Wednesday, October 18, 2006

Toolz

A couple unrelated blog posts I read got me thinking about tools.

1. Clint Hocking posts some commentary on AiLive's "LiveMove" motion capture app for designing motion input for the Wii. The video below is worth watching to get an idea.



It's basically motion capture using the actual player input device for the capture, and then some kind of matching algorithm to determine if player input matches the original sample.

Way easier than, say, force feedback programming which I tried my hand at (shudder) a decade ago.

2. And now for something complely different.

Reading a post about writing (sorry, I forgot where), I got pointed to Jer's Novel Writer, a simple niche-targeted word processor. (mac only, doh!)

While it doesn't have all the features of Word, it does have some stuff specifically aimed at those that are writing a novel. Some examples include integration with a database in which to store information about characters, their roles, histories and other references in the text (e.g. you add Susan in chapter 12, and are reminded that she was last seen in chapter 2, where you killed her off), maintenance of a separate outline (switch back and forth between filling in the detail or changing the high level flow), and margin notes for everything from sentences to fix up later to possible flaws in the narrative ("has bob already met susan before this point?").

For what it's worth, the author of the tool is also a 5 time winner of NaNoWriMo - another "Jam" type of challenge - this one to create a 175 page novel in 30 days.

Anyhow, this point isn't so much about writing novels, but rather, to pose the following question:

What niche-specific tools are we missing for this "next generation" of games?

The LiveMove tool is one example.

However, with all the talk of better, deeper narrative and player involvement, what do those tools for designers need to look like, and do they already exist?

When people speak of more beleivable characters, it's generally spoken in the context of shaders, walking animations, and physics simulation of bouncing bosoms. Not that these aren't important, but what about making the character's behaviors and place within the game more beleivable?

For example, if there's a state machine for a given character's AI, does that state machine grow/shrink or otherwise change throughout the course of the narrative (if there is a narrative)? If so, are we back to "explain to the programmer to hook this up to make it happen like this", or do the tools designers use need to start incorporating this kind of stuff.

Maybe people are already tackling this kind of thing. I sure hope so.

3 comments:

Anonymous said...

You're thinking of what I called a "drama engine" which is a bit nicer than "character AI engine" though not as descriptive. Storytron is such an engine but highly monolithic and inflexible to anything other than Storytron specific content. Facade is another, but comes with high production and content costs (scripting joint behaviors is time consuming, and at least 500 are needed per relationship per act. Hence the three million dollar budget for "The Party").

Drama Princess is a modular character engine developed by Tale of Tales, but from what I've read and spoken with its creator about, it seems limited, particularly in the lack of any language. Which is okay, I doubt they're trying to license it.

Enter my company. We're planning on developing an engine once we've stabalized our revenue with a casual game release (its slightly more innovative than the typical match fare, but nothing revolutionary). I'm currently considering how to make a casual game with it, but the key is it depends on being tied to the verbs of a game engine. Then a writer/designer (or programmer/designer, what have you) can spend a deal of time scripting, testing and tuning individual, algorithmically distinct characters. It could probably work for most genres and a range of interaction models, we'll see, if we can make a profitable game with it that'll be enough, but we'd try to bring it out from there I suppose.

BTW, I think this is right up your alley.

Andy said...

The best part of the tool is the ease of integrating the capability- and that is critical- almost as critical as price (ie lots of middleware guys dead on the side of the road who thought non-core middleware elements could go for 40k per game) - There was a company out of Beaverton that had a similar speech tool for command and control- they were thinking more about medical applications than gaming. What was REALLY impressive about the tool was the ease of integration from a development standpoint. It was also really cheap, but wisely never pursued the gaming market in favor of the medical niche.

I don't see command and control via camera, mic or flapping wing controller gaining ground. Other speech vendors have tried and not had wild success either, same with the camera- which I think might have more success with a fitness sort of market-

What I do see as a cool capability in the realm of AI is enabling users to access a game through different devices (or different price points on the same device)- but not in the same way. In other words- I'd like to come into someone else’s FPS as a rodent, snake or swarm of bees and wreak hell on players- in other words an entirely supplemental business model for games and an entire element of AI that the programmer didn't have to think about.

and yeah- now that I have some time on my hands, I'm going to have some fun looking at the game market from another vantage point

I don't see command and control via camera, mic or flapping wing controller gaining ground. Other speech vendors have tried and not had wild success either- however I think you were in the room when someone REALLY influential in the industry said the next big thing would be new ways to interact with the game.

What I do see as a cool capability in the realm of AI is enabling users to access a game through different devices (or different price points on the same device)- but not in the same way. In other words- I'd like to come into someone elses FPS as a rodent, snake or swarm of bees and wreak hell on players- in other words an entirely supplemental business model for games and an entire element of AI that the programmer didn't have to think about.

and yeah- now that I have some time on my hands, I'm going to have some fun looking at the game market from another vantage point

Andy said...

sorry about that C&P double error- I cut and paste that from another window- since the blog comment capability sucks.