Suppose that in the near future, animation technology reached such an extent of hyper-realism that it appeared nearly the actual same as real-life human actors, and movie studios began using this Hyper-Realistic Animation (let’s call it HRA, assuming that’s not actually in fact an industrial term already) rather than human actors.
Such HRA would offer some advantages:
No human lives would be put at risk in acting (in case dangerous stunts are required.) You wouldn’t have to overcome any human qualms. The animated characters could be portrayed eating cockroaches or sticking their hands in a bonfire or jumping off a cliff and you, the director, would have no liability issues to worry about. Etc. etc. They wouldn’t demand $25 million salaries like Johnny Depp.
But when I brought up this possibility in conversation to someone, she told me she would not ***like ***the idea of human actors being replaced by animation, no matter how realistic the animation was/is. In her view, human acting - for all its flaws - is much preferable to computer animation.
What say ye, Dopers? We as an audience seem to be perfectly happy with animation (such as Pixar) as long as it is its own separate niche, but if HRA actually *replaced *human actors in the industry, to the point where there were no more human actors and actresses, would you consider it a pro or a con?
Overall I would still prefer live actors, but there would definitely be some advantages to HRA, movies that are effects driven would be much easier to produce and would hopefully be better for it
E.G. imagine a remake of Starship Troopers with the mobile armour suits as described in the novel
Anything you can do with HRA you can do with actual human beings plus CGI effects. That’s been demonstrated plenty of times in recent years, so the argument about not putting human beings in danger pretty much goes out the window. The decision of whether to use HRA vs. Human Actors plus CGI vs. Human Actors plus older effects technology comes down to artistic choice on the part of the director.
And sometimes you just want to have that feel of unreality that the HRA (or even not HR animation) gives you. I don’t think A Scanner Darkly would have worked any other way. I myself liked the animation in Beowulf, although I know that it put other viewers off. Going back even further, the 1958 film The Fabulous World of Jules Verne (Vynález zkázy in its original Czech) was deliberately filmed to look like the original engraved illustrations from Verne’s books, and it worked pretty well at conveying a particular sense and mood.
I’m impressed with the animation style of the Amazon Prime Video series, Undone. It takes rotoscope animation to a new level. There’s no uncanny valley effect, but the facial emotions are rendered hyper-realistic and spot-on—the perfect blend for this series. In particular, the anguish of Rosa Salazar’s character, Alma Winograd-Diaz as she plunges into psychosis is riveting (or, is it something other than psychosis?).
Live-action or hyper-realistic animation would not have worked nearly as well for this series.
Undone is a thought-provoking, mind-bending story. I’m hoping the last episode cliff-hanger means there will be a second season and beyond.
There are two different things that go into making CGI humans (or other creatures). On the one hand, there’s the modeling, and on the other, there’s the animation. Even if you make a model detailed enough that it includes all of the facial muscles needed for smiling or scowling or whatever, you still need to decide when to smile and to scowl and to do all of the other things that human faces do. And the easiest way to do that is to have a real human doing all of those things, and set it up so your modeled character mimics the actions of the real human exactly.
For instance, in Avengers: Endgame, where we see an aged Captain America, what we’re seeing on screen is not light reflecting off of Chris Evan’s skin and captured by a camera’s sensor-- What we’re seeing is a computer simulation of what light would look like, reflected off of a simulated face similar to what Chris Evan’s face is expected to look like decades from now. But that simulation is moving in the same way that Chris Evan’s real face was moving when that scene was shot, so we still say that it’s Captain America as played by Chris Evans.
Maybe someday computers will be advanced enough to do this entirely on their own, but when that happens, it won’t be a question of computers vs. people, because the computers themselves will be people: To get it right, the computer would have to understand the full range of human emotional response.
Currently, any computer animation is going to be mo cap (with animator tweaks) or just solely done by animators. Both can be done in a close to photorealistic way. In the latter, the animators are essentially fulfilling the role of an actor. Then there’s the voice acting element. I don’t think you’ll ever be able to remove the human element, certainly not in our lifetimes.
I think it has it’s place (e.g. video game cutscenes, bringing an actor from the dead), but I generally prefer a real human being if possible.
I think it is generally better for children’s stories, when animated to be more stylized and not too realistic.
Right, thanks for the feedback from all thus far. For clarification, I’m not asking about the technical aspects of animation, but just viewer preference - if audiences are likely to feel that there is something “hollow” or missing when there’s no actual human on screen, only a CGI figure.
At some level, there’s never an actual human on screen. What’s on the screen is a pattern of light that looks like a human. It used to be, that pattern of light was usually produced by shining a beam through a piece of plastic with a pattern of pigments on it, whereas nowadays, it’s almost always done by shining a beam through a CCD controlled by a computer interpreting some digital data. The question then is just how that pattern of pigments or set of digital data is produced.
Beyond that, people are rarely STILL, even when we’re not talking or performing some specific action- we twitch, we breathe, and we move in weird little ways that it’s probably really hard to accurately model for a CGI person. I suspect that sort of thing is a big part of the uncanny valley. So like you said, old Cap looked the way he did because of the CGI, but he moved like he did because we had Chris Evans doing the actual moving.
Eventually it’ll be possible to have a computer scan enough video of people moving, standing, etc… for it to be able to gin up a convincing simulation, but for the moment, I don’t think it’s quite there yet.
I don’t think the OP himself is fully aware of the distinction. Was Old Captain America a filmed live actor, or a character created via animation? The simple answer is, he was both.
Well, all I can tell you is that CGI people piss my Millennial film-maker son off. He called me after seeing the last Star Wars. I’d liked it, and he now thinks I’m an idiot because “my god, it had that moronic CGI Luke ‘n’ Leia!” – the jedi training sequence. He went on for ten minutes, saying “Look, if you’ve written yourself into a corner where you have to CGI people, just go back and rewrite it!”
I said that CGI Peter Cushing/Moff Tarkin and CGI Tom Hanks in Polar Express (poster boy for the uncanny valley) were much worse, and he agreed.
Those things really bother me, too. I still can’t think about Lion King without seeing the computer wildebeests, or Toby McGuire’s Spidey without the Rag Doll Physics of “him” swinging around.
Doesn’t SOMEone look at the footage before it’s seen by the public? How stupid do they think their viewers are? I’ve seen scenes in a lot of movies, esp. Marvel and DC, where four seconds of Rag Doll Physics (or another CGI misstep) pull me all the way out of the movie. And often make me angry (apparently for years… I didn’t know I was still pissed off about Obviously-Not-Tobey On A String).
So, to stick precisely on the ‘Purely CGI’ topic, my kid and I would say if there’s ANYthing that shows it’s not real people… go get some real people.
When you watch the old Sinbad movies, do you get pissed off when you can obviously tell that the skeletons are stop-motion models matted into the scene?
I think thats a fine distinction to make/ask - Smaug in the Hobbit was Cumberbatch with motion capture tech as was Smeagol/Gollum with Andy Serkis (and a few apes/monkeys in other films), while Bolg/Azog were ‘fully’ CGI without any human behind them (outside of voice actor)
Where would you classify the differences between them? And I think thats what is amazing with how far we’ve come -
