I’ve been learning my way around iClone and managed to create my first video! It’s not quite the finished result yet, but it’s close enough that I shared it with my brother. He, however, created a video on the same subject using Gemini. The result was remarkably good, and it made me stop and think. I’ve invested quite a bit of time and effort into learning how to create an iClone video.
So here’s my dilemma: if AI can produce a better result in a fraction of the time and with a fraction of the effort, what is the point of doing it all manually? What’s the point of programs like iClone?
I’d very much like to hear your thoughts, fellow iClone users who have invested their own time and effort in creating animations.
Have you faced the same question yourself? Has AI changed the way you approach your projects?
Answers will vary for and against from person to person depending on many factors from their skillset, how deep they want their projects to be, cost, quality control, consistency to name a few.
Me personally, im at a crossroads… I learnet to use iclone properly around the time of the pandemic… i got very good at animating characters… i’ve even made a whole city for scenes in my 3d film production but where im stuck is now where its time to render…
What style do i use… i’ve used Nvidia Omniverse and im happy with the way it looks… (still some learning to control all the render settings)
I can use that style but my dilemma is with AI being able to take a still image from my scene… it can animate a character to walk and move.
What do i do? just render a still frame high quality then generate everythhing from the perfectly rendered stills…?
or keep doing what im doing… or a mixture of both.
With AI you dont always have full control…
Im worried if i render 1 frame taking 2 hours to render just 1 perfect frame… if i use that to AI generate… the quality will be too perfect and i will always be forced to use AI to continure the visual consistency.
Thats my dilemma.
I’ve mentally settled the idea to continue doing what im doing and create in iclone, render in omniverse… and only use AI for short fillers im unable to animate myself manually.
I have been using AI for almost a year and it is not an EZ button. Yes it is convenient, but to get the AI do exactly what you want can be a challenge. Try to create a simple scene of a young woman sitting at a bar. She orders a glass of wine from the bartender, takes a sip and places the glass back on the counter.
Having the woman drink the wine works well with nice physics for the liquid. However, the bartender serving the wine is something else. The results are hilarious sometimes, with in one case the bartender taking a sip of the wine before giving it to the woman.
The examples you often see are action scenes with a lot of movement so the details get blurred anyway. But more intimate scenes are harder as is consictency of look and environment.
I’m using a hybrid approach where I create the environment for the scene in iClone and the character in Character Creator. With AI Studio it is then easy to integrate that with AI generation.
There is another aspect and that is that image references in a prompt give much better results then using text prompts alone. So here again, iClone can be useful for creating those references. And sometimes using a video reference for a complecated animation would be the way to go.
In sum, your learning iClone is not a waste of time in my opinion.
A while ago I started a thread about my AI experiences, which may be of interest:
I spend hours in iClone and Character Creator, and I’m fine with that. I like the viewport. I like turning a model around, nudging a pose, dragging a light across and watching the shadow move on a face. Sorta the same reason I love photography.
For me it’s like knitting, or working a crossword, or putting a jigsaw puzzle together. You lose track of time. It’s working with your hands, taking an idea out of your head and putting it somewhere you can see it. That’s a big part of why I do this.
For me the question is what the hours are for. If all I wanted was a finished clip, AI would win most days. The hours are where I make the choices: where the light falls, how the head turns, when the character takes a breath. With a prompt I get a result, but I didn’t make those choices.
I use AI and I’m glad it’s here. It’s quick, and I use it for tests and moods. It doesn’t replace the part I enjoy.
So yes, for me the hours are worth it. Even if AI could do all of it, I’d still open the viewport.
in both music and visuals - i find myself taking a step back, and refocusing on what makes a good song or image - is it telling the story i want to tell, from this perspectives, AI and Traditional animation techniques are just tools …using a photo here and some mixed media paint there for the purpose of telling a better story. A better visual doesn’t improve a bad story, but a good story survives being presented as a stick figure animation.
