Concepts and Costumes

Some more information has been released from the NinjaBee secret vault.  Microsoft is allowing devs to create costumes for avatars, and we’ve taken the opportunity to make some fun stuff, themed around the pending A World of Keflings (sequel to A Kingdom For Keflings):

A World of Keflings Avatar Store Preview

(this graphic is copied from there, but it’s worth clicking through for some extra details)

A World of Keflings Costumes

…and some of the pieces of those costumes are found here.

It’s a curious thing, this new MS policy, allowing devs to have a bit more to offer in the avatar customization arena.  Not only is it a way to maybe monetize interest a little more, but it’s also advertising of a sort, not unlike branded t-shirts.  It’s also fun; I can’t wait to see someone meandering around in a NinjaBee costume.  I think it’s a good thing, and I’m very curious to see where things go.

Beyond this being an announcement for my company, though, I get to show off some concept art that I did for the male Arabian-flavored costume:

Arabian Male Concept

I spent a fair bit of time painting this costume over a generic avatar, offering this as a proof of concept for what we might offer for avatar costuming.  The final male Arabian costume doesn’t quite match up with my original concept, but it’s fun to see that it’s at least similar to what I dreamed up.  It’s especially fun that they kept the little trim decoration design.  I doodle that sort of thing all the time in my notebooks and sketchbooks, and it’s fun to see it get through the production pipeline.

Y’see, concepts don’t always come out in the final game as intended or originally conceived.  Larisa lamented this a while back, wherein I responded with a bit of a peek into the dev process in the comments section, here quoted for completion’s sake:

Might I chime in as a game artist with a background in film production? I’ve worked with some fine concept artists, but inevitably, given engine or time constraints, the final implementation of concept art will not match the concept perfectly.

Some companies do try to match it as tightly as possible, and others use concept art as mood pieces to set the emotional tone for a piece of the game. Still others are mere color studies, a great many others are merely experiments, and it’s even likely that the bulk of concept art is merely sketch work. It’s impractical to polish it all up to “museum” quality. I’d go so far as to say that we’re not likely to be seeing even 5% of the concept art created for the project, and what we do see is likely a cross section of varied types. Straight up production pipeline concept art almost never makes it out of the studio.

Even if it did, probably only 10% of that art is faithfully represented in the game down the most granular details. There are simply too many compromises to be made in the translation from fine art to game art, most especially in the change from 2D to 3D.

If the final game art can match the mood and spirit of the concept art, evoking the emotional response that the devs desire, even if some details are lost along the way, the production is successful.

Think of concept art as a sketch (as so many of them are for various reasons; trying to make the equivalent of the Mona Lisa in 3D would be ridiculous in game production schedules) that must be translated into something playable. Much as the translation from vision (or reality) to painting can lose fidelity, translation from concept to final can lose fidelity.

The best production pipelines don’t sweat that loss. They find the most important parts of the art and focus on those and let the unnecessary bits go. That’s where the artistry and skill comes in on the production floor, making the choices on where to spend time. That’s where the art director steps in and keeps both the concept artists from going too far and the production artists from endlessly chasing miniscule details. It’s a matter of scope… and a LOT of game devs don’t do that well, to be blunt. That’s an article in itself, though.

To be sure, it’s possible to take concept art and match it to a high degree of fidelity, as in a Pixar movie, but the practical realities of game production mean that a different approach is necessary. (This, both in the production cost and the lack of prerendering… Pixar’s 30 frames per second are typically rendered over the course of DAYS, while we have to render in real time.)

So, concept doesn’t always translate directly to the final product, for a number of reasons.  I also intend to dig into this a bit more with Yet Another Artisty Article over at the official NinjaBee blog… but I’m not done with that article yet, and it might need to fit into the promotion schedule.  I’ll mention it when it comes up.  There’s only so much detail I can get into for a variety of reasons, but I really want to show a bit more about how things go from concept art to final game asset.

Until that time, then…