I think Alpha Centauri is pretty close to a true strategy game. There you got all the elements of Civilization plus flexible unit customization.

One of the biggest strategic elements missing from almost every strategy game is the idea of "supply lines". In many games, once you build a tank, you can send it anywhere without worrying about its fuel and ammo. In a game that requires supply lines, the player would have to worry about managing his empire's economy properly to suit the militaty needs. Escorts to supply lines would have to be made. A successful attack on enemy supply lines would mean that strong enemy units on the other end would be rendered useless quickly. It would be no longer enough to build a few dozen tank units and send them rolling over enemy empire, cause if supply lines are stretched too far those tanks may not even reach the enemy.

I guess most games don't use this idea cause it makes war much more tedious.

Reply to Dauntless:
You are absolutely right, of course it's better to think of the game as a whole before jumping into detail. The problem is that its hard to imagine the entire game without keeping in mind some sort of specific details. At the start of the game making process, you get some vague idea for a new game. Then you try to play that game in yor mind, you immidiately come up with some specific details for the game. You write those down and then try to imagine the game as a whole again, keeping in mind these new specific ideas. Every time you write down a specific detail you can better imagine the game as a whole. And every time you add a new specific idea, the game as a whole changes slightly. So it's like evolution of ideas in your mind. That's how I look at it anyway

By the way, I'm interested in writing design document for a strategy game, if any of you want to help me then send me an email.

Edited by - berserk on August 1, 2001 7:21:48 PM

There was a US Civil wargame that was pretty basic but did do strategy and tactics. You started with your armies and made strategic decisions - I got the South taking NY once - but when contact with the enemy was made your armies scaled down to tactical units and you could do elementary tactics.

A real RTS game would be divided into different phases. You would see the world / space map and have your objectives, resources etc.. laid out. You then devise your strategy and allocate troops. You make strategic moves. Then the second phase kicks in and it goes tactical with each army or navy.

The difficulty is that say you have 10 armies. You would have to work out the tactical battle for each before going back to the next strategic move. That''s perhaps why you don''t see many real RTS''s about.

WarMage...you have a point.

quote: Original post by Dauntless
I started the topic as a "Holistic" thread because I think it is a common practice to go from the specifics to the general, but I''m thinking that it''s probably better to look at the generalities first, and then start nailing down the specifics last. I think this is a better strategy becase you never know what sort of interdependencies might crop up. If you are working on a little "black box" (whether it be your graphics engine, your unit design, client server model or whatever) that''s all well and good....but black boxes need to have well defined interfaces for other objects to call on. And you won''t know what interfaces you''ll need unless you have a general idea of what will be needed for the game design.

That''s why I suggested the macro envisionment of the game first, THEN get down to the nitty gritty. Of course, the fun is in the details so we want to jump to it first. But I think in the long run it may cause for backpeddling to take into account certain things that hadn''t been forseen.

I think the big picture does get looked at first. (You decide to build a MMORPG where everyone plays a car. Or it''s an underwater RTS with mermaids and dolphin riders. Etc.) But I think this VERY high level design often gets handed of to programming without the proper intermediate level work. Programming is NATURALLY very low level. So, you get unit based code being developed.

To make Dauntless happy I think we need a LOT of time spent by the game design team to really work out everything in the game. They are the only ones who can take a macro to micro approach. We also need some innovation because I don''t think the kind of strategy Dauntless envisions exists yet. Someone has to figure out how to make it fun to play "Sim-Admiral" or "Sim-General".

SIDENOTE: I have also heard that micro to macro approaches can create wonderful synergy between units. I think that would be hard to force from a top down approach (although it is admittedly "lucky" when it happens bottom up).

All these ideas require a higher investment in activities before the programming starts, and that probably scares most producers. Deciding not to code for a month while you do game design probably sounds like it will cost a lot more - although it could actually bring the schedule in...

I still see the big questions as: "How do you create a game that allows for fun strategy - not just tactics?" AND "Will extensive macro to micro design create something with strategic considerations as a critical element of the game?"

I think there''s a difference between big picture and looking at things holistically. Looking at things from a macro level requires looking at the "big picture" but it''s not enough.

What do I mean? Well, take bodybuilding for example. The big picture is, "I need to make my muscles bigger". You know that in order to get bigger muscles, you need to workout and lift weights. Doing high-intensity low-rep exercises builds muscle the fastest. That in turn is what will give me bigger muscles. Then if I work on each bodypart in turn, this will make me more muscular.

If you are thinking along those lines, you are still thinking on a micro level. Instead, you have to think about how everything interrelates....such as rest time, the order of working out, what kinds of food to eat, etc. Only when you think about how other factors effect the actual exercising are you truly looking on a macro level. Sure, doing the above example is looking in depth at the main constituent of exercising, but it doesn''t take into consideration the other factors that affect building bigger muscles as well. And trust me....it''s a common trap.

So to me, holistic thinking is looking at how everything relates and affects one another. For example, by taking a look at a game design this in turn will necessitate how the programming will follow. I guess the most succinct way to put this is, don''t focus so much on the end that you fail to see the different means to get there.

If you are thinking of a game that''s about race car driving....what''s the first thing you think about? Probably the car itself. The better car will be the one with the bigger engine, or the better handling, etc. But what about the driver? What about the pit crew? What about the design of the race track itself? I think if you focus too much on the end units and not on the "supporting structures", then much of a flavor of a game gets lost.

To be honest, that''s why I''m not a huge fan of most RTS''s, as they are really not that much more than an adult version of marbles or collecting Matchbox cars. "See, my unit/marble is more cool and effective than your unit/marble." DashZero is right in that the game I''m looking for doesn''t exist as a RTS (although Close Combat comes awfully close).

What I''m really looking forward to is what might be better termed a Command Simulator rather than a strategy game. But I think that what I''m talking about in thinking of things holistically is important to all genres. If you are designing an RPG, is all you are looking at for the "fun" how to increase your stats? If you are making an FPS, is your main design criteria about what neat gun to think of or the cool level with no camping sites? If you are making a RTS, is all you are thinking about is the cool combination of units that can take on anything else? Hopefully people understand what I mean by holistic and will see how to apply it to different genres.