"It takes 6 months to put a robot together" - Transformers-2 vfx supervisor Scott Farrar

If you thought all the visual effects magic in Transformers came easy, think twice. "It takes roughly about six months to put a robot together. It's like going into your workshop and making those parts, except it is a computer graphics workshop," shares the visual effects supervisor for the Transformers movies, Scott Farrar, in an interview with Bollywood Hungama.

“It takes 6 months to put a robot together” – Transformers-2 vfx supervisor Scott Farrar

Scott Farrar joined vfx house Industrial Light & Magic in 1981 as a camera operator on Star Trek II: The Wrath of Khan and went on to Oscar nominations for The Chronicles of Narnia in 2006, Backdraft in 1991 and A.I.: Artificial Intelligence in 2001. The original Transformers was nominated for the Academy Award for Best Visual Effects in 2008.
Excerpts...

What exactly are you doing on the pyramids for Transformers: Revenge Of The Fallen?

The foundation of visual effects today really starts with match moving. What does that mean? It means that if we want anything that we create in computer graphics to look like it is on the ground or climbing or moving on surfaces, we do have to measure very carefully all the surrounding terrain because that has to be recreated in the computer. If it doesn't match, the feet will float or what has been created won't touch. So it is ultra specific. It is pretty high end.

Your colleagues are working on the Pyramids, taking exact measurements. What is that for?

Well there are a couple of different things that are happening. For instance we have a little character that walks up the doorstep - that is pretty simple. You can see how high the steps are, all that has to be measured. It is nice and flat, so that is easy. This (on the Pyramid) is more difficult. The stones are uneven and of different sizes because they are worn with weather and time. So we have to photograph from different angles and we have calculations where we can take the photographs and use those to create measurements. For instance...that stone is higher, this stone is lower, that sort of thing. So it is pretty cool because it is very sophisticated at this point. Therefore the shots (created by computer) do not look fake. It really looks like the guys are there.

So at this point there is supposed to be something big climbing the Pyramid?

There is and we do all these measurements - some of them super accurate, based on what the camera sees. The other thing is that we might recreate some of this situation later; we do a lot of photography where we recreate the model of the Pyramid in the computer. So we might cheat a little bit, we might change the shape of some of the blocks. We have to be ready for anything because we have a rough shot design. We know we have down views, helicopter views, views from below; but we don't know what the cut is. And we don't want to start shots with Michael Bay if we don't know what the cut is. Why? If you start a shot that is not going to be in the movie it is very expensive - and we don't want to do that.

What would you use as the stand-in for a Transformer like Optimus Prime?

It is very sophisticated! It is a window washer pole! I believe in low tech for high effect. How it works is like this...if you have robot and I want the actor to look at the robot then you could have a big head (to represent the robot) but those are really heavy. So we just have the pole and the actor uses that to get an eye line. Shia has become very good at this. The actor has to sell the shot, otherwise you don't believe it. So we work very carefully with it. We just use that to help them react. That's what it is all about. We also use them to represent height of the robot so the camera operator can aim at it and judge composition.

Do you ever have faces for the robots?

We do have faces. They are cut-outs that we have blown up from the art work until they are actual size. Some of them are so large that when Michael says he wants so and so, like Megatron, and the wind pulls the face it's terrible; so he'll say, forget the face. So the pole works just fine.

How much interactive stuff like smoke is computer created?

Tons, I did Roger Rabbit and then we were scared to death of physical effects because we liked to shoot them as separate elements and layer them in after the act. But that looks pretty fake so we are fortunate today that we have these very sophisticated compositing tools - compositing just means putting all these pieces together layer by layer like a sandwich, building from the background to the foreground. When we shoot these movies with Michael they are very dirty - all the smoke, flames and explosions.

That was kind of new for me on the first Transformers, but it is fantastic. You cannot believe how complicated a piece of smoke is that you photograph. If you look at a dust cloud there are all these colours, that are self-shadowing and there is tonal change as it moves and the way it reacts to the light. There are all these things that you would never guess. So it is fantastic for us to do that.

The first Transformers had amazing effects. So how can you top them?

It is hard. We were trying very hard to be photo-real in the first one. The problem with a lot of the shots that we did was that we were still learning. Now we have refined a lot of the tools so that we can have 10 people working on the same character and the character will look he same all the time, rather than constantly adjusting. So we have tried to make it easier on ourselves. But how do we top the first one? There is a higher production value this time; the scale of this movie is huge. When I was a kid Ben-Hur was a big film with scope and scale that was unbelievable. Well this is like Apocalypse Now and Ben-Hur rolled up into one. Plus it has got comedy and a great story. For us it has very, very challenging effects because of the size and the numbers of the characters and the complexity of the characters. Optimus Prime had 10,000 parts and we have characters in this that are five times more complicated. And all those parts have to look correct and they all have to move correctly. Plus, 2 sequences are photographed in IMAX, so the resolution is.

Can you give an idea of the time and money it takes to bring just one of the robots to life for 10 seconds?

That is a great question because no matter how many times they appear in the movie it takes a certain amount of work. It takes roughly about six months to put a robot together. This may be surprising, but you have to build all the pieces. It is like going into your workshop and making those parts, except it is a computer graphics workshop. The men and women who make these characters, make the shapes and those shapes have compound curves, which is complicated. Then some shapes have 4 to 16 layers of information in the computer, so that it looks like plastic or glass or shiny chrome or brushed steel, plus all the pigments of colour. That is a lot of stuff, for every piece. The building of it is one thing, that takes 12 to 16 weeks, and then you go into paint and textures. Then there are the people who connect all the pieces and that can take even longer. You have to work it all out so that basically the skeleton hangs together in the computer. Then you are about ready to start putting it in shots, but even then it is not finished. You keep working on it. The money? It is millions of dollars for this kind of thing. It is all based on how many weeks it takes to do each thing.

Do you have an office full of computers with a team working round the clock?

No people have to go home and sleep! We have mandatory breaks and everyone is involved in athletics to keep in good shape because you have to be able to sustain the time that you are sitting down. We have 250 people working on this movie. Towards the end there will be about 300 people working on it. They are divided into character animation, lighting and rendering, paint people, people who work on the plates and the final compositors

Is creating Devastator an exciting challenge?

Right now Devastator is made of many large, earth-moving vehicles. I just saw a rough of that and it is looking very cool. But it is in the early stages. We go for photo-real parts so we copied the real road graders that are used in the movie. My guys photograph those things; we use all those parts. In the first movie we had about 6,000 still photographs of engine parts or car parts, like disc brakes and clutches. We drew from that so that the things would look real. It is the same thing with Devastator; we have got to make it look like heavy earth-moving equipment.

How hard is it shooting such a huge robot like Devastator with humans?

Scale is a difficult thing. If you have a robot all by itself against the sky then you don't know how big it is. With Devastator his hand is larger than all of us [a group of people]. So when John Turturro is running about below Devastator you will only see little bits and pieces. Then it looks pretty cool.