I''m not sure if "consistency of concept" is the best phrase for what I have in mind, but here''s what I''m thinking... In multi-player games, it seems to be vogue to suggest creating a "universe" where a player can be at "any level" of gameplay. For instance, a war game where "everyone" in the army is being played by a real, live human being. The general who''s ordering entire battlions, the captains, the "looeys", all the way down to the guy with the rocket launcher. First, as described, that "game" already exists. If you wanna play, your local Army/Navy/Air Force/Marine recruiter will be happy to talk to you (just don''t believe a word he says and sign anyway ). Secondly, the same game cannot reasonably contain that many different "levels of detail" operating simultaneously. To the general (or whatever), the scale needs to be *huge*, and because of that, the time-factor of the game needs to be incredibly sped up for the game to be playable. While, at the other extreme, to the grunt, every fraction of a second counts. So if you speed up the time-factor to make the general''s "game" playable, it becomes impossible to be a "grunt", and vice versa. And, finally, to get to the actual *point* of this post, the "game" played by the general is totally different from the "game" played by the grunt. They are logically 2 different games, with only superficial similarities. As such, they should each be their own game. Whether or not you consider a game as "art" or not, is irrelevent, and C.S. Lewis''s assertion (horribly paraphrased) that "anything in a work of art that does not add to that work of art, detracts from it" still applies. If a feature doesn''t "belong" in the game, adding it (or keeping it in there) does not make the game better. Instead, it makes it worse. In our example, we were either trying to patch a tactical game onto a strategic one, or vice versa. Choose what kind of game you''re going to create, and then create that game. Just because something "would be cool" doesn''t mean it has any place in the game, and implementing it only detracts from the time you have to do the *real* game. This where the game design becomes a "discipline". You have to know what''s right for your game, and what''s not, resisting the "really cool" temptations. Of course,this doesn''t mean that elements of different "genres" cannot be successfully mixed. They can, but only when it makes sense in the context of the game. DavidRM
Samu Games

David, in my "infinite-levels-of-abstraction engine" (tm) (see I''ve got to go up against Revolution''s inifinite polygon engine with something...maybe an infinite monkey engine would suffice) I wasn''t suggesting playing more than one role at once, it was a question of being at a level of abstraction and playing the mission at _that_ level, with the next mission being at a slightly higher level. This way you choose the speed and setup that''s right for you.
There was the question of what is feasible with AI to control everything you don''t want to at any given level and I think it''s perfectly possible to simplify the whole gaming world to work at your level. The important aspect is to keep the frame of view of the player as realistic as possible while cutting down on the amount of processing power given to outside events. By frame of view I mean more than just what the player see''s on the monitor and what is off screen. I mean that having a platoon that takes up one pixel because of the scale, working to a simplified set of rules which encompass the game mechanics, the graphics, the AI and the sound.
I''m also wondering where the similarities and differences occur between strategic and tactical games and at what point one melds into the other.
I''m not talking about this from an abstract "what if" viewpoint either. I''m a professional AI games coder (sorry, I still love saying that, it brings a smile to my face just thinking about what I do) working on this kind of design in my spare time to try and work out the logistics and see if it _is_ possible (of course it is possible, it''s just a question of when, how and what level of processing power it would take).
There''s nothing wrong with thinking about writing a standard game based in a fixed genre trying to work in one direction but the thread this post is an offshoot from was talking about next gen games and there''s always going to be a little fantasy and a lot of imagination mixed in with the realistic potential out there.
I''m just trying to get a grasp of the possible.

Mike

I don''t have a problem with a single game, with a single player doing what you describe. That''s kind of a cool concept, really.

It''s when the suggestions become targeted at multi-player games with some guy somewhere being "the old man" and some other guy somewhere else being "the grunt". For squad-based combat, I''m sure that can work. It''s when the "layers of abstraction" between player A and player B become too many that I get dubious as all hell, because now the game is trying to be "all things to all players", and that simply doesn''t work.


DavidRM
Samu Games

well said, David. I agree with you.

Mason McCuskey
Spin Studios - home of Quaternion, 2000 GDC Indie Games Fest Finalist!
www.spin-studios.com

I agree to a point.
It would be an interesting experiment to have several distinctly different viewpoints over a war game.
You can''t speed up or slow down the game but you _could_ have one person giving orders in realtime for a general strategy.
How fun this would be for the ''general'' is questionable, but with thousands of troops they''d probably have their hands full time-wise.
However, I doubt anybody would implement a game like this on a whim but it could be the next executive training toy.

Do the army have systems like this for training tank battalions in simulators?

This is the sort of game I am thinking should be developed. As you say it has so many drawbacks, but the idea of large-scale warfare consisting a mixture of human players and AI opponents is most interesting.

Humans can choose what side they wish to fight for, whether they wish to be a soldier, maybe a tank driver or a helicopter pilot. Then a general can give orders.

It is interesting but I guess unmanageable. Well, not unmanageable but potentially unprofitable. I mean, when you get killed in battle is that it? Not too appealing.

Shame really because I think this grand scale of multi-player interactivity using different roles fighting for your human leader is most interesting.

To make a something as this possible, regarding to gameplay. The "general" role in this case has to be designed to fit the time-scale of the others, and it''s not impossible I believe.

Take for example Starsiege Tribes, which looks like a fps but is a lot more. In Tribes it''s possible for a player to act as a commander (ie "general"). He may hide in the base with a command-laptop and issue orders to the other players out in the field. The other players may set up radar-sensors so the commander can detect incoming enemies etc. When the commander see something that should be done he may assign a waypoint to a player or a group, etc.

This works out... and it''s a little bit like what you''re talking about I think.