Tales of the Rampant Coyote

Adventures in Indie Gaming!

Picture: Worth 1000 Words, Costs 5000

Posted by Rampant Coyote on October 22, 2012

My favorite death in the original Zork adventure game came from my attempt to brush my teeth using a tube of what appeared (according to the description) to be toothpaste.  Apparently, it was actually “All-Purpose Gunk,” a kind of fast-acting glue. The result was… well, it was actually hilarious, and not entirely unexpected. But one of the fun things about text adventure games in those days was how you could explore the possibility space with the text parser. Sadly, most of the time it only ended in irritation as the game refused to recognize a seemingly obvious approach to solve a problem. But occasionally, particularly Infocom games, your more bizarre actions would be rewarded with some amusing text, or at least recognition that the game was responding to your request.

Is this doable with graphics? Maybe not at the time, with 48k computers, but eventually. The Space Quest games were somewhat notorious with their graphical death-scenes. No doubt each of those death scenes cost considerably more than the quick two or three sentences it took to describe my demise in Zork. Actually, said death scenes were frequently re-used, and included a sentence or two of captions to give the animations a stronger context and more amusing.

But here’s the problem: As graphics get more realistic, we naturally tend to represent all interactions with graphics and animations. But as the realism goes up, so does the expense.

Therefore… interactions are getting more expensive.  ‘Cuz of realistic graphics.

When faced with budgetary concerns (and we always are), game producers / project managers are therefore encouraged to cut interactions.

Without breaking too many rules of predicate logic, we can suggest here that more realistic graphics has had a chilling effect on the level of interactivity. I hesitate to equate “gameplay” (game mechanics) with the level of interactivity – as great gameplay is not necessarily sacrificed as interactivity goes down (often the opposite), but certain kinds of games are potentially negatively affected.  In particular, RPGs and adventure games, which have a greater emphasis on experimentation and exploration than most other genres.

As an example – let’s say there’s a spell that in an RPG causes characters to “hulk out” and go on a rampage. Naturally, you’d want to see that (Heck, I’d LOVE to see Chloe “Hulk Out” in Frayed Knights).  But this would result in serious model changes and new animations for every potential target of the spell in the game.  We then have three choices: Suck up the cost, simplify the representation, or cut the feature. Simplifying the representation might mean doing the old trick of throwing a particle cloud or icon over a character (a la “Berserk” in many Final Fantasy games) to abstractly represent this state. It’d be nowhere near as cool as seeing a realistic hulk-out effect, but may be the best compromise.

Or not. Video games remain a visual medium, and short-changing the visuals is not often a recipe for success, particularly in the highly competitive, hit-driven AAA side of the fence.

Another option that occurred to me just after writing this article (while thinking about the above Final Fantasy shortcut) is to tightly constrain the interaction – reduce it to a (repeatable?) cut-scene. This allows a scaled-down version of the interaction without all the messy variables of different contexts.

I think this is one reason why I’m remain excited about the “retro” trend in indie games. Yeah, sure, they aren’t quite as viscerally satisfying as their more graphically over-the-top cousins, but by being liberated somewhat from the need to show every interaction in exacting detail, they have potential access to a larger library of interactions beyond “Buy,” “Shoot,” and “Destroy.”

Is there a fifth option? More? Another way to compromise?

Filed Under: Adventure Games, Art, Retro - Comments: 6 Comments to Read

  • McFunkypants said,

    I’d propose that the advent of procedural content (simulated art, emergent effects, “morphs” on arbitrary meshes and shaders, generated sounds, physics controlled animations) you might be able to actually pull off this “wide spectum of possibilities”.

    Imagine the hulk effect, for example, being just a vertex shader that “fattens” a mesh, while physics morphers could add more friction to sim anims to make her “shuffle” when walking.

    Done right, we wouldn’t need to create separate art assets for an infinite variety of emergent visuals.

  • John Evans said,

    Fallen London. StoryNexus. Failbetter Games has tackled this problem head on.

  • @rje said,

    Procedural content can help, but it certainly doesn’t completely eliminate costs.

    Things that come to mind:
    * Testing/iterating until you find the right procedural effects
    * Trying the effects against a wide variety of input
    * Fixing edge cases that don’t work as intended (e.g. – ‘the female elf rogue with armor x, y, z has horrible clipping issues with the hulk effect’)

  • Chevluh said,

    While it doesn’t eliminate costs, it can efficiently prevent their rise. As Chris Hecker put it, the automatic texture unwrapper they had for Spore did a worse job than a human by far, but it did work 36000 times faster. Maybe the female elf rogue doesn’t matter when it’s one of a thousand characters, and the iteration will be rewarded in the long run by automation.

  • Robert said,

    I often see this argument coming up in the endless debate of “nostalgia” versus “progression” (I’m not saying that this is one, just that I encounter it a lot in that context) and I feel it is quite a onesided argument. I do agree with what you say, I just feel it’s not the whole story.

    I feel that visuals, most notably 3d visuals in an ‘open world’, can provide a lot of both experimentatino and exploration, if done right. The costs will still be there, but I’m not sure if it should go hand-in-hand with a loss of interactivity.

    #1: Experimentation: Physics, destructable terrain. In the easiest form, this should liven up combat by making enemies defeatable by environmental conditions. Fire, falling boulders, breaking floors. This is something way harder to pull off in the case of Zork as you can ‘create ‘ events that the developers haven’t thought of. Working with lighting, 3d environments, water, sounds, there is a lot more ways to ‘evade’ combat. These are just low-level possibilites, you can go further if you think about it up front.

    #2 Exploration: While a ‘new-school’ world might not have the same volume as an ‘old-school’ world, (compare Skyrim to Arena), you can work a lot on ‘quality’. I’m not talking pure visuals here. Let’s just think about hidden doors. It used to be ‘invisible’ (press space at every wall) or semi-obvious (think Dungeon Master). There’s a lot more subtlety possible on hiding it. Or let’s think about flying/swimming, there’s a lot of possibilitiess still open there. Multi-level dungeons (beside the discrete ‘levels’ of a dungeon), you can play a lot with visual cues to make exploring literally more-dimensional.

    In short, I feel that right now we are quite constrained by following the evolution of videogames, indeed cutting down on interactivity/features, because we are walking in a straight path. I believe there is a lot more possible – with the current technology- then is apparent now. So for that reason, I wholly agree with you and believe that the trend of going retro is good. It might remind developers of cool things they thought of that weren’t possible back then, but by cutting down on the ‘new-school’ complexity, can be done now.

  • Adorna said,

    I’ve been thinking about art styles and retro visuals a while ago. Partly because most very detailed 3D art doesn’t seem to attract and resonate with me like some of the older low res 2D art does, but on the other hand, I tend t long for higher res art in older games. (and some of the current android rpgs)(personally, I found Corpse Party a lot more scary and engaging than Silent Hill)
    In fine art, quality graphics aren’t usually about photo realism. Having some parts of an image more detailed than others gives emphasis and meaning to an image sometimes.
    I’d really like to see what you can do with game art, if you’d go for a reduced look that works with textures. I wouldn’t really know how to go about it, though.

    I’ve deviated from the topic, sorry. It just came to my mind again, because I keep thinking that if we’re no longer restricted to low res 8 bit graphics by computer power abd the highres graphics don’t work well for every game, maybe the answer isn’t to reduce the game to fit the graphics but look for graphics that work for the game.