Look at my post, #34, for some reasons.
Arnold kind of looked like a real person in the first Terminator, but the robot at the end? Have you watched it lately?
You’re jaded. Although even in 1933 a lot of people were sophisticated enough to spot the effects, there were reports of peopl;e being bowled over by them back then. Similarly, before the release of the 1925 The Lost World, Arthur Conan Doyle took footage of the animated dinosaurs and showed threm to reporters in New York. Even when I was a kid I wouldn’t have been taken in by the effects (some were great, but some were truly awful. I assume they weremn’t by O’Brien, or that he was having a REALLY off day.), but they fooled the reporters, who wondered where Doyle got them. At the first showing of Winsor McCay’s Gertie the Dinosaur, some people reportedly thought the cartoon was a full-sized mechanical device.
as we get more sophisticated, infomed by out viewing, our view of older special effects changes. A lot of stuff in the first Jurassic Park looks blatantly false to me now – the lighting on the velociraptors in the kitchen seems completely off, and they appear to glow slightly. But I;m still utterly blown away by the work in, for instance, the most recent King Kong.
To take another example, look at Forrest Gump. Say what you will about the plot, the effects shots were brilliantly done. “What effects shots?” you say. “Exactly,” I say.
Things like removing Gary Sinese’s legs. Or Forrest standing in a crowd of demonstrators at the reflecting pool. Things that don’t look like effects shots because nothing is exploding or dripping ooze.
I respectfully disagree. I am of the opinion that stop-motion photography, while harder, almost ALWAYS looks less realistic than CGI. Especially when you are stop-motion photographing something human-scaled interacting with humans. It always tends to have a jerky, unnatural quality about it. Compare the skeltons from Jason and the Argonauts with the ones from The Mummy.
It’s really the mixture of special effects that makes a scene work, not one particular effect. Take the scene in Aliens with Ripley fighting the queen with the power loader. Some of the long shots were stop-motion photography. But they are interspursed with animatronic and puppet/guy in suit close ups so you aren’t able to get a good look at the weird movements. The T-Rex scene in Jurassic Park was also a mix of animatronics, puppets and CGI. Using any one technique would have made the scene look fake.
I forgot to add it to my list, but another thing that makes CGI stand out as false is pattern repetition, especially with crowds or partical effects. Like if it looks like you have the same 5 pikemen over and over again.
I don’t see what you’re disagreeing with – I agree that wel-done CGI looks better and more realistic than stop-motion. The question was about why it was harder.
QFT. Modern effects are great and technological progress is inevitable, but there’s something amazing and magical about looking back on fantasy films of the 80’s and realizing that all the puppetry and animatronics were done live by real people in front of a camera. If they didn’t get it in that take, they didn’t get it. There was no digital post, it was a photo process stage performance. Films like Labyrinth or Return To Oz are wonderful examples of this art; it’s far more amazing and impressive to me when I look back on characters like Tick Tock or Jack The Pumpkinhead and wonder just how they made those guys work. Today, fantasy characters may be infinitely more dynamic, but they don’t captivate me in quite the same way because they’re all ones and zeroes. I can think of only two CGI characters who were big exceptions to that rule because they were so groundbreaking: the T-1000 and Gollum, neither of whom was the result of special effects for effects’ sake. I for one also liked the animatronic and stop-motion Terminator skeletons. Compared to the third film where everything went CG, I like the physical presence of the first film’s physical endoskeleton on celluloid, even if it was only a bust. I respect the craftiness of the design, the quality of its physical construction, and the skill of the puppeteers that controlled it. The first film may have been made on a shoestring and the quality of the stop-motion was probably below par even then, but the image of that endoskeleton limping his way down a dim corridor still gives me very “analog” chills.
A digital character is only as effective as the viewer’s emotional connection to it.
How important is realism anyway? I can’t honestly say that many of the walking dead from the original Dawn of the Dead movie looked like anything other than people in really bad makeup. It’s still one hell of a movie though. The latest Star Wars movies featured CGI but for the most part I thought it looked fantastic. Especially compared to the AT-AT Walkers from Empire Strikes Back.
Then again I’m one of those people who don’t find the effects from 2001 to be all that impressive these days. Impressive for the time it was made, the attention to detail, and the attention to realism? Certainly. I don’t find that it looks particularly real though.
Marc
Case in point: When people talk about the special effects in the Lord of the Rings movies, you’ll hear many mentions of the Balrog, or the Eye of Sauron. Ever notice how nobody ever mentions all of the scenes (nearly the entirety of the movies) where hobbits are interacting with humans? Every time you saw that happening, there was some sort of special effects going on, and nobody even realizes it. Now, that’s good work.
I’ll agree on Gollum, but I hold the exact opposite view concerning the T-1000. For me, the whole movie looked like a proof-of-concept to show off the CG they were able to do with the T-1000, which (I thought) was already starting to look dated by the time the film was released. I think that the third one, regardless of the technique used for the effects, used them much more intelligently than the second one: Include enough effects to show what the monster is capable of, but most of the time, she’s just a human actress. None of this nonsense of “walk into a wall and invert to turn around”, which serves no purpose other than impressing the audience.
I think one other issue is the cost. CGI is sometimes less expensive than other forms of FX. But at the same time, it adds up. A lot. They can sometimes amount to vast fortunes of cash, which gets burned through like tissue paper. I think it’s often a good idea to just get the shot done. If you need to touch to up later, so be it, but at least it’s done and you have something to start from rather than paying an artist at ILM a month’s salary to work on a 3 second shot.
And of course, it’s been mentioend that it messes with the actors to have to interact with things that are not there. Doable, but harder.
Well, that particular example isn’t really typical of the sorts of things I had in mind.
Legitimately curious: I was under the impression that for the most part, that was a practical effect done with a trick chair. Was I misinformed, or are you referring specifically to the parts where that doesn’t apply?
Trick chair when he was in the chair, but at all other times he was wearing bright blue socks and they were digitally removed. Or rather, they were digitally replaced with an image of the background.
As always, you have good work and bad work. Muscular simulation is implemented and more or less standard practice on high-end character setups. I had to animate a cat for the last movie I worked on, for that I was given a cat mesh and a workstation, I built a real cat skeleton under the mesh and I attached actual muscles to it, getting the mechanics to work took nearly two months. The end result was quite good if I may say so myself 
, a very flexible movement and you can actually see the muscles and bones sliding under the skin, specially on the shoulder blades.
Rest assured that CG animators know, or should know that; it´s standard course material (I´m self taught though) The problems are A) that there are, sadly, many crappy animators around and B) time constrains that only leave enough room for rough animation. Also I´ve noticed that in big screen movies there seems to be more care taken on the purely visual than in the animation quality, I´m not sure why is that; in my personal experience I can say that sometimes a director demands things that simply are not good from an animation point of view.
Oh boy, all the times I was this close of starting Pit threads about that…
Most people work on very simple concepts about what character motion entails, it´s chock full of very subtle things that usually are not even allocated a minimum of thought.
For example, a bad animator (or one seriously timeconstrained) would make a character open his mouth by just moving the lower lip and teeth down, maybe so far as rotating them a little as the jaw pivots on its socket. The real deal is to make the whole face and even the neck work together with that motion, as the mouth opens the corners of the mouth move inwards as the lips are stretched, the skin pulls slightly down on the nose, more on the nostrils than on the middle; the cheeks sink slightly and slide over the muscles, maybe a double chin appears but if not the skin under the jaw has to swell a little; heck even the ears move a little, etc. And on top of that you have to find the way to make the mouth open the right way to express whatever you´re trying to show in that particular moment. It´s not the same thing opening your mouth to get a bite off a burger than an incredulous jaw drop.
If you don´t have all those subtle clues you see a guy opening his mouth and something´s not quite right even if you can´t really point to any or all of that
Okay, so we’re on the same page. Thanks!
It seriously bothers me a lot that in all three Spider-Man movies, they hand animated all the digital stunt doubles, instead of using motion capture (or a hybrid of both).
When I went to a seminar that had the Sony Imageworks Animation Supervisor (Anthony LaMolinara) giving a talk about the first Spidey, he said he preferred hand animating because he thought motion capture wasn’t good enough, it looked fake. Well, guess what, dude - hand animating something and getting the centre of gravity and inertia wrong every time is even worse than bad motion capture!!
He continued doing the same on the second Spidey, and didn’t work on the third, but even in the third film they kept with the same method, and it continues to stick out like a sore thumb. Very disappointing.
There are lots of times when it couldn’t be faked that way. At several times he’s out of the chair and swinging around or moving energetically around, and it’s pretty clearly Garu Sinise, and not a legless body double.
they did the same thing with Rosa McGowan in the “Planet Terror” section of Grindhouse (there’s a bonus DVD that shows the effect work. There are times when she’s got her leg hidden (like on the hospital bed), but most times she’s wearing a green-colored leg cast (to keep her knee and ankle immobile), and they difitally removed the leg and composited the machine gun in. They did this even in long shots where they could easily have used a body double.
My feeling is, don’t make CGI the star of the movie. If CGI has to be used as a character like Gollum, make sure a real person like Andy Serkis is pulling the strings.
Ridley Scott used CGI to recreate old Rome in Gladiator, but he used it for background and didn’t have his cameras focus on it like magpies stealing shiny things.
I think Mummy 2 is the most egregious example of letting CGI artists whack off too much. Like the scene where humans with scimitars take on living dog statues with shiny dark blue and gold textures. There was so much bright color with no regard to diminishing shades, it was just one big writhing mess of yuck. It seemed like a lot of scenes in that movie were the “Hey, look what we can do!” variety that were slapped in the story regardless of importance to pacing or plot.