We saw Avatar the other day, and did the whole 3D, IMAX nine yards. It complicated things a little — there are many more showings of the plain-old 2D version, and for that matter of the 3D in non-IMAX theaters. But after all we’d heard about the experience, it seemed the only way to go.
My original intention with this post was just to provide a thumbnail review, along these lines:
James Cameron, damn him, has done exactly what he said he’d do: delivered a kickin’-good movie with mind-blowing special effects and cinematography. He may not be king of the world — any more than Orson Welles was in 1940 — but…
Etc., etc.
But the more I thought about it, the more I realized I was interested mostly in one thing: one facet in which the film didn’t disappoint, exactly, but also didn’t (probably couldn’t) quite succeed. Before getting into that, though, let me say:
- The 3D effects in Avatar — at least, as viewed in an IMAX theater — go way beyond the lame, unimaginative poke-the-audience-with-a-sword precursors. When little flies and moths beset the characters in the jungle, you may have to fight the impulse to try swatting the bugs away. Or, like me, you may find yourself looking over your shoulder to draw the projectionist’s attention to the need for an exterminator.
- Motion-capture technology, likewise, has leapt ahead since even the (justly) celebrated tools which Peter Jackson and Andy Serkis employed to bring Gollum to life in The Lord of the Rings — particularly in capturing facial expressions.
- Technology aside, you’ll recognize Avatar‘s plot and love story from numerous “civilized man goes native” films that came before (Dances with Wolves, anyone?)…
- Yet, you may still find yourself welling up from time to time.
- I thoroughly enjoyed every second of the film. Thoroughly. (At some moments, indeed, I felt that I may have been undercharged despite the almost $14-a-pop admission price.)
So what didn’t succeed?
The images at the top of this post are screen captures — frames grabbed from the online version of the film’s 3½-minute trailer. The one at the top shows the main character, Jake Scully, just as he’s drifted off in the “interface pod” which allows him to animate his alien avatar’s body from a distance. Below that is the face of Jakesully (the name compressed, per the captioning, into a single word), one of the title’s avatars: the creature, a member of the Navi race, whom Jake “becomes” when in the interface pod. The frames both in the trailer and in the movie are just a few seconds apart, so can be viewed almost as before-and-after pictures.
Oh, yes, there are obvious differences — blue skin, big yellow eyes, pointed ears. But look more closely.
The first thing you may notice is how remarkably real the Navi face seems. The reflectivity of the eyes, for example: highlights appear in exactly the right spots, given the overall lighting, and the irises have those little sunbursts of light-and-shadow which you can see in the eyes of terrestrial mammals. The eyebrows (and the eyelashes which rim the bottoms of the eyes) aren’t just flat digital brushstrokes, but have individual hairs. Jake Sully has a slight crease below his left eye, and so does Jakesully.
But something seems just… just off nonetheless. Even allowing for the obvious differences, something makes Jake Sully’s face different from Jakesully’s. Maybe it’s the pores and creases, both of which are relatively plentiful in the human face but smoothed out in the Navi. Whatever it is, the photograph and the animation are clearly two different media, technologies, choose your word. If you met Jakesully on the street, you would never ever mistake him for an actual living creature.
In wondering about and investigating exactly what it is which separates the subjects of those two images, I came across a bizarre-sounding term which may explain the “this one is real, this one isn’t” feeling: uncanny valley. In short, what we might be seeing is the intentional artificialization of reality. From Wikipedia (emphases added):
The uncanny valley hypothesis holds that when robots and other facsimiles of humans look and act almost like actual humans, it causes a response of revulsion among human observers. The “valley” in question is a dip in a proposed graph of the positivity of human reaction as a function of a robot’s lifelikeness…
[The] hypothesis states that as a robot is made more humanlike in its appearance and motion, the emotional response from a human being to the robot will become increasingly positive and empathic, until a point is reached beyond which the response quickly becomes that of strong repulsion. However, as the appearance and motion continue to become less distinguishable from a human being, the emotional response becomes positive once more and approaches human-to-human empathy levels.
This area of repulsive response aroused by a robot with appearance and motion between a “barely human” and “fully human” entity is called the uncanny valley.
…the concept of the uncanny valley is taken seriously by the film industry due to negative audience reactions to the animated baby in Pixar’s 1988 short film Tin Toy.
Never seen Tin Toy? YouTube to the rescue (the creepy baby in question makes his appearance at around 50 seconds into the film):
I don’t know — revulsion seems too strong a word to me, or maybe I’ve just got a strong stomach. But I get the point: if you want an audience to squirm, without knowing why they’re squirming, just put perfectly human faces and movements on an artificial creation which the audience knows cannot possibly be human. If you want to avoid discomfiting your film’s audience, and can’t make a completely perfect human-like creature, be sure not to try too hard: give your automatons (like the Navi) some at least faintly artificial elements.
The (perhaps intentional) artifice in the onscreen Navi extend to the characters’ movements as well. (Which may explain Cameron’s decision to film the human characters, rather than to animate them as well as the Navi. In only a few scenes do both filmed actors and animated creatures both appear, and — for me — the effect was disconcerting.) When the Navi are running or jumping, their movements are just a little too fluid, a little too “slidey.”
(In online fora, I’ve seen enthusiasts argue that, hey, this action takes place on a different world: gravity itself might be different, and the very air might be thick enough to make movement vaguely like swimming. Okay, but that doesn’t explain why the humans — in those combined film/CGI scenes on the world’s surface — seem to move just like they do on earth.)
The point is that this isn’t a flaw. It’s not, y’know, a bug, or a sign of limits to current technology. (It wasn’t until later, when thinking back on the film, that I even put my finger on what seemed false. Like I said, I was too busy enjoying myself while there in the theater.) I’m just suggesting that maybe what I thought of as a weakness is there (as Microsoft says of certain software “bugs”) by design.
If you’re curious, here’s the longer trailer for the film (a larger, higher-quality version is available by clicking on the link which follows); keep your own eyes peeled for possible attempts to work around the uncanny valley problem:
____________________
P.S. Cameron himself is on record, as it happens, worrying about the uncanny valley. See this feature at ReelzChannel.com, among others. For a negative review of the film which deals quite a bit with the uncanny-valley effect, see this scanners blog entry (including follow-ups and comments tacked on at the end) at the Chicago Sun-Times site.
(That reviewer, Jim Emerson, seems to have missed a subtle point about avoiding the uncanny valley: if you can’t create a perfect non-human human, stop trying so damned hard — because those attempts will just push the audience even further into the trough of discomfort.)
P.P.S. Whatever it means, I have to say I just love the term “uncanny valley.” I’d love to sample a ranch dressing from there.
That’s a great explanation. And I remember being creeped out by that baby in the Pixar movie when I saw it for the first time. I never knew the term, but I’m with you — I like it.
This is such perfect timing and so funny. My husband, evidently, explained to me what the Uncanny Valley was about this time last year — after we watched the film adaptation of “The Polar Express” and were rather aghast.
I was cleaning off my desk the other day and found the little graph he had jotted down to explain the entire concept to me. I approached him and said, “what the hell is THIS?” And he re-explained it to me.
I, like you, love the term itself.
And, yeah, that baby is CREEPY.
Darc: If I had to pick somebody who remembered the Tin Toy baby from his first viewing, it’d be you.
A while back, thanks to the Ally McBeal TV series, just about anybody with a computer was subjected to this, in the same vein:
Somebody even packaged it on a CD, with numerous variations, as a screen saver to enrage one’s co-workers.
Jules: Good old B — he’s got the most amazing array of… of stuff at his mental fingertips.
I think I’ve seen him “in person” on 7-Imp twice: once on his birthday, when he was just spoiling himself :); the other, in response to a question I asked about that brief little moment of silence just before a teakettle starts to shriek. Lots of people responded, actually, but we were all just sort of mutually befuddled/mystified. Not him, though. He knew EXACTLY what it was.
(And given your own lit/art geekism, coupled with B’s version in sci/tech, I picture your kids — cute as they are now — growing up to be adorable little polymaths: nodding gravely when someone older is spouting patent nonsense, but looking at each other out of the corner of their rolling eyes when the speaker’s attention wanders.)
I think you’ve got your ‘million dollar idea’ there, John – Uncanny Valley taking on and trouncing Newman’s Own. Assuming you’ve got a little bit of savvy in the kitchen, or can buy it in cheaply.
And ooooh, Sigourney Weaver as a redhead? I’m there.
My husband has been bugging me to see AVATAR and I hadn’t put much thought into it, but reading all this information has actually solidified my decision not to see it. I get vertigo in IMAX. I’m worried I’ll feel the same with 3 D. I think I’m worped out on sci fi movies from raising two boys. Interestingly, my 18 year old has no interest in it. Is it oriented to adults?? And that creepy baby isn’t creepy because of uncanny valley. It’s creepy because it’s artistically UGLY. It’s poorly rendered. It has none of the cuteness of a baby. The skin is tight, the coloring is off, the diaper looks like solid rock. It’s creepy because it’s creepy. It doesn’t need an explanation. Why did they make the toys cute but the baby creepy??? It shouldn’t have been that hard. Clearly someone not around a lot of babies.
Froog:A disconcerting moment in the film — not sure it was an uncanny-valley one — was the first glimpse of the Navi avatar who is controlled by Weaver’s character. Each avatar is “grown” from a blend of Navi DNA and that of the, er, inner human, so they resemble the appropriate human. Here’s Dr. Grace Augustine’s avatar:
Squirrel: Okay, don’t let me get caught up in an intra-family movie-going fracas. But I’ll add that (depending on where you are) you can actually see the film in any of three separate formats: standard 2D, 3D on a regular screen, or 3D on an IMAX screen. So maybe you can work something out. (Fwiw, afterwards I talked to a couple of people who normally get 3D-film motion sickness. They said they felt sorta like that at the beginning of Avatar, but the feeling passed.)
“Oriented to adults”: well, I don’t know. Arguably, it’s a “message” movie — it could be called environmentalist, New Age-y, anti-Big Business, and/or pacifistic. But I sort of feel that’s like arguing that Titanic was a plea for building safer ships. Maybe your 18-year-old is just supporting Mom?
Yeah, that baby was pretty clunky as well as creepy!
Just a quick comment. I get vertigo at regular IMAX movies…you know the ones where the screen sort of wraps around to the sides and the seats are all essentially vertically on top of the seats in front. I went to see it at an IMAX theater which had stadium seating, and the screen was just REALLY big. I was amazed how okay I was seeing it this way. I can understand not wanting to see it for lots of reasons, but if the dizziness thing is the one thing keeping you, try it on a regular screen. It was really okay.
And I love the dancing baby. I remember trying to show it to a bunch of elementary kids in their classroom, but the computer was so slow downloading it took all afternoon. I’m not so sure I love all the variations but the original was still a hoot!
And I agree with Squirrel, that baby was just plain ugly.
I love this uncanny valley idea, but I don’t think that’s what we have with Avatar or the Tin Toy baby.
The baby is supposed to be unsettling – it’s the enemy, the threat, The Destroyer. Maybe it’s also supposed to seem monstrous, unreal, from the perspective of the toys? Or maybe it was just done in a hurry. That looks like a fairly early Pixar production, and not one of their more grandiose efforts – there are lots of other things about – inconsistent shadows and highlights, the little balls seeming to hover above the floor from certain angles – that you’d be unlikely to find in one of their features.
Whatever the reason, the baby isn’t anywhere near to being a realistic baby. Quite apart from the problems of skin tone and texture, its hands and feet – and, worst of all, its eyes – are all much too small. Babies have BIG eyes – that’s one of the key things that makes them seem cute to us. A baby with small eyes is an unnatural thing, a squinty monster.
As I understand it, the ‘uncanny valley’ idea suggests that we would be more creeped out by something that’s very, very close to a human, but not quite human – perhaps a situation where the similarity is so close that we can’t consciously identify what’s off, but we know there’s something; the kind of high-level subsconscious processing of very subtle visual information Malcolm Gladwell talks about in Blink.
I think, for example, that when you watch slow-motion film of sportsmen running, you see all kinds of small details of the stresses their bodies are under – balling muscles, tiny shockwave ripples through their almost non-existent body fat, jowls flapping side to side. When you watch in real time, you don’t notice that kind of thing, aren’t consciously aware of it at all – but I bet you notice it subconsciously, and I doubt if any CG animations are yet paying that amount of attention to modelling movement.
When you watch CGI – particularly of living creatures – however well it’s done, however closely it’s modelled on real-life motion capture, you still pretty much always know it’s CGI. And, generally, you know it consciously, and can point to the things that don’t seem right – as you did JES, with the strange movement of the Navi. I think we’ll only be in ‘uncanny valley’ territory where we can’t really tell consciously any more, where we just have this overpowering gut intuition that it’s not quite right but we can’t tell why.
Neither the baby nor the Navi are anywhere near to being believably human, or believably humanoid, so I don’t think the ‘uncanny valley’ notion properly applies.
The reason the Navi are so spooky, I would suggest, is that they are so other: roughly human proportions, but lizard-like skin and feline facial features. And BLUE skin. Blue is not a colour that occurs much in the natural world – well, not in skin, anyway, certainly not mammalian skin. Also, we get especially creeped out by even slightly non-standard proportions, particularly in regard to the eyes (do you remember my post about the the mysteries of IPD, JES?). These Navi, with their broad, flat noses, wide-set
eyes, and HUGE irises look really, really weird. It’s not how close they look to humans that unsettles, I think, but how far away they look from anything we’ve ever seen.
Imagine how overloaded our brains would be if we ever did meet an alien creature face-to-face. I recall being scared out of my wits the first time I encountered a live chicken, when I was only about three years old. I have a very similar experience – running away, shrieking in terror, convinced that aliens had landed – the first time I met a girl with strawberry curls and freckles (I’ve learned to quite like the look now; but when I’d never seen it before, I found it quite tramatizing).
cynth: When the City Commission here decided to build an IMAX, it was with the idea that it (and the “science center” in which it’s housed) would attract attention from tourists. I guess there’s been some of that, but not much; up to now, attendance seems to have been pretty spotty. Now even that is threatened — now that a chain multiplex (20 theaters) has converted one screen to IMAX format. One way or the other, though — before the last few months, I’d been to exactly one IMAX film (at Cape Kennedy/Canaveral, in the ’80s), but have since seen two more. Now if I can just get IMAX closed-captioning…
(Such a thing may actually exist. I just haven’t checked yet.)
Froog: Whoa. I’m almost embarrassed to accept that as a comment; it ought to be elevated to full post status!
In the Alien movies — Sigourney Weaver again! — remember the androids? Until the one was decapitated, revealing its eletronic and hydraulic inner workings, Ripley had assumed it to be human. The sight of something so “human” and yet simultaneously so other was pretty bad, but I think nothing the audience felt could have matched what Ripley must have been going through. She must’ve felt like Wile E. Coyote five seconds after stepping off a cliff over the uncanny valley.
(Some gripes about Avatar are that the Navi are too humanoid: that any alien life forms we encounter will probably not even remotely resemble anything we know of terrestrial life forms, at all — human or otherwise. Thus, or so goes the argument, that the Navi are bipedal and bilaterally symmetrical, have hair and eyebrows — all this is just stupid. But for dramatic purposes, they have to resemble humans at least a little, otherwise the audience won’t really empathize with them in the inevitable conflict.)
In the My Name Is Earl sitcom, the title character’s brother, Randy, is afraid of birds, particularly chickens, particularly roosters. (This is true, in part, because he can’t remember what the proper response is if suddenly confronted by a rooster: do you play dead, or punch it in the nose?) In the episode in which we learn this, he suddenly finds himself in a wire pen, face to face with an ostrich. He — Randy — immediately falls to his back, all four limbs in the air, and screams for bystanders to call the police. Heh.
Ha – I don’t know that show, but I should go and check it out. I think that’s probably a pretty sensible response if cornered by an ostrich – they’re big scary critters. I couldn’t eat chicken for some years after that freak-out experience.
Yes, poor old Ian Holm getting his head wrenched off – that was one of the great, unsettling surprise moments in the cinema. I suppose androids that lifelike – and the replicants in BladeRunner – have gone beyond the ‘uncanny valley’ and gained acceptance.
I wonder if the ‘uncanny valley’ response isn’t something like the cognitive disarray, the reeling discomfort, the anxious double-taking we go through when we encounter someone with a really odd look – prominent body-piercings or a really extreme haircut or…. well, shaved eyebrows are the thing that unsettle me.
Froog: I don’t have shaved eyebrows. But in that vein, I think when someone walks in on me while I’m changing out my hearing-aid battery they experience some of that cognitive disarray.
You can almost see it in their faces as it happens: the initial response of Yikes, I’m not going to embarrass him so I’ll pretend like everything is normal — only to realize that “normal” implies the start of a conversation, which is obviously pointless. So they stand there wringing their hands in misery for a second or two. It’s one of the few times when I feel that being hearing-impaired puts me one up on those who are not, so sometimes — not always — I find little bits of stage business to prolong the moment.
Thank you! Ever since we saw Avatar the day after Christmas, my husband has been *insisting* that the uncanny valley has been crossed, which I think is a ridiculous claim. His logic is that they were very human-like creatures, but did not repulse. My response is that they weren’t human *enough* to cross the valley.
I think Cameron did a nice job of toeing right up to the edge of the valley with this work, but the fact remains that we are told point blank that these are aliens! These are cat people! Easily 60% of their emotions were relayed through body and especially ear movement. Crossing the uncanny valley has much more to do with being able to replicate the thousands of “microemotions” that a human face can convey through the slightest muscular changes. Watch an episode of “Lie to Me” and you’ll know what I mean.
The bottom line is that once I walk into a movie about aliens, I’ve already agreed to suspend my disbelief and accept that these are *not* humans. Therefore to say they’ve made huge strides in approximating humans is ridiculous.
Xena: I’m no expert on the uncanny valley (or the happy one, for that matter :), but I share your interpretation. What you said, exactly:
If you know in advance you won’t be seeing humans — and that it’s an animation, for that matter — well, there may be many things about the film you’re not comfortable with, but the ambiguous human-ness of the characters isn’t one of them.
Thanks for stopping by and commenting!
Indeed CG animations ARE paying that amount of attention to movement. One specific example is when Jake leaves the building after first waking up in his avatar and goes running for the first time, during the slow-mo part.
My thoughts on the uncanny valley, as I’ve been familiar with the term for quite some time, I really feel it only applies to representations of humans, not kinda-sorta-almost like humans, but things that are supposed to represent humans, and nothing else. This is where I feel Cameron sidestepped the issue altogether by the Navi being decidedly NOT human. Are there some issues with the implementation of the Navi? Yes. Most certainly. Are they show stoppers? Not a chance. Given the same technology, could humans be created the same way as the Navi were? Not without falling straight down into the valley. We’re not there yet as far as jumping the valley with human representations IMO. I think though, that the tech that was created/improved in the making of this film has pushed us closer, but it’s not quite there yet, and I think the uncanny valley is not applicable to Navi. (I’m probably repeating myself several times, it’s late)
I did also notice some animation strangeness, in both Navi and creatures. Two that stand out in my mind, and were in the same sequence, the bonding with the banshee, when Jake gets thrown over the side and pulls himself up he seems magically boosted up in the end, and once he has bonded, the head of the banshee, just for a moment, has a very “mechanical” movement to it, almost looks like stop motion animation, just for a brief moment.
Thank you for stopping by, Phrog, and especially for the comment.
My favorite special effect in the first Jurassic Park film was not, really, any of the scenes in which velociraptors were tearing about the place, or being scooped up in T Rex’s jaws.
(Well, I liked those scenes quite a bit for their entertainment value. But I think maybe they were so entertaining that I didn’t stop to analyze them critically, or even experience the slightest awareness of them as special effects.)
No, my favorite effect came at a moment shortly after the Sam Neill character arrives. I don’t remember the specifics, but everyone is out in the park-like landscape marveling at all the “gentle-giant” herbivores and they come across one lying on its side. (Was it sick? dying? Had it been shot or tranquilized? No idea.) The humans carry on a conversation and I think one of them — maybe even Neill’s character — sort of absent-mindedly reaches out and touches the dinosaur’s flank. It’s breathing, and we can see the skin rise and fall in a hyper-realistic manner.
That was the only moment where I think I could have believed I was looking at a real dinosaur, instead of a CG creation.
In human prehistory, there were probably quite a few moments of uncanny-valleyish responses when one human(oid) species encountered another — like Neanderthals and Cro-Magnons, or whatever. If so, this might suggest that the little thrill of characteristic revulsion/discomfort has evolutionary value. Like: Clearly, this is a human here before me. But somehow it is also NOT a human. It has no business being here! (Bashing follows.)
There’s a surprisingly good version of Beowulf made a few years ago (with Gerard Butler in the title role and Stellan Skarsgaard as the Danish King, shot entirely on location in Iceland), where trolls are depicted as late survivals of some giant Neanderthal-like species.
I think that Jurassic Park moment you mention, JES, (and any moment where a live actor actually has to touch an artificially created creature – at least in a close-up shot) is going to be done with animatronics rather than CG. Not that the CG animations don’t have a lot of subtle detail like breathing and drooling; but where you’re watching the belly rise and fall under Sam Neill’s hand – that’s a model with something inside moving it. Possibly the same issue with the scene Phrog mentioned in Avatar, taming the ‘banshee’ (they weren’t called that in the movie, were they? but I can’t remember what the Navi word for them was).
I don’t recall the other scene Phrog mentioned, where Sam Worthington’s avatar runs in slow motion – but I’m sceptical as to whether even there the animators really caught (or even attempted to) all the myriad minute details that you see in slow-motion film of real people or animals. And that was a special case: short scene, slow motion, close-up on the main character. I was talking about the depiction of the Navi in general: when you’ve got ten of them running through the forest at the same time, you certainly don’t have that kind of micro-detail lavished on them.
Yes, there was a good bit of animatronics in JP.
Banshee is what the humans called them, and as many times as I’ve watched it I can’t remember what the Navi word for them was either… hmm.. guess I know what I’m watching tonight heh. I can say it was not an animatronic scene. It really just looked to me like somebody dropped the ball for about a half second or so of movement.
Well, all I can say is, watch it again. The scene hit me square in the face the first time I saw it. Sure, they may not have tried to capture every nuance of a runner, but they got the cheeks springing in a very convincing manner. I realize this was all captured with their new facial rig, but even still things need to be cleaned up to be not too blatant to look unrealistic, and not too subtle to be missed. The same with the facial quivering that is seen a couple times through the movie during times of high emotion.
As far as the same detail being put on more distant groups of runners, well I can’t say one way or the other if it is or not. I will pay more attention to that on my next viewing. I do know you don’t spend extraordinary resources on things that just aren’t visible on things like this, especially since you don’t really get to appreciate those kinds of details unless they are magnified and slowed for you to catch. So in any case, it’s not an issue of the animators not trying to get that kind of detail in there, it’s more an issue that it would be a waste to do so. Normally such work is reserved for hero shots. In case you are wondering, yes I am an animator, and I am quite aware of these kinds of fleshy movements myself. This is one reason I was happy to see such detail on the screen.
I heard a lot of complaints – or ‘less impressed observations’, anyway – that most of the time, the movement of the Navi didn’t look all that convincing, Phrog.
There may be larger issues of ‘smoothness’ or ‘rhythm’ of movement or convincing weight transference in that; things that are immediately noticeable at a conscious level (even if one can’t precisely identify or articulate what the problems are). However, my suggestion was that the micro-detail – the stuff you only notice at a sub-conscious level, such as Malcolm Gladwell was talking about – may do as much or more to undermine the effectiveness of the illusion as these more ‘obvious’ shortcomings. And you notice this stuff (subconsciously) even in the background figures, even on a crowded screen. This is why I think actors in costumes (so long as they’re really good costumes) still trump CGI most of the time.
I suspect there may also be a phenomenon that you become more attuned to these shortcomings because of the contrast between the ‘hero shots’ and everything else.
Perhaps we’re not far off the day where there’ll be software for modelling this micro-movement of muscle/flesh that will make it a much less labour-intensive process. Then maybe the bit-part players will start to pass the ‘Blink’ test too.
When we get to that stage, maybe fully CG human characters will be possible too?
I suspect that’s still a long way off, though.
You sound like your disagreeing with me, but I’m unsure where I said I thought the movement was perfectly convincing. In fact, I do believe I’ve said it was indeed, not perfect, and in fact I believe Cameron himself has said that probably 10% of the scenes have some issues. I mainly was touching on the single point that they did make a reasonable attempt at the slow-mo look of a runners face in all it’s springy glory in response to reading your post stating you didn’t think that animators were even paying attention to that amount of detail. Now I realize on re-reading it that I took it a bit out of context where you seem to be focusing on wide shots with multiple characters and I was relating a hero shot. Indeed untill we get a blu-ray player (only watched it once in hi-def on a friends player over a month ago, otherwise I’m stuck with watching it standard def) I don’t have a clear enough view of the movie to say one way or the other.
I’d like to hear what YOU think about things, not what you heard from somebody else or whatever. You go watch the movie, study it first hand and then talk to me about what *you* saw in the movie. If you pick a scene out to dissect, I don’t have any problem looking as closely as I can at that scene and giving my opinion, at least as well as I can make out, as it’s what I’m doing anyway.
It’s not a case of “may be”, there were definite weight transfer and smoothness issues in some scenes, but if we’re still talking about the uncanny valley, vs talking animation tech in general, I’m still of the mind that the uncanny valley is N/A in this situation (and also disagree with what Cameron says in this instance, all I’ve heard him talk about is UV equating to photorealism, which I feel is not accurate), although the tech developed for this, and still being perfected, will get us much closer to believable CG humans. That’s what Cameron’s aiming for in the end, if it weren’t obvious enough.
In spite of all the problems I could point out in the movie I think it’ll be one of my favorites for a long time. Not all art is, nor has to be, perfect. As far as what animators *attempt* to do, you can bet your bottom dollar they did attempt to get as much detail and realism as they possibly could given any constraints placed upon them. If there’s one thing we animators do, it’s research reference material with a passion in order to intimately understand these very details to the fullest possible extent.
[...] a sense of repulsion people feel towards things that are close to human, but not quite. Even ‘Avatar’, with its beautifully detailed graphics created with the latest technology in film-making, suffers [...]