What exactly is a 3d Artist?

I long for someone to write the book:  “The Succinct Guide to Explaining 3D at Social Occasions” (as suggested in this CGTalk thread).  I’ve tried different answers to the terrible question: “so what so you do?” but none have been truly successful. What usually happens is something along the lines of:

Me: I work in video games.
Them: Oh that is soooo cool! So what exactly do you do?
Me: I’m a 3D artist.
Them:(blank stare):… “Hmmm.. and what exactly is that?”
Me: I make the stuff that you see in the game.
Them: Oh so you draw it.
Me: Well not exactly I make it in 3D, in the computer.
Them: Oh so the computer makes it for you….

Then either you leave it at that or you try to explain what a 3D artist does and end up having to detail the entire video game pipeline or even worse how a 3D software works: “So you see you have this entire 3D world that is replicated inside your computer…and you can build things…” (more blank stares). This usually leads to painfully long and boring conversations…

As I am hopeless at explaining what I do for a living on social occasions, I thought I’d give it a go here… Computer graphics is a very young domain and few people outside the industry understand what it entails. A lot have, watched animated movies, played video games and maybe even viewed a few special effects making of’s, but what it’s really like, what we do on a day-to-day basis, that remains mostly unknown.

“3D artist” is actually a very general term. In most studios, employees are specialised in one aspect: modelling, animation, texturing, lighting, rendering etc. I will detail each of these tasks in later posts. Computer graphics are used in various domains: the entertainment industryof course (movies, video games, animated features, etc.), but alo commercials, architecture, medicine, the military and many others.  The major difference for the 3D artist though,  is whether they are creating something that will be viewed in real-time or precalculated. Let’s say you have created a 3D scene using all the previously mentionned tasks. If you are making a movie, you will then have to produce 24 images per second (one for each frame). You will place one (or several) cameras in your scene to capture what is happening from a certain angle. The computer will then calculate the view of the camera  for each frame and produce an image. When you have all of the images you can replay them at a rate of 24 per second and you will see your animation. It doesn’t really matter how much time the computer takes to calculate each of these images.  As a matter of fact these numbers can be quite mind boggling: one frame of Pixar’s movie Car’s took an average of 15 hours to compute… which means 15 days for only one second of animation!  On the other hand,  anything that uses interaction has to be calculated in real-time (video games are the most obvious example for this).  You cannot know what the user will do in advance ; you have to adapt the 3D scene to his/her actions. The computer then has to calculate the images on the fly so that the user can view them immediately… and this can happen up to 60 times per second! For the artist, this means they will have to use various techniques, shortcuts and simplifications to ensure that the computer can perform that fast. On top of looking good, what they create has to be quick and easy for the computer to calculate… This is the reason why your video game doesn’t look as good as Avatar, the  movie! 😉