Viewpoint - Digital Backlot Economics
Issue: Volume: 31 Issue: 2 (Feb. 2008)

Viewpoint - Digital Backlot Economics

 
Joel Payne is a 3D environmental artist currently working in the games industry.
 
"Digital backlots could be used more effectively in gaming, saving time and money."
 
The term “digital backlot” loosely describes a production in which the shooting occurs entirely against a greenscreen or other blank backgrounds, with the “sets” added later through the use of computer graphics.

It’s difficult to say which film was the first to use the virtual backlot, because three such movies were all released at about the same time in 2004. The first to hit the screen was Immortel (Ad Vitam), a French film based on the graphic novels La Foire aux immortels and La Femme piege. The second was the Japanese film Casshern, which was based on an earlier anime of the same name. The third to be shot on a true digital backlot was Sky Captain and the World of Tomorrow. It hit theaters in September of that year.

Of course, examining the films’ release dates does not determine who came up with the idea first. That being said, Sky Captain director Kerry Conran set out to work on making his own film without standard set-building limitations. The concept Conran developed would become Sky Captain and the World of Tomorrow, which utilized the digital backlot to its fullest. Despite all its new bells and whistles, Sky Captain did not win box-office success, which begged the question, was the use of the digital sets worth the effort?

In 2005 came Sin City. Based on Frank Miller’s series of graphic novels, the movie was shot entirely against a greenscreen with Sony’s CineAlta high-definition cameras. The feature proved very successful, grossing $158.7 million worldwide, and indicated that a digital backlot was a viable option.

This past year, at least one film shot on a digital backlot was released. 300, based on Frank Miller’s graphic novel about the Battle of Thermopylae and directed by Zack Snyder, received critical acclaim. I think it is safe to say that there will be more of these types of films in the coming years. For one, digital backlots offer certain advantages over shooting on a physical set. Not only does the director have total control over the environment, but he or she can also create sets that couldn’t possibly exist on earth.
 
Reusing Assets
While digital backlots are fairly new to the feature-film world, they are hardly new in the gaming world, where artists have been using digital sets from the very beginning. Take, for example, Mortal Kombat, with its photo-generated fighters atop hand-painted sets, or Myst, with its fully-filmed live action over beautifully detailed 3D backdrops. While the film industry has had mixed success with digital backlots, they have been used in gaming all along. But have they been used in the most effective way?

With costs of developing game titles growing each year and consumers expecting more and more from their 360 degrees of visual bliss, the game industry has to return to its roots. No, I’m not talking about another Myst, but modularity, reusability, and prop outsourcing are becoming frequently used expressions. We’ll have to start thinking about how to build structures once and redress them again for future projects by taking a page from what Hollywood has been doing for years.
 
The gaming industry should take a play from Hollywood of yesteryear, when film studios reused generic sets, facades of buildings, and street corners.

Universal Studios, Paramount, and Warner Brothers have been using their practical backlot façade streets in countless movies over decades­—redressing, repainting, and reusing the same beautiful old streets. Perhaps fans noticed the repurposed New York street from the Universal lot on both the Spider-Man and Transformers films, but did you know it also was used in the film Interview with a Vampire? There is a great advantage in the concept of building it once and redressing it for reuse, and the game industry will need to embrace this if the current development cycles and high visual expectations continue.

Midway Entertainment is already sharing internally what the developer considers a digital backlot database among all developing companies within its own family. From standard textures to generic buildings and vehicles, Midway is capitalizing on the novel idea of not having to reinvent the wheel for each new project.

Let’s face it: We’ve come to a point in game-making history where the models aren’t going to date themselves as quickly as they have in the past. In the last generation of games, it is no longer commonplace to see the polygonal edges of the models. With the addition of high-quality normal-mapping techniques, we can pack a tremendous amount of complex detail that can make the flattest of single surfaces bubble with dimensional joy. High-resolution specular maps and complex global-illumination techniques are making games look more real then ever.

Having worked on two CG cartoons for television back in 2002, I find it exciting to know that the level of detail in the games I’m working on now (in real time) far exceed the image quality that at one point took hundreds of computers to render! Wow! What an exciting time to be in the game industry.

I witnessed a similar graphical revolution back in 1993 as a 2D pixel pusher. At the time, we were limited to drawing at 320x240 resolution, but just as I was getting the hang of things, our next project switched to 640x480. With double the resolution came double the time to paint in all those wonderful details. I went from being able to hack my way through as a computer artist to actually having to fulfill my obligations as a true digital painter, and I had the pixel bandwidth to do it!

So here we are again. No longer can we complain about not having enough to work with, nor can we mod or hack out a Quake clone with minimal traditional art skills. We have the true freedom to do what we want and have it look as we intended. We can’t just slap on a photograph we’ve made, tile it onto a wall, and expect to live up to what consumers are demanding from us. However, this higher level of detail adds up to extended production costs and time. So how do we harness the power of this next generation of art and still stay under budget?

While smaller game developers don’t have the luxury, in most cases, of owning their modeling assets, where should we draw the line on the benefits of reusable 3D models? It would bode well for all game publishers to consider the cost benefits of allowing developers to retain the basic core building blocks of a project because in the long run, it’ll dramatically lower the costs of game development. I’m not talking about the recognizable features of iconic details that make intellectual properties what they are; but let’s face it, a door is a door, the Eiffel Tower is the Eiffel Tower. Do we really need to build it over and over again? What about that pine tree I see in every game, or that trash can or crate that is frequently used to fill in dead spaces.

If you really think about it, it’s mostly the texture and lighting styles that define “the look” of a game and make it unique. Take, for example, the film The Day After Tomorrow. Many of the buildings were just simple low-poly boxes, but they had complex normal-mapped textures applied to them, making them look real. I have it on good authority that those models have since been used in other films.

Thinking about building game art in terms of modular modeling pieces should be a staple of every project plan. Building on a standard grid size not only maximizes pixel consistency, but it also sets standards within a company that crosses over from one project to the next. Organize your backlot just like Hollywood does, with wall sets, doors, light fixtures, building trims, cars, and more, and think about how many ways that one item can be used down the road before you build it.

If sitcoms can keep us entertained by using a single living room set for five years, then we as game artists should be thinking about how to get the most use out of the spaces we build—and then be smart about how we build them.

A 16-year veteran in the DCC industry, Joel Payne is a senior 3D environment artist currently working on Silent Hill 5 for Foundation 9. His work can be seen at www.DigitalBacklot.com .