SpittingTrashcan-

Firstly, I think your idea has great merit. I do agree with someo of the things said in repsonse, but I also believe that they can be overcome with slight changes in your way of thinking, while still leaving the core of what you want in the game.

MSW-
I think you are worrying to much about little details. Does it really matter exactly how the multitued of actions will be chosen.

And what about different heighted people?? It would probably be easiest if the computer did this automatically. You could create any shaped creature, but as long as it has a head the computer will know where to aim the "head shot" at.

Anyways, I think the idea sounds great, although it would take a lot of effort to get some of the features working. And MSW I am not saying you are completely wrong, just some of your thoughts seem to be picking at very fine details .

Doolwind

DuranStrife,

You''re absolutely right. When a game gives near-unlimited realism, any lapses become just that much more annoying, as they remind you that it''s just a game after all. When a game is severely restrained and rules-based from the get-go you don''t mind how unrealistic it is... so long as it follows its own internal rules!

The hope I have is, through increased realism, to make the interface "transparent": that is as intuitive as possible, so that not only can the player guess from his real world experience what can be done, he can also largely guess how to do it. The net benefit, besides less need for a manual, is the ability to do away with distractingly artificial elements like status bars, huds, and numbers popping up over this that and the other thing.

My model in this aspect is Grim Fandango, which attempted an icon-free interface. Grim was an adventure game in which your character was steered with arrow keys. If he got near something interesting, he would turn his head to look at it; you could then hit keys to make him comment on what he was looking at, manipulate it, or pick it up... you could also make him reach into his jacket and cycle through his inventory; he''d pull each item out in turn. It was very easy to pick up and play, but some of the puzzles were fiendishly difficult. I thought it was absolutely the neatest thing...

Doolwind,

MSW has his points. It''s just a matter of design philosophy. He doesn''t mind if games are games, which is perfectly workable and a good design philosophy; I''d prefer them to be more on the order of simulators, which is significantly more risky and difficult to pull off.

MSW - this might address some of your use concerns. To make the "moves" system more easily usable, it might be a good idea to make some of the moves context-sensitive. One example is the basketball player who would want to switch between offense and defense depending on his possession of the ball. Another is the graffitist, who''d prefer only to spray when there''s a wall in front of him. A wielder of multiple weapons would of course like to have a separate skill set bound to each. The Zelda games on the N64, with their context-sensitive B moves, are a good implementation of this.

Just throwing some more stuff out there.

Biggest problem I see with your idea: Its a simulator. Simulators are fun for people who want to learn how to do something; they''re not so much fun for people who just want to play a game. Its come up before, but I don''t think you quite realize how much of an issue it is. You''re talking about a very complex control scheme (regardless of the specifics).

1) It would seem to me that at the heart of your argument, what you''re really interested in is a combat system that FEELS more realistic.. not necessarily one that IS more realistic.

2) I agree that P&P RPGs are simplified, numerical abstractions of reality due to the inability of humans to handle much more complex systems on their own, and that computers can handle more complexity. But even computers are going to require some level of numerical abstraction of reality, since thats what they deal with -- numbers. The statistics can be hidden, but they''re still there, and some players like seeing that.

Also with regard to P&P RPGs, Neverwinter Nights (which you criticized in your first post) is a computerized interpretation of an existing P&P RPG (which is extremely popular, I should add) -- the mechanics are SUPPOSED to mimic the P&P version.

3) Yes, there are people who will build mods and skins and so forth, and there are people who will build moves for your hypothetical game -- but regardless of how sophisticated the physics engine is, the real issue is that its a lot of work for very little return.

4) Most skills have to be defined for the computer beforehand, because it needs an understanding of specifically how that skill affects the game environment.

I''m sure my comments above don''t address everything you''ve said, but some things to consider perhaps. I was once trying to build a really complex, innovative, flexible combat and skill system; then I realized, that in order for the game to be fun, it needs to be ACCESSIBLE -- not just to those that have practiced for hours upon hours, but also for those who have spent five minutes getting a feel for the game. As one individual mentioned, let the player''s own skill be reflected in combat tactics, not in the act of combat itself. If anything, the combat should be handled MORE by the computer and less by the gamer.

Good ideas, ST, but ultimately probably not a best-seller.. if I wanted to learn how to fight with a sword or perform a roundhouse kick or paint like a pro, I''d go buy myself a sword and join a combat re-enactment club, or get into Karate, or take an art class. But I don''t. Let people play a game, not practice skills they don''t need in a simulator.


Brian Lacy
Smoking Monkey Studios

Comments? Questions? Curious?
brian@smoking-monkey.org

"I create. Therefore I am."

Mr. Lacy,

Good points all. Once again I seem to have misled people with terminology. When I said "simulator" I wanted to conjure up images of such physics engines as the algorithms used by car designers to make safer cars: "crash simulators". I was not attempting to remind you of such games as "flight simulators" in which all the joyless complexity of flying a jumbo jet is flawlessly emulated. Complex physics, simple interface is what I'm going for here.

In response to specific arguments:

1) You damn right. I don't like the abstract turn based combat systems of Japanese CRPGs very much. When I'm in a fight, I want to choose my strategy, line up my shot, and let 'em have it myself. I want to be able to use whatever vicious tricks I can come up with. I want to do it and I want to do it with style. If there was a fighting game that let me customize my weapons and moves, I'd be first in line to buy it. I'm looking to get that Die by the Sword ASAP.

2) Yup, computers are number crunchers not thinking machines. But they can crunch a lot more numbers. The average D&D stat ranges from 1 to 18 (for humans). Are all 15 STR people equally strong? Realistically, no. The average C++ unsigned int is at least 0-64000. Let's use all of that numerical range, is what I'm saying. That number crunching power means more granularity in statistics than human players could possibly handle, with the net result that the jags are evened out almost to a smooth curve. Don't bother telling people that their strength is 35675 - just show them where they are on the bell curve.

Now don't get me wrong about NWN. I love it to death, and I know it was an attempt to port a P&P game accurately to PC, with all its associated stats and rolls. I'm just saying there are plenty of games which take the same approach as NWN. NWN is the apex of one mountain; I'm just trying to climb another one.

3) I say thee nay, my friend. Check this out: One crazy modder. Now tell me that people aren't willing to mod the hell out of games. New moves won't just look cool (as skins do), they'll offer in-game prestige (Nobody can stop the Spinning Cobra Clutch!), economical opportunities (Sure I'll teach you the Seven Strikes of the Monkey. For $5000. Each.), and tactical advantage (What the- *WHUMP*).

4) Which is why the physics engine must be so sophisticated: it must extrapolate the effect of a move from some simple rules, the move itself, and a whole lotta math.

I appreciate the advice on complexity vs. fun. My hope is that by including a large set of pre-built moves, a simple use interface, and a thorough move builder, I'll be able to create a scaling complexity: for the average non-athlete character, just press some buttons and get things done; for the consummate strategist, the widest possible range of options. And recall that I do include a major role for the computer: the player has only to "point at target, press button" in most cases. The player doesn't have to learn how to fence, box, or paint if he doesn't want to; all he has to do is tell his character to learn it, or at worst build it himself or ask someone else to. I hope I've made things more clear.

[edited by - SpittingTrashcan on October 7, 2002 3:46:28 AM]