of rigging. And I could probably break down the characters into more modular parts, like separate the face from the body.
Hey Rob! I think you're right it might depends on the rigging. But The approach of having the parts separated is a good idea. I was thinking maybe you can use solos for changing the faces and nested inputs in order to change the colors and the elements on top of the head. I put together this little test that might help you. I'm sharing a video showing a bit the project file. Let me know if any of this makes sense. If you're already familiar with solos and nested inputs ignore this last part, but just in case I'm sharing two useful videos:
Thank so much Lala. That all makes a lot of sense. We're going to try an approach like this. Really appreciate you taking the time to reply!