It would be a hell of a lot of work, but why not give the user a choice?

The game im working on is an RPG, but you can see the same sort of problem there, so what we have done is implemented a way to let the player choose.
1. Turn Based: Old FF1 Style, just in case a player wants to use it.
2. Active Time: FF6-FF10 Like, lots of people like it.
3. Real Time, This Was PHENOMINALLY HARD to implement, as we had to create an entirely seperate battle system and enemy placement, it still has a hell of a lot of bugs and we cant decide wether or not you should be able to kill civilians, and levelling is an issue too

Sandman - you will note that I was referring to "decent" players of an RTS game. If you seriously think that you can tear up a decent player with that kind of "chaotic mass" you''re in for a shock. The strategy is actually getting to that stage - it''s quite difficult.

Dauntless - This is what I originally said :

quote: Germany did in fact have to build all of its factories from scratch. The difference is where you think the war began.

If you say that the war begins as soon as the first offensive is made, then yes germany had some factories built. RTS games start the war before the war factories are built - which is not unreasonable. And Germany certainly did have to build its factories from scratch.

Saying that a game doesn''t model what happens in the real world (which is what you seem to be arguing for) does not say anything about the level of strategy in the game (which is what you seem to imply).

I did actually play a *lot* of Kohan, and it is a great game. But you still have to think fairly quickly, which seems to be the downfall of a lot of those who dislike RTS games.

quote: Original post by Argus
Sandman - you will note that I was referring to "decent" players of an RTS game. If you seriously think that you can tear up a decent player with that kind of "chaotic mass" you''re in for a shock.

I say: "If you can''t beat a decent player with a chaotic mass, then your chaotic mass just isn''t big enough."
Some games have gone some way to countering this by utilising scissors/paper/stone balancing methods, but you can nearly always beat a careful defence by making sure your chaotic mass is as diverse as possible.

Also: your counter-definition of the word ''strategy'' is fair enough as far as the general public is concerned, but in military terms there is a clear distinction between strategy and tactics. Often there is another level, ''operations'' in between the two, although some people call tactics "operational strategy". Either way, one good definition of strategy is "the art of distributing and applying military means to fulfil the ends of policy" . This is compared to tactics which address exactly how these military means are applied. As Carl von Clausewitz said, "Tactics is the art of using troops in battle; strategy is the art of using battles to win the war."

I''m not saying your definition is "wrong" - language means different things to different people. I just believe that as far as game design is concerned, I would think that it serves us best to have clear and precise definitions so that we know what we are talking about. And in this case, a clear distinction between strategic and tactical games would be helpful when discussing them.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost ]

Argus:

I gave an example of how my 'unbeatable' fleet could be defeated. I have no doubt that it is theoretically possible to defeat such a force.

Unfortunately, the reality of the game, is that it is too difficult. Maybe there are a few really 1337 starcraft players who could implement such a defense, and pull off a victory against all odds, but for your average Joe playing the game, he doesn't have a snowballs chance in hell. So if I can beat all but the very best players with an amorphous mass, I would call this a dominant strategy.*

Of course, against most players I would be lucky to get a force even a quarter of that size. This is because the real key to the game is peon pumping, something which I suck at.

I want a strategy game that places good strategic thinking over fast mouse clicking.


Edited by - Sandman on January 14, 2002 8:36:44 PM

Sandman, Kylotan - I realise that a chaotic mass of large enough size can defeat any kind of defence (provided game balance is reasonable), but as I mentioned before, the strategy in those kind of RTS games is actually getting to that point. And believe it or not, there''s more to it than "peon-pumping". I might just as easily sum up strategy in conventional turn-based wargames as "attack the enemy". But I know there''s more to it than that.

As far as my definition of strategy, I was using the term as it is used in game theory. I''m fine with people using "strategy" in the large-scale wargaming sense, but when people say that RTS games have little to no strategy they mean it to imply that there is little thought involved. If that''s the case, then the word is being used in the game theory sense but defended as though it was being used as the wargaming term..

Most RTS games require strategic thought and fast mouse clicking. If you don''t like games where you have to be pretty quick with the mouse, then fair enough, I can''t blame you for not liking the current crop of RTS games. But please don''t insult RTS gamers by claiming they are lacking in strategy.

Argus

I suppose a lot of this is semantics, but yes there is a wargaming strategy (what I really want to see in games) and there is the general definition of strategy. But look at the definition more carefully, and it still seems as though RTS are more about tactical considerations than strategic ones. What do I mean?

Tactics are the specific means used to acheive an end. Strategy in many ways determines what tactics will be used and when. I think most RTS get stuck in the tactical mode. It takes so much time and effort to enact the tactics, that not enough is left for the strategy. I also think that the scope of the strategy is not as much as it could be. If only certain other factors were taken into consideration or if the command interface was different, than more strategy (i.e) gameplay could be had.

I think you are trying to say that RTS do indeed offer a strategic level of thinking. And, yes there is a degree of strategic thinking, but I think it''s not as grand as it could be. I think what I and several of the others here have been trying to say is that the model that RTS use for their strategic implementations can be made better.

Here again, I think is the confusion between a tactical consideration and a strategic one. I''m not sure exactly what your style of strategy gaming is, but I yearn for a game where I don''t have micromanage every unit. Perhaps you feel that I and the others are attacking RTS depth of play...but that is not my major gripe...as I think there are some strategic considerations to be made. MY concern is the means to an end. I think that you can formulate your plans, but the interface to get there is just too tricky. I also think that there are many other considerations that simply aren''t factored into strategy games that I believe would make even the best Korean Starcraft player reconsider how to do things...

For example:

1) morale- units don''t follow orders just because you said so, they may be too afraid to act
2) unit integrity- Army units aren''t just made mish-mash. There is a TO&E that defines how a unit is organized
3) unit cohesion- what happens when elements of a unit get scattered?
4) Logistics- that''s all great that your tanks just blew a hole in the main enemy line...except now they outraced their supply line and are vulnerable
5) Leaders- what happens when your major just took a bullet to the head? Will the Lieutenants rise to the occasion?

And these are things just off the top of my head. To add just these simple little things would make gamedepth and gameplay FAR more interesting. And this isn''t even talking about a better command and control structure that I talked about before.

I think you hit the nail on the head about the pacing of the game. Again, this is tactics to me. While some may enjoy speed-chess, I don''t think the Masters would say this is the best way to play chess. In many ways, I think RTS are like the QuakeIII''s and Turn Based games are like the Rainbow 6''s. I think the time factor takes away from forethought...and this is especially true when applied to wargaming style strategy.

As for the notion that we are attacking RTS gamers for their not thinking strategically, I do not denigrate them, I blame the paradigm of the game. Those elite level players have excelled at what they do, partially from training, and partially because they know how to play within the confines of the game. But just because you are elite level Starcraft player doesn''t mean you will excel at a game like Close Combat, or even Battle Isle. In other words, RTS players have become at what they do because they know the limits of the game.

If my game idea ever came to fruition, I don''t know how many skills learned from RTS would be suitable in my game. My game idea breaks with so many RTS molds, that I think they would be very lost. Again, this is not to say RTS gamers are stupid...many will pick it up quickly, but the point is....the level of strategical thinking required would be deeper and broader at the same time. If you play one FPS, usually you can pick up the nuances of another FPS pretty quickly. However, with what I have in mind...players weaned on RTS games would have to adapt to new skillsets and different ways of thinking (again, I''m not implying that they are stupid, merely that what they have learned in other RTS probably won''t really work with my ideas) So I''m not insulting RTS gamers, in fact, how some of those elite level players can do what they do amazes me. I''m not attacking the gamers, I''m wishing that RTS game designers would break away from the clone mold that they all follow and would start coming up with new ideas to broaden strategic play.

Argus: Dauntless has pretty much said everything I wanted to say. Conventional RTS games are not completely devoid of intelligent thought, but they don't seem to encourage the sort of strategic thinking I am interested in, As I see it, there are two basic classes of 'strategy' in Starcraft, Economic strategies and combat strategies.

Economic strategies seem to me to be the most important, yet to me they are the least interesting. Stuff like, "how many probes should I build before I start building a pylon?" leave it too late and you waste time hanging around for extra psi, do it too early and you waste time hanging around for extra minerals. None of this has anything at all to do with military strategy. If I want to build stuff I'll play sim city.

Combat strategies (perhaps 'tactics' might be a better word) are a little more interesting. However, they are nowhere near as interesting as I would like them to be, and they seem to be more a case of learning to use the right combination of units at the right time. There are some interesting terrain considerations early in the game - good defense of choke points is essential - but once air units arrive these considerations are pretty much unimportant. Furthermore, actually implementing a complex tactical manouevre is simply too difficult for most players - so they have to make do with the good old chaotic mass.

Now I think Starcraft is one of the better RTS games. The interaction between all the units of the three races can be great fun to experiment with, and come up with fantastic new tricks. However, each time I try one of these tricks, the implementation of it proves less efficient than the chaotic mass. So I end up going back to the chaotic mass.

An example of a 'failed tactic' of mine: The hit & run carrier attack.

Problem: protoss carriers are immensely powerful, but they are incredibly slow. Furthermore, they are rather vulnerable to things like ghosts, science vessels, scourges and defilers. On large maps with lots of minerals, they can get wasted very quickly. This tactic was designed so the carrier fleet could launch a sudden attack on a weakly defended section of the map and retreat before heavy resistance arrives.

Execution: Build a suitably strong carrier fleet. 12 carriers is usually enough. Build two squads of 6 scouts, and make sure all squads are accompanied by a couple of observers and at least one arbiter. Make sure you research the arbiter's recall ability.
Use your scouts to locate a weakly defended base. Use the arbiter's recall ability to teleport the carriers in, and let them wreak their havoc. Withdraw the scouts a safe distance. When resistance becomes too strong, use recall again to get the carriers out to safety. It is worth leaving an arbiter hanging around your main base in case you are sneak attacked - this way you can recall the carriers right into the middle of your base.

Practice: Proved too fiddly to actually implement properly. Carriers would drift apart while you are getting ready to teleport them out, so you could often end up only teleporting two or three in at a time. Arbiters would get locked down in the couple of seconds you spend not looking at them, etc. In the end it resulted in losing more units than I would have done if I had just sent in 20 carriers and forgot about all that teleporting stuff.


Edited by - Sandman on January 15, 2002 6:19:25 AM

ROFL I just kinda finds your StarCraft strategies quite amusing... Believe me, I used to play like that (Carriers are good => Build as many carriers as possible), but it doesn''t work in the long run. People will strike at you earlier than that, a quick muta rush or a vulture drop could make sure you don''t get those carriers. Just defending StarCraft here, I guess. :D It _is_ about more than just building that large mass.

On topic, I think it''d be great to have the opporunity to select between real-time and turn based. Although, that would prolly mean a tremendous amount of work for us, the developers. Baldur''s Gate has some kind of weird choice between the two with its auto-pause function, but that hardly makes it turn-based.

The strategy above is aimed at longer games on a large map.

I am perfectly aware of the fact that rushing to carriers leaves you very vulnerable at the early stages of the game - this is one reason why I don't do it. I always try to get a decent basic defense going before I even start building carriers. Perhaps I should have specified this in my description, along with exactly how many probes you need to maintain production, how many stargates you need to get your carriers quickly, what order you should build everything in etc, but I didn't really think that much detail would be necessary. Besides, I never said I was the best starcraft player in the world. I really can't be arsed to learn a hundred different build orders that enable me to get different units ultra quickly, I make it up as I go along. I also fail to see the relevance of your counter argument - the strategy did not fail because I got killed before I could build the carriers, it failed because the successful implementation of the manouvres was too difficult. Perhaps I should have picked a different game, whenever someone criticizes starcraft you always get a few people popping up with some variation of "Oh yEaH stArCrafT R000lZ joo 3 oN 1 BGH n0 zErG RusH ^_^ KEKEKEKEKEKEKEKE!!1!!"

Anyway, the purpose of this game is not to brag about how amazing or crap we are at starcraft, it it to point out how the strategies are abstract. The key to these conventional RTS games is good resource management (peon pumping) and deciding when to attack - (rush early with a weak force, or later with a crushingly powerful one?) Deeper tactical considerations seem largely redundant, unless you excel at micromanagement.

PS: for semantics purposes, replace the word 'strategy' with 'tactic(s)' in this post.


Edited by - Sandman on January 15, 2002 7:59:15 AM

Dauntless - I understand what you''re saying, but the route you take to get there has a few problems.

quote: I''m wishing that RTS game designers would break away from the clone mold that they all follow and would start coming up with new ideas to broaden strategic play.

Believe me some try, but nobody buys their games. Did you buy Kohan? Dark Omen? War Wind? Kohan in particular has all 5 factors that you mention would be nice.

Also, note that grandmasters of chess play under a time limit. We might think of it as turn-based, but chess is a real-time game (it''s just got a real nice interface). As someone mentioned in another thread, turn-based strategy is really just coarse-grained real-time strategy (or real-time is fine-grained turn-based, however you like to put it).

But yes, RTS games need a better interface and new ideas - however designing one is a *lot* harder that it sounds. Just getting to the stage where we are now has taken a decade.

Sandman - You talk about what you believe strategy is in RTS games, yet you can barely play them. Is it surprising that what you claim to be the truth is dubious at best?

quote: The key to these conventional RTS games is good resource management (peon pumping) and deciding when to attack - (rush early with a weak force, or later with a crushingly powerful one?) Deeper tactical considerations seem largely redundant, unless you excel at micromanagement.

It may seem this way to you, but how one divides their forces, where they strike and in what manner all play big parts in RTS games. I can accept that the interface makes things difficult to pull off - it''s annoying when you don''t have the finger speed to pull off the tactics you want I agree. But user interface is the only major problem with RTS games - quite a lot of them have fairly deep strategy. And I''ll tell you why.

Most RTS games are tested very thoroughly for strategic diversity. It took SC over a year to reach strategic equilibrium (people having a rough idea of what strategies to execute and when). This is with around about 30000 people playing at a time, and with very quick games (compared to turn-based). I''ll wager that if you put those kind of numbers through most turn-based games, that game would collapse very quickly into a certain set of strategies.

The sad thing is that everytime I play an RTS I see the same people complaining "there''s no strategy, it''s just click click click". But in a game design forum ? Maybe we should be trying to design a better interface instead ?