why don't computer generated graphics look real

[Description]

It doesn't look that good. Very smooth and rendered. No grain, imperfections, or detail.

Metal poles are supposed to a have a brushed, reflective appearance.

 

[Sahakiel]

Originally posted by: Shalmanese
www.archive.org has the siggraph winners from a few years back I think. you can download most of them for free. Apart from that, "The flight of osirus" was a short clip in the animatrix which was rendered by square and has some fairly advanced stuff. Final Fantasy still hasn't really been surpassed yet AFAIK simply because nobody can put that much money into rendering stuff.

Of course, this is all purely CG. If you want assisted CG, crowds are routinely generated from a few extras, the entire crowds in the titanic were CG based. most likely the reason you never notice it is because its so seamless.

Flight of the Osiris was done by Square.

 

[drag]

I think it has more to do with the artists combined hardware/software limitations then anything else.


I've been trained in art somewhat and what you see has more to do with perceptions then anything else.


Look at paintings for instance, a good artist can take pigments and create a illusion of reality, even if you were to analyze it scientificly It would be wrong a hundred thousand ways, but it doesn't matter.

For instance if I were to create a picture of a person and measure out porportions of a human being to 100% accuracy it will still look fake. In order to make a figure look "real" I have to distort the figure subtly.


The face is very important. You can make the entire figure one blob of color all mixed and swirly, but if the hands, feet and face are rendered perfectly then the figure will appear to be "real"

I make the hands and feet bigger. For some reason our culture enphisies those features in a human, especially hands are very important.

Other cultures are different, for instance you look at cultures that emphisis stregth in the legs and then their pictures you would see oversized buttocks and thighs. To us it's obviously silly, but to them it was realistic stylization of the human figure.

BTW, you know what most people notice right after the oversized lower extremities of these figures? They ask: "why do they have such tiny hands?"

Also you have to create a illusion of 3-d space on a 2d image. When people draw flat images most people have huge problems with their feet.

This is because when you LOOK at a person's feet your seeing them from the top. However when you look at a person you look AT them straight on. When you look at a person you look at individual parts and the proccess of viewing these parts they change, distort, and morph according to the angle at which you look at them.

So using these familar images in your mind's eye you draw the figure straight on. In reality, howerver, the head you look straight on and the feet your look down at them. The image you have created in your imagination you generally see the WHOLE body straight on and so you try to draw the foot as if your looking straight at it. This is very unreal so it looks all screwed up.


Moving animation is even more difficult and complicated then a single flat image.

Next time you watch the show "Cops" look at how the colors and shapes of the images are totally distorted by the proccesses of recording the image.

It's all pixlated, colors are muted. You can't focus on the badge, no details come out in the face or anything. The lights from the camera shine straight on.

However you know it is "real" because that's "real" image your seeing.

If you made computer graphics that are perfect 100% representations of space and time it wouldn't seem as real as the real footage because it's not screwed up in the same way. That screwed up-ness is what makes it real, because we assume it must be there. If you make the visual cues the same as what people see on "live" TV, then it will appear to be more real.

Also people need to aviod concentrating on details on every little thing.

The overall composition matters very very much. You must create the "suspension of reality" of a story or image sequence. That is how much a person is whilling to sacrifice is ability to discern reality.

The more compelling a image is the more control over the perceptions of your audiance. It's kinda like hypnotism. You just can't go out a hypnotise a bank teller to give you free money no matter how good you are, it's impossible. (or nearly impossible, some people are more scatterbrained then others). If a person beleives they can't be hyptontised, or they don't want to be hypnotized then you can't do it.

So in a realy good story or a image that is very compelling and taps into a subconsious level, then if as a artist you subtly hint that "this is realy real fake image" then your audiance will go "this is a realy real fake image I am viewing" simply because people LIKE being fooled. People pay to go to movies to be tricked into thinking this is real for 2 hours.

Also visual composition matters very much, you must create a image that controls the movement of a eye.

As a artist you get trained in "focus" point of a picture. I can (on occasion) go up to a painting or a image and after studying it point out different things and go "you will look or at least study at particular image first" then the eye will travel here and here and here.

For instance the image of the street that was shown to be "real" looking. You have a gigantic reversed L shaped mass of darkness This prevents your eye from moving off of the picture to the right and the bottom, it's like a gigantic arrow point towards the bottom right of the image. And what's there? Notice how the image is mostly green-ish, cool colors. The walk light is orange and bright, a warm color. That's a big contrast with the surrounding image. There is also a person there walking. People always attract attention, you can have one big picture of a pasture with a little spec of a person off to one side and that will be automaticly a focus point simply because instictively we look out for other humans.

Well that combination of warm color and person make that area the major focal point of the image.

Notice how the red lights of the interesting car keep your eye from wondering off on the lower left hand corner. They trap your eye and make it follow the line of cars that "point" directly at the orange "walk" light. Hense drawing your eye back into main focal point.

Also the bright sky draws your eye upward to the skyscraper, it's much brighter then anything else and thus is another focal point, it makes your eye follow allong the top of the building and the lines pull your eye back downward to the cars.

Check out how the street lights are. They are sweeping lines that contrast sharply the image in the background, black on the bottom with bright highlights on top. You eye likes to follow lines and either you go to the building with the downward lines of windows and edges which pull you back downward to the cars or you go the other way directly towards the light and the person walking. And just incase you may miss it the lightpost shows a red stop light, a high contrast color, but a cooler subdued version of red that ties it into the tail lights.

So the sum total of what happens is that you have a "swirling" effect. Your eye goes in a circle. The blue-ness and cool colors indicate wetness, (also notice how the bright reflections on the sidewalk make the figure stand out in a otherwise dark on dark scene) and a mist seems fills the air bit. The shinyness of the street and the figure with the umbrella indicates that there is/was a rainstorm.

The city is a very busy place crowded, however nothing is going on. The cars are standing still and sharply rendered, the poles are sharp and clean.

This is what tries to create the final emotional tie that finishes the illusion.

It's obvious busy, but you only see ONE person. That person is blurred and thus calling on our experiances with photographs and moving images indicates that this person is moving alone while the rest of the city sites still and apart from the person.

Everything is calm and clear after the rainstorm but the person is alone, moving for some internal reason (you don't see something that she is going for do you?). Maybe slightly ghosty?

Also the Red lights indicated "stop" the orange light indicates "don't walk" or caution. Water and wetness also indicates "cleanlyness", like how fresh everything seems after a summer shower. However the figure is protected from this by the umbrella which is a further seperation from the surroundings. Also other suggested human beings are in their cars, trapped, but protected, the figure is open and unprotected relatively to other humans.

Also it's weird because their is no reason to stop, no cars are comming in that direction and it seems safe. While others are standing still that person still moves.

Depending on how much this strikes a cord semi-conciously (how does this relate to you and your self image?) has a big effect on the "realness" of it.

Because no matter how good it gets it ISN'T REAL. None of it is. It is mearly a concoction of electronic signals. It's a flat 2d space in a 3d world. Even what you watch on TV isn't real, even "live" TV. It's a illusion, so obviously fake and distorted that it's not funny, but because you believe it's real and want to beleive it's real then it seems "real".