[PIZOIG main page]


Gaming for yet another new millenium!

See also: [pizoig assets] [AI topics, etc] On this page: {Introduction} {Speed Concerns} {Real/Time Modifications} {Game Philosophy} (towards a philo of what games are) {Behaviour} {Links}


To me gaming is much more than "just playing games". With the advent of our more and more fragmented world, the role that games can play will (i think) grow to one of great signficance. This file addresses many of the issues that are of signifcance to this dolphin. In the words of the monolith (as relayed in scribe Arthur C. Clarke's "2010: Oddysey Two" Use thse worlds in peace.


In the same way that we can use a skybox to speed up the way that a large place is rendered (eg, using a skybox to fill in the skyline of New York City, so as to concentrate on the action at street level), we should be driven to load things into the system as needed. This is reflected in the concept of "JIT" (Just In Time) which was developed in the last century by Japanese firms that couldn't afford totally parallel development and deployment of systems. (In the competitive behavioural model: If you can't keep up with the big dogs, you fall behind; but, if you haven't got it then fake it). As such, systems should seem full-blown and immersive, but we all know the limitations imposed on us: Even when more powerful processors become avail (Sony PS-3 will have something like 4 or 5 separate processors), we will just in turn turn up the heat and make things more extensive. For this reason, all systems design should be able to at run-time load as needed (or at least pre-load) parts of the Back/Ground system or parts that seem likely to need to come into play soon. This idea has been around a long time and is most commonly known thru MicroSoft's DLL concept. That is, we can add modules to an existing system and then tell the ADMIN part of the system about them. Thus, the game, it's universe (and now the characters and their behaviours) can be upgraded *while* the game is either in play out there. Again this goes back to "upgrades are available, do you want to load them?" concept.

Real-time Modifications

Naturally, with Aritifical/Intellegence (AI) programming, we can only "go so far" when we upgrade a character or its abilities (or indeed the plot, b/g etc). One approach has been (like Doom3) to include a dynamic programming Inter/Face that allows us to create/modify the game. In many AI systems that are based on the LISP (list processing) programming language, they include an embeded lisp processor (similar to the embedded Java engine in browsers, cell phones, etc). Thus, *at run time*, we can bring in not only new data modules, but programs and data structures and modify the way that the AI works completely dynamically. But. Should we? When a player gets used to how a particular character or thing behaves, and we "upgrade" it, then this modifies the whole nature of the game.


One essential aspect is (big drum role here) is: randomness vs predictability. In game play and important aspect is the nature of the player's behaviour (refer to [
Assets]). One of the obvious predictable behaviours is of course that of "habit". Examples of this are Arthur Dent ordering tea on the HOG -- and thus throwing the HOG's computer into total chaos. Thus: computer can't not grok "tea" (any more than Slarti can sass this "cheese" of which you speak) result: chaos An even more apropos eg is in Robert Sheckly' short story "fools mate" where a battle computer is fooled by a devious strategy. At the other end is the "Pelig" character in Philip K. Dick's "Solar Lottery" wherein a random plan IS the plan to evade detection (and hence prediction & capture) by telepaths. Thus: total randomness (at times even lacking ANY goal) crashes the TP network. Finally, the most random element is (no drum roll, loan eagle's cry from far above) is: The mysterious stranger. One of the clearest examples is in Sergio Leone's "The Good, The Bad, and The Ugly". This is further explored in Spielberg's 2nd Jurasic Park ("Hammond expected this. So, he had a backup plan. [what?] Me."). To model each of these requires a bit of or thurmatergic cupidity (or at least a bit of aracand legerdermainitry ;). When we see this for the hero's journey, most (if not all) of the rules of play are rendered useless (refer to the 1-gig flash drive containing the "Sword of one million faces" ?? in the World WarCraft ep in SouthPark). Thus, we are faced with the rule: The stronger you make the hero, (protagonist) the stonger you have to make the enemy. (antagonist) eg: Star Trek TNG: Warp 9.99 (routinely), vaster knowledge, etc gives rise to: "Q", "Borg", etc. In terms of gaming as model toward extending intellegence (and not just towards pureile enterainment), our goal is provide a path towards an evolving set of capabilities in gaming -- both in its technology as well as its goals. Player prediction.. [Intro Game Dev . com]

Game Philosophy

One thing must be stressed: The role of games as ways of dealing with the world will grow in the comming years. Games (currently) are of much the same ilk and rarely mimic real life (not that there is anything wrong with that). As far as i can see games can go in these areas: 1) Murder mysteries (a particular fav of mine). 2) Heist Adventures (ditto). 3) SF adventures (double ditto). For the most part (from what little i know/see) most of the game world is shoot-em-up westerns (mod'd for the new age, doan tchuh knogh?). As such, the need for not only more realistic characters and such (as well as interactions and the env's within which things happen -- eg, the real world (ie, rendered), created worlds (ie, the logic of the game under play), as well as dream worlds (ie, dreams happening inside the game under play), false worlds (eg, the "floor 13" world created, or better yet the "holodeck within a holodeck within a holodeck" in the second "My Dear Data" episode of Star Trek TNG), etc.


Intro Game Dev . com]