Itemization and the Visual Contract

It was an epiphany in polygons and pixels. I stopped dead, my current task forgotten. Blooms of cascading sparkles were bursting from a semi-transparent globe floating just above head height, and falling in glittering arcs to the turf below it. In what was a visually dull game it was a shock of colour and extravagance.

I was transfixed, with the rest of the newbie players now gathering beneath it in awe. Every now and then it moved at an alarming rate for a short distance, and stopped again. We ran, sluggish in our low level lethargy to catch up to its new position, unsure if it was dangerous to do so but unable to stop ourselves. Suddenly from somewhere near the pulsing globe vibrant blue notes began appearing and floated away, dancing to a magical melody beyond our hearing.

"What the heck is that!" I typed in breathless chat to my in game friend who was with me on his low level alt.

"It's a high level bard," he said. "That's her runsong effect. You get it at level 51. It levitates you, makes you invisible, and gets faster with each level. It's the fastest runspeed buff in the game."

The sparkles stopped abruptly. A bold silver and blue plate-mailed figure popped into view where the globe had been, and dropped sharply to the ground. I could see the blue notes were emanating from an ornate sword in the bard's offhand. I had never seen a weapon with a particle effect. I didn't know they existed. Why didn't anyone tell me this?

"That's your class's epic weapon," my friend replied to my feverish questions. "The Singing Short Sword."

We get a Singing Short Sword!?

That was it. I didn't know what was required. I didn't care how long it took. All I knew from that moment was that all my effort, all my play, would focus on one thing. I had found a clear, compelling purpose: I was getting that sword. I was not going to stop until I had it. And I didn't, though it took 18 months.

In RPGs, the desire for more powerful items is a large part of what drives players to play. Itemization is one of the most important aspects of these games; as important as combat systems. Yet it is a strangely regular dissonance that powerful items sport appearances less impressive than lesser items. There have been times when I have looked at the high level prize in my character's main weapon hand and had the thrill of winning it reduced by a bemused and unfavourable comparison to the one it was replacing.

Reflecting upon my experience with hysteria-inducing blue notes, I would wonder why this was ever allowed to happen.

The importance of item appearance was driven home to me years ago, ironically not by the many RPGs I played but by a short stint in a totally unrelated genre. There is a virtual online 3D world built around the entire premise that people will pay to look better. The game was precursor to the unexpectedly successful Second Life, and offered essentially the same thing - a 3D social experience. Players formed chat groups by standing next to each other and then talking in chat bubbles. If you weren't in the group, you couldn't read the conversation. Players could buy anything from beach buggies to condominiums, and spend time hanging out in them with online friends, but it was the chat groups that gave the interaction genuine currency.

The thing that stuck with me after my brief time in the game was not the (for its time) bizarre notion of playing out an interactive soap opera, but how the game created an environment where looks really mattered. I felt like a teenager again, worrying over what brand of jeans I was wearing. Did I look as good as the others. Did I fit in. Wearing default anything branded you immediately as either a newbie, or - which was much worse - not cool. The better looking the clothes and skins you wore, the more likely you were to get attention and be accepted by other players; the more likely you were to be socially elevated.

Those improved looks weren't free of course. The better clothes were expensive, and cost serious ingame cash (called "Therebucks"). This in turn was purchased in bulk with real money. It was a brilliantly intuitive leveraging of the simple idea that people care about how they appear, even in a virtual world - perhaps especially in a virtual world.

Now no-one needs to tell RPG players that what their characters look like matters. We were giddy with excitement the first time a game displayed equipped items on our avatars. Suddenly that +1 sword felt real. There it was, resplendent in our character's hand. See? Look, it moves with me as I turn. How cool is that. Conversely, the first time we proudly equipped a new acquisition only to have it look less impressive than the default item it replaced, something small inside us died. "This just should never be," it breathed in disbelief as it fell spiraling into the sucking abyss of our disappointment.

I remember the first time it happened to me. I rotated my avatar, trying to come to terms with its new reduced look. I swapped the old weapon in and out with the new one, comparing them ruefully. I double checked stat improvements, confirming that I couldn't live without the upgrade. I was more powerful, but looked less. What a letdown. Who the heck designed that?

There are many possible schemes for a graduation of item appearance to avoid this catastrophe, from the simple 4-staged full costume switches in the original Diablo, to increasingly detailed individualized docking meshes in modern MMO systems. But one thing holds true - the more powerful, the rarer, the more special, the more difficult to obtain is an item, the more spectacular or unique its appearance must be.

This is an implicit visual contract all players intuitively expect. The schema can be broken and played with, but only after it has been established. Any break with it then elegantly confers narrative and gameplay cues. A plain looking dagger that is unexpectedly potent gains attention by being uniquely undesirable; an extravagantly ornate staff that is otherwise useless becomes an oddity. But such devices aren't possible if there isn't first a contract of visual grading.

Every aspect of items is involved in this contract. Mesh detail, texture detail and resolution, animation, shaders, environment maps, particle effects, colour. All elements are visual cues that reflect item function and power. These are gameplay money, and should never be used on artistic whim. It must be clear that if it looks better, it *is* better.

Part of the reason some games have lost their way with itemization is a recently mandated design tenet that the first minutes of a game are the most crucial. If players are not intrigued enough by their initial experience, we are told, chances are they won't continue playing. Now whether you buy into and want to propagate such a philosophy of attention deficit game design or not, it has meant many games give some of the best visuals away at their beginnings. This mass market aspiration may give the first few minutes more impact in isolation, but it does so at the expense of the rest of the game. It's a brash and inelegant non-solution that soon comes back to spoil things.

Item power and visuals are competing in the same space as all other game effects. Once you've hit maximum visual volume, there is nowhere to go, despite Nigel Tufnel's wisdom. When it comes to visuals, it doesn't go up to 11. It stays at 10 and bleeds out laterally, blurring into a tacky sideshow mess of meaningless, unreadable noise.

It sounds like I'm talking about art distinctives, but I'm really talking about design. If you overload visuals, you reduce the value of all visual effects including those on items, and so reduce the power of the language you are trying to establish. This subtley undermines all item value because it sends the message that visual effects don't really mean that much. In this respect, less really is more.

The answer is to provide understandable - and consistent - graduation. Players will see powerful items, intuitively know what they are, long for them not just because they are powerful but because they look powerful, and love you for it.

The alternative is cranky RPGers, and no-one wants that.