Is the reason why 3D animation varies so much because of the artist’s skill, or is it the sophistication of the technology?

Didn’t I write a really long answer to this question at some point? Is it lost forever? :(

Auggie found it. :)

All of the above. The quality of the animation depends on both the skill of the creator and the program used. It also depends on what exactly you’re watching. Certain games, for example, don’t have the capacity for high end graphics – simply to keep it usable for the player base. Very realistic or high end graphics would make certain games very difficult to play, especially for PC as opposed to console gaming.
The 3D items you see in a game or an animation are based on a polygon mesh, which is made up of various parts. The more vertices you have in your mesh, the more defined and well shaped your object will be, and also the texture you lay across the mesh will have a better capacity for detail and realism. Unfortunately, the more vertices you have in a mesh, the more demanding it would be on a graphics system.
The quality of a texture tends to be based on the ability of the artist, however, you can take a flawlessly detailed texture and slap it on a lousy mesh and it will look like crap. Though, it sounds more like you’re talking about the 3 dimensional aspect of what you’re seeing, and not so much the visual aspect. There are certain things that can affect how a texture will appear on a mesh, and certain qualities (which again, add to the demand on your graphics system) that can create a more realistic look. A mesh with an editable alpha layer and an editable, functional bump map can be far more versatile and realistic than a mesh without those capabilities. Again, a lot of that boils down to the dedication and skill of the creator.
Of course, if you want to compare a Pixar movie to a big PC game with a lot of players, the difference in graphics is going to be shocking. However, watching a movie and playing a functional, interactive game (especially with a large number of people on a variety of machines with different capabilities), there are just different aspects that have to be taken into consideration in order to keep it functional. I’m sure cost is another factor.

Some of the things you’re mentioning are also artistic preference. Realism is not always the ultimate goal.

It’s mostly the artists skill. The program allows for the vision to be carried out, but it is not automatic by anymeans. The artist is programming the animation, the computer is computing the actions.

A little bit of both

@ANef_is_Enuf Very realistic or high end graphics would make certain games very difficult to play, especially for PC as opposed to console gaming. I know computers but not so much when it comes to how computer graphics and animation works. Breaking it down to ridiculously simple high-end graphics would be hard to do on a PC because It has to be dynamic, the PC memory has to divide space or time with other applications, a console clocks faster? A PC with a very fast CPU, tons of RAM and a fast front side bus could not do it smoothly enough? Would having L3 cache help?

The 3D items you see in a game or an animation are based on a polygon mesh, which is made up of various parts. The more vertices you have in your mesh, the more defined and well shaped your object will be, and also the texture you lay across the mesh will have a better capacity for detail and realism. Unfortunately, the more vertices you have in a mesh, the more demanding it would be on a graphics system. When I was watching Finding Nemo, and Monsters Inc. am I watching a version of polygon mesh technology or some other process? In order to get Sully’s fur to look, well, furry was that the skill of the artist that is rendering it, the computer running it, or the program interpreting it? When you say _graphic system are you speaking of hardware or software?

The quality of a texture tends to be based on the ability of the artist, however, you can take a flawlessly detailed texture and slap it on a lousy mesh and it will look like crap. If the texture is not the mesh, what is the construct by which the texture is made? How do you know if you have what you want if you can’t create it one the frame, mesh, etc it is intended to be on?

Some of the things you’re mentioning are also artistic preference. Realism is not always the ultimate goal. Why would you spend all that time and effort using a 3D construct when you do not care for realism, or render shadows that fall anywhere near accurate?

I don’t know computer stuff. Someone else will have to answer that.

Pixar, How We Do It
Yes, it’s a very similar concept, but more in depth. To make Sully look furry is a combination of mesh, texture, mapping and animation. There are multiple processes that go into creating a realistic digital animation.

Your mesh is the 3D “frame.” By mapping the mesh, you can take a 3D object and turn it into a flat image. With that map, you are able to paint a texture (sometimes referred to as a skin) in a way that when stretched around the 3D mesh, it will line up and create an illusion of a realistic person, place or object. For example, this face texture looks quite odd if you didn’t know what it was for. However, it has been lined up and drawn with a map to fit a specific 3D mesh, which will, in turn, look like a head when they have been combined.
Knowing what you’re going to end up with is a matter of knowledge and skill. Someone who has never created a texture or a mesh would probably have no idea how to go about it without learning the process, because it can be a bit complicated.

If you’re talking about misplaced shadows, then that is indicative of a lack of skill on the part of the artist. If you’re talking about hair that doesn’t float or limited details, then it is probably an issue relating to functionality over appearances. If you’re talking about cartoony looking features compared to realistic or photo-realistic details, then that is most likely an artistic decision.

In terms of the technical questions, each vertex would be represented by 3 numbers in memory ( x, y, z ), then multiply that by the frames per second, and you have an idea of the number of points needed to be calculated per second. Shadows are done by ray tracing, which I believe is computationally intensive, as are reflections. Also there are the textures, physics collisions, environmental effects such as smoke, fog, haze etc. Modern graphics cards are designed to process many simple calculations in parallel, which is ideal for these types of applications.

With regards to the cartoony look of the animation, this may be done partly to avoid falling into the uncanny valley. @ANef_is_Enuf has some terrific answers.

Another reason is that the 3-D display technology (or more precisely, technologies) are still developing.

Well, there are basically two kinds of computer animation. One is hand drawn and the other is motion capture. The motion capture is when you take a person and coat it in a mesh, and then have them act out. The animator then takes the computer generation motion capture image and then draw on top of it. An example of this would be Gollum in Lord of the Rings

In hand drawn animation, the animator draws the animation directly to the computer, like they would draw animation on paper. The computer then will fill in the gaps, (called “inbetweening”). The computer doesn’t do a very good job of inbetweening, so the animator has to look at each frame and tweak it. Pixar uses exclusively hand drawn animation, and ILM’s Rango was all hand drawn (i.e. no motion capture).

The rendering takes all the computing power. Pixar uses Renderman. ILM used the Death Star but reitred it, and WETA (who did Lord of the Rings) have other hardware.

@Hypocrisy_Central At the same price point, a console will tend to have better graphics than a PC for several reasons.
First, the console is a computer optimized for graphics at the expense of other aspects, while a PC is a general-purpose computer.
Next, consoles are designed to work with TVs, which until recently had much lower resolution than a computer monitor (typical HDTV resolution is 1920×1080; my fairly elderly computer drives two monitors for a combined 3200×900 (or 3200×1800 with the virtual desktops) and it can display at high framerates and levels of detail across both at once).
Consoles run software which is built and optimized for that particular platform. Because PCs are all different, most PC software can’t take advantage of such optimization (Gentoo users are grinning smugly at this).
If you are willing to spend tons of money, a PC can give you a technically superior gaming experience, but is it worth it? The answer will be different for everyone.

@Rarebear The motion capture is when you take a person and coat it in a mesh, and then have them act out. The animator then takes the computer generation motion capture image and then draw on top of it. An example of this would be Gollum in Lord of the Rings The computer might capture the movement, but you still have to put skin, clothes, armor, etc on the image. How would that be done? If you have motion capture, do you need actors the right scale of the 3D image that will be built upon, what technique is implemented?

@Hypocrisy_Central Right. The computer captures the motion, and then the animator adds the stuff. The rendering process completes it.

In terms of your second question, I’m not sure. If you really want to know the answer, I’ll ask my animator friends to weigh in.

@Rarebear et al. Can someone explain what the processes are going on here? Is this a step in general computer animation of game animation creation? Did the woman form the armor and other items are being built on, part of the base program or what it motion captured, or scanned in from a real in the world model? Does it help to have a tablet do the armor building?

@Hypocrisy_Central “If you have motion capture, do you need actors the right scale of the 3D image that will be built upon, what technique is implemented?”

Perhaps I’m misunderstanding the question, but monition capture will give you the relative scale of the movements of your skeleton. Once you have these, the computer can easily scale it proportionally up or down to any size they want relative to itself. This seems so obvious that I’m wondering if I’m misunderstanding something.