January 24, 2009

Design Lessons from Visual Arts Jobs

I've learned several important design lessons from my work in various non-design jobs. See the full introduction, here.

Job: Graphic Designer

I did some freelance design work to pick up some extra cash in college, for department newsletters and things like that.
1 - Your job is to give people what they don't know they need
"I'll know it when I see it" is the worst thing to ever hear from a client, whether the job is in graphic design, art, writing, or game design.

A large part of this problem is that the customer doesn't speak your language like another designer or an art director would. Teasing out what they want can take months of iterations, and involves time-consuming tactics like making many mockups each time for them to choose between.

Spend more time up front getting a list of goals and requirements before working on a rough draft. Refine that draft before you waste time on a polished version that may be rejected, and constantly reiterate your goals. Don't ask a client if they like what you've done, ask them how well you think the deliverable achieves the goals you've laid out together. You'll get much more useful feedback.
2 - Don't give clients details to focus on too early
Clients with no vocabulary to discuss design details will often latch onto the first thing they can articulate about the design that they dislike. This often takes the form of an inane detail that becomes a huge distraction for them.

Clients will fixate on the font choice, a word selection, or a paper color, when really there's something about the design as a whole they dislike but can't articulate.

If you have a really difficult client, don't put in any details at all until you get them to agree on the rough layout. Then go away for awhile and insert all the details at once, making sure all the details are in service of the goals you've agreed upon.
3 - The client wants to feel in control
Talk difficult clients through what you're going to do, and make them think it was all their idea.

When you bring back the final perfect deliverable, insecure clients will always have one tiny piece of feedback that they want you to implement. This will always happen, no matter how much they like the design, and it's just because they want to put their mark on it, and assert themselves over you.

Usually these details will be so harmless that you can just put them in, and everyone will walk away happy.

Job: 3D Artist

I spent several years making 3D models and textures for game engines and virtual reality, as well as some 2D interface work.
4 - Your work must be usable by your customers
I use the word "customer" here to refer to any person who has to utilize the work you've made, which may be a player of your game, or another person working on the game who needs to use your assets.

A modeler who makes beautiful characters that don't deform well when animated isn't a good modeler. By the same token, an engineer who writes code that no other engineers can interact with or debug isn't a good engineer, and a designer who designs games that normal people can't understand isn't a good designer.
5 - Versatility is as important as skill
A mediocre artist may be able to draw a nice picture in one very specific style or medium. A master artist can represent the same subject with skill in any style or medium.

In game development, there are many ways to solve almost every problem, and we need to constantly evaluate which solution is the best for the goals we're trying to achieve.
6 - Never add details onto a rushed structure
Good artists aren't in a hurry to make their work look finished. They've learned the counterintuitive lesson that fine tuning a sketch for a few minutes is much more useful than trying to correct a problem once the details are already laid in:


Granted, if you have photoshop you can just select the eye and move it, so it's a bad analogy. But imagine if you had to copy and paste every pixel individually and by hand, and you'll be a bit closer to what fixing this sort of problem is like in game development.

Game designers need to make sure to iterate on a single map, character class, enemy group, or power set until it's right before making any more content that might be based on a flawed concept. Updating the entire game after the fact will take weeks for something that could have been corrected in the test content and designed into the rest of the game with little to no extra effort.

It's incredibly tempting for teams to start making tons of content when the first batch isn't even working right yet. It never saves time, and you'll always regret it. Be the voice of reason.
7 - Iterate on paper, when it's still free
All the great artists of history have created sketches or studies of some sort before a major work. Yes, even Picasso. For the cost of a pencil and some paper, Michelangelo made sure he wasn't going to waste a 15 ton block of marble.

By spending an extra month in the design phase, you can save yourself a year of production time. Think of all the extra content you could make in that year, if you spend more time getting your initial game design nailed down.

No comments: