quote:

---

Original post by MadKeithV

What you see is not what beginners see. How long since you were a beginner? Sure, picking up a unit and telling it to "go there" is easy, but getting your active troops to do what you want them to do is a lot harder than getting your static base defenses to do what you want them to do.

---

Hmm, beginner must have been just when i got my hands on wc2 and red alert.

Controlling a mobile unit, will probably always be harder than controlling a tower - in some way or another, unless you lessen the degree of control the user actually has.

quote:

---

You are talking static gameplay though! Moving a mobile unit somewhere useful does not equal active gameplay, that''s just rearranging the static scene of the battle. The movement itself is what sets it apart from static base defense. Having a contingent advancing from one side to distract while another circles a strong point in the defense to infiltrate an area left unwatched because of the distraction is active. Making doing things like this easier is going to improve gameplay for everyone.

When playing becomes more than choosing the places your troops stand (still an important part of course), when the player has an intuitive, easy method for defining the paths, actions and reactions of his units, when he can define how they should react should certain potential situations occur, and when he can do this without constantly worrying about the pathetically idiotic way the AI handles the situations where you CANNOT control the troops, that''s when you take the pressure off the beginner to just do the whole "base-building around resource gathering points" instead of trying something exciting.

Much of the problem lies in the actual controlling of units. You don''t need to control static structures at all. You can concentrate fully on building your units and planning the next extension of your base while your static defenses blow the bejeezus out of everything that comes near. Clicking the "repair" button every now and then is a silly way of trying to force the defender to keep paying attention (what about a "do not repair" button? That''s a condition that will happen much more rarely!). Now, if you can allow the defender to not have to pay so much attention by building a combination of static defenses (which, in my opinion, should have weaknesses as a whole, and not just as individual types of unit), and mobile defenses for which you can issue quite a complex set of orders in an easy-to-understand way, you''re on your way. This way, you can still have a pretty worry-free base defense, but it requires more active effort on the part of the defender than just ploinking down heavy turrets at specific intervals with regular spacings of air defense units.

---

If I can agree here or not depends on what were actually talking about

I agree if were talking about what a RTS should probably be according to its name - a STRATEGY game.
I disagree if were talking about what a RTS currently is - A TACTICS (+ economy) game.

If things like unit micro-management was to be removed from SC/BW (just as an example, as thats the game i played most), you''d end with a boring game for anyone but beginners (and maybe programmer like ppl who like to self-program/customize some AI parts.)

Anyone above casual-gamer level will simply not use these new features (if possible), because he has more confidence in his own skills than that of some AI.
If the AI now does better than a human could do (or it simply is impossible to disable), the non-newbie had about nothing left todo while playing (maybe playing a 2nd game simultaniously ? )

In a "real" RTS the amount of units should probably not be controlable as in a standard RTS, either by an enforced command structure or increasing unit numbers to a degree where noone can hope to control them all.

quote:

---

Make static units a little less attractive, active ones a little more, and the whole experience improves, offense AND defense.

For instance: what about the player that engineers a visual "gap" in his static defenses, but has a carefully planned ambush by hidden mobile units, a scenario programmed by the defensive player to lure an attacking force into the base in such a way that for minimal losses on the defender''s side, he can deal a great blow to the attacker''s resources?

Of course the really seasoned player can do this right now in most games, but it takes a level of control and comfort with the interface, and indeed a level of "twitch" that many casual players (including me) cannot keep up for the majority of a match. I can do it once or twice, but I want to spend most of my time planning what I want to do instead of clicking the mouse around like a madman.

---

The experienced player wont need lots of time to plan what he is going to do next. (Or maybe the actual clicking around doesnt need much thinking time ?) So both can be done at the same time.

If my enemy walks right into the trap i created, i''d rather decide right at that point what i will do now, not plan/"program" ahead for all eventualities that could happen.

Maybe i feel that him running into my trap is just to distract me from his real atack, so i move my forces to where i expect his real attack to come (atleast somewhere he does not expect them.)
I sure could do the same with pre-planned waypoints, but i would still have wasted time on the planning earlier.


In summary:
It''s an issue of catering either for the novice or catering for the one who mastered the game already. Doing both at once might be very hard to impossible or will atleast take up precious development time.

One idea here would be to offer automatisation for the newbie, but allow the experienced player to do everything manually - with better results, if done right.

quote:

---

PS: I know I can be a bit aggressive in my writing style, so I apologise if I have seemed to attack you (it''s not my intention), and don''t let that annoying "moderator" tag stop you from pointing out when I''m wrong. I love a good discussion on a topic I can identify with, and what good is a discussion if everyone agrees!

---

Dont worry, i tend to have an agressive writing style aswell I think, so I know how to take what you''re writing

quote:

---

Original post by Anonymous Poster
If my enemy walks right into the trap i created, i''d rather decide right at that point what i will do now, not plan/"program" ahead for all eventualities that could happen.

---

What I''m worried about is when the enemy walks into the trap while I''m really busy doing something else. Most games that I have played tend to really fan out as the game progresses, with different mini-bases across different parts of the map, and sometimes many fronts to deal with at the same time. Now, individually, I know full well what I want to do on each of those fronts, but I''m not fast enough with my mouse action to do it all, while still keeping track of whats going on, during the course of the entire game.

It would be a relief to know that I can try to organise an offensive, without being 99% sure that the AI will do something incredibly daft on the home front if I am attacked there while leading the battle, and that if I leave the offensive behind to manage the defense, the offensive will be completely annihilated because of more AI stupidity. If I had time and the tools to plan ahead a (little) bit so that at least I can go back and forth reasonably and keep things from a complete meltdown, that would make the game more enjoyable to me.

Please note: this is stuff that I''m still thinking about, and as such doesn''t have a firm opinion driving it.

More and more, I find myself questioning whether or not stalemates are something we want to eliminate. It seems to me that, between players of equal skill and with equal resources, stalemates will arise frequently, if the game''s design allows it.

In chess, for example, it''s fairly accepted that the player controlling the white pieces has a slight advantage because he goes first. So, between two equally skilled players, the one controlling white will tend to win; stalemates are often viewed as a minor sort of victory for black.

Tic-tac-toe is, of course, the classic example of a game that will almost always end in a stalemate (tie) unless someone is just playing poorly. Anyone who''s watched War Games can tell you this.

Stalemates also often occur in history as well (witness: the cold war); people are used to the thought of "the balance of power".

Some games, of course, are set up to not allow for stalemates, or to make them as rare as possible. In classic Risk, turning in cards allowed for players to suddenly produce ever larger armies, which could drastically alter the control of the map in one turn in the later game. Nine-men''s morris is similar to Tic-tac-toe, except it''s designed to prevent stalemates. Most of the game variations in Myth prevented stalemates; in king of the hill, for instance, play continued until one person was the victor (there would always be a victor because there were finite units available, and thus anyone still in the game could attack anyone else).

So, when we look at RTS stalemates, these questions occur: should stalemates be possible? If not, how can they be prevented, and if so, how can they be made less frequent?

Going from history, stalemates should be possible; however, some would argue that that makes a game less fun. Personally, I think that two people playing perfect games (which is impossible) would have to end in a draw, because otherwise, you have an imbalance in your game. The examples I listed above that don''t allow for stalemates generally have some pre-existing condition to favor one player over another (nine-men''s morris favors the first player), or have some semi-random imballance (turning in cards in Risk, etc.) These don''t make sense for an RTS, unless you''re setting up specific campaigns (player A has to defend the fort for x minutes, players B and C must breach his defences), for obvious reasons. So, I believe that stalemates are possible, though we should strive to make them as rare as possible.

So, what are the characteristics of an RTS stalemate? A) both players run out of units that can attack the other player''s base; or B) both players can produce more units than they lose in any given battle. Obviously, option B) only occurs with infinite resources; once a player runs out of resources, they can''t produce more units, and this will break the stalemate. So, option A) is the only one we need to worry about (since every standard RTS map has finite resources). When will it occur? i) When both players final mobile attack units (capable of attacking buildings as well as units) manage to kill each other, or ii) when the last mobile attack units are destroyed by immobile ones. Can these two be prevented? Well, i) can be blocked by not allowing two units to attack simultaneously; this works for "instant attack" melee units, but unfortunately, this prevents things like launching a missile and dying before it hits, or poison attacks. I''m not sure if there''s a solution here. ii) can be prevented by having all immobile units be, in effect, manned (there''s a soldier in the machine-gun bunker, or manning the missile tower), and allowing the player to destroy the immobile defense to "regain" the mobile attack unit it contains (example: bunkers in Starcraft). This may not work with some specific game unit designs, but it does make sense, and so seems to be a viable option.

Ok, that was much longer than I expected, and probably fairly rambling. If you stuck through to the end, though, I thank you; and any comments/suggestions you have, especially on the latter part, are more than welcome.

-Odd the Hermit

quote:

---

Original post by Sandman
I think I''d agree with you there - so we get to the root of the problem which is: Micromanagement.

---

For me it''s the source of most fun atm, replacing it might be a good idea, just removing it not.

(I hate warcraft 3 for it''s auto-cast, amongst other things)

quote:

---

As you say, micromanaging a defensive force is a thousand times easier than micromanaging an offensive one. All your units are in the same place, so no need to wander around looking for them. All your buildings are in one place, so no need to worry about defending isolated outposts and the like. Furthermore, your units rarely need to move - you just concentrate them around the most likely attack points and move them in response to an attack. The rest of the time you can just sit back and watch your factories pump out more units for your final swarm, and watch your enemy (hopefully) fail miserably to breach your impregnable defenses.

---

I see more of a problem in centralized resource gathering, which basically is the reason that turtleing is possible at all (while still keeping chances at winning and not just keeping alive longer).

(Or defense units being simply overpowered, but thats no micro-managment problem either then)

quote:

---

Better controls for setting up aggressive actions would help, provided they don''t go too far and overcomplicate the interface.

---

Agreed, aslong as they dont oversimplify the gameplay either.

quote:

---

Also, more interesting defensive structures might help. For example, what if defensive structures required a power supply? Taking out the power supply would neutralise the defensive structure. A few of games have used this idea (C&C and SC spring to mind) but none of them (to my knowledge) really take full advantage of it. In SC for example, it''s too easy just to make sure that every building has 4+ pylons next to it. What if the generators did massive splash damage when you blew them up? Then you''d have to be a bit more careful how densely you packed them, and where you put them in your base...
What if your defensive structures had to be manned? What if your opponent then captures them from you?

---

Fully agreed here

quote:

---

1. High quality map zooming that clusters together similar units as soon as you''ve zoomed out too much to see individual ones. No matter what the zoom level, you have a good overview of the troops that you can see in that zoom level.

---

Agreed, but in most games the map-size and number of units is not big enough to make that really matter. A quite neccesary feature for a grand-scale RTS tho.

quote:

---

2. Unit type / Unit capability selection criteria. Allow the player to select troops by type when lasso-ing. "Select all the infantry in this area", "select all aerial units".

3. Customizable user interface. Allow the user to set up new buttons for selecting troop types, for instance. "Select all troops in this area belonging to unit 24", "Select all troops with damage level > 50%", and things like that.

4. Gestures instead of waypoint-setting. Just draw a curve for your units to follow somewhere, instead of having to build a path out of straight-line segments. As additions: allow the naming of points on the curve to act as traditional "waypoints" ("gather all troops from unit 11 at waypoint delta", allow soft curves that have some pathfinding in their area so that the curve modifies itself to follow the terrain more logically.

5. Give me GOOD AI, or no AI at all. There is nothing more frustrating than the computer constantly trying hard to ruin your strategy by doing something "smart" to unburden you.

---

Agreed, although i''d like to see such a kind of path-planning in action first, to see if it''s really usable. Definatly good ideas anyway.

quote:

---

6. Simple programmable unit behaviours (i.e. programmable AI). Now that''s something that''s very easy to say and very hard to implement, but it has to be possible. Units have a very solidly defined set of atomic actions they can perform (move, turn, shoot, repair, some more for very specialised units perhaps), and building on that level there has to be some set of higher-level actions and commands that make it easier on the player.

---

If it fits in your gamedesign and you got a very good interface for such a system, this could be interesting to try out.

quote:

---

Why don''t people play Free-For-Alls more often? I have yet to run into many of the bordem causing afflictions in FFAs.

---

I think FFA games are very static in many occassions.
Early agression can kill you, without much gain if you succeed.

In regards to a stalemate in Total Annihilation.. I once played against a friend for four hours on one of the large desert maps. We had the classic stalemate with impenetrable defenses, no matter how many units or nukes were sent. Also, we each had radars all over the map, so we knew where the other was at all times.

That was, until I realized the power of the Radar Jammer. Recognizing that we were in a stalemate, I sent a radar jammer and a construction bot out on a mission to build a Big Bertha about halfway across the map, but still under my air defense control and in an area that was not in a "direct line" between my base and his base. It took a while to get the Big Bertha built with that one Kbot, but knowing that the game was in a stalemate, I was confident that this would be the tiebreaker.

And sure enough, when my bertha was built and started firing rounds into his base, his defenses crumbled and I marched my massive army right in.

Moral of the story: There is no such thing as over-balancing a game. In fact, TA is one of the most balanced RTS games of all time. The bottom line is that it''s the players that cause the stalemate and can ultimately relieve it.

quote:

---

Original post by Anonymous Poster
If things like unit micro-management was to be removed from SC/BW (just as an example, as thats the game i played most), you'd end with a boring game for anyone but beginners (and maybe programmer like ppl who like to self-program/customize some AI parts.)

---

Not sure I agree with you there. However, you would have to add a lot of strategic depth to compensate for the fact that the players are no longer spending so much time trying to get their units to do what you want them to.

In any case, we aren't really trying to remove the micromanagement, just trying to add more tools to make the burden a bit lighter.

One feature I'd like is some kind of sticky grouping policy. How often do you really need to give individual soldiers orders? Not very often. And when you do it's usually to stop the AI from doing something stupid. I'd like to be able to assign units to a squadron, and thereafter, selecting any unit in that squadron selects the whole squadron. None of this faffing around with ctrl+1-9, or having to rubber band the whole group each time I use them. Furthermore, the squadron should move and fight as a coherent unit once it is formed - none of this crap where half the squad goes off towards the target at normal speed, quarter of it goes off really fast and gets decimated by the enemy, and the rest gets stuck behind a tree and wanders off to the other side of the map looking for a way around.

[edited by - Sandman on September 12, 2003 11:53:01 AM]

In response to Odd The Hermit''s thoughts on stalemates.

While two mathematically perfect players shouldn''t have one consistently beat the other, not all balanced games need tend towards stalemate in practice. In modern particle physics, the concept of symmetry breaking is important. Basically, like a roulette wheel, a system which at high energy is symmetric ''drops'' into an asymmetric state at low energy (when the wheel stops the ball is on one number not spread over all of them)

In game terms, I describe this as being balanced ''high'' or ''low'' - ''high'' balanced games, while theoretically possible to keep in a symmetric state tend to fall off into victory for one player very quickly (like Z). ''Low'' balanced games tend to settle into a stalemate situation where any slight advantages tend to disappear rapidly or be overwhelmed by underlying fluctuations (Risk for instance tends to swing in favour of the player whose turn it is as card sets get cashed in). Stalemate-type games tend to end up with one player getting bored and throwing everything into a premature attack, and suffering crippling damage followed swiftly by defeat as the other player presses home the advantage.


On RTS games lack of strategy: one effect that I find particularly debilitating (as a relatively low ''twitch'' player) is that guiding an attack force takes you away from easy building of replacements - meaning that my attacks not only lead to my losing resources in an uneven trade at the enemy base, but also losing tempo in my building/research. Playing defensively, on the other hand, I find I end up sitting around twiddling my thumbs, so there is plenty of ''downtime'' in which I could be exploring if it weren''t for the fact that I expect to lose any troops I leave unguided in the field...

Here in UK, there''s currently a TV series, "Time Commanders" where a team of four contestants replays a historical battle (from the winning side) against a computer opponent and with (potentially) full information on the state of the battlefield. In that, two "generals" oversee the battle by passing orders to their two "lieutenants" who in turn talk to half a dozen computer operators who actually control the troops. Of the two episodes so far, one team ended up all looking at the same things at the same time, and suffered a humiliating defeat, while in the other team, the two generals delegated authority to their lieutenants, focussed on the big picture, had pertinent details and suggestions offered by the lieutenants, and outperformed the historical general despite a few newbie mistakes. Obviously, with current mainstream games, delegating authority to AI isn''t an available option, but it''s something I hope is in the not to distant future. Certainly Dauntless was working on an RTS type game with real strategic potential. Whether he''s managing to produce something that lives up to his design, I have no idea, but it should at least be feasible.

quote:

---

Original post by Sandman
Not sure I agree with you there. However, you would have to add a lot of strategic depth to compensate for the fact that the players are no longer spending so much time trying to get their units to do what you want them to.

---

Isnt that basically what i said ?

If you remove micromanagment, just make sure it is replaced with something else.

quote:

---

One feature I''d like is some kind of sticky grouping policy. How often do you really need to give individual soldiers orders? Not very often. And when you do it''s usually to stop the AI from doing something stupid. I''d like to be able to assign units to a squadron, and thereafter, selecting any unit in that squadron selects the whole squadron. None of this faffing around with ctrl+1-9, or having to rubber band the whole group each time I use them.

---

What happens if you want select units from both groups at some point ?
Double-Clicking or something to achieve this would be nice tho.

quote:

---

Furthermore, the squadron should move and fight as a coherent unit once it is formed - none of this crap where half the squad goes off towards the target at normal speed, quarter of it goes off really fast and gets decimated by the enemy, and the rest gets stuck behind a tree and wanders off to the other side of the map looking for a way around.

---

One part of this is formation-movement, which is already implemented in a few games afaik. (Hopefully with a button to switch it off)
The other part is just good pathfinding.

The squad fighting as a coherent unit would be more than just movement. And i wouldnt like that in a game which is more about tactics than strategy.

Of all the flaws listed on the thread, I think commander omniscience is the most damaging to gameplay. The ability to keep tabs on every unit all the time is a lousy system. Granted, a game in which you send your guys off Lewis&Clark style would be less fun to play, since you'd just be gambling without any real-time intelligence of their actions, but seeing exactly what happens to them can hurt strategic factors.

"Scouts" don't have to come back to give reports, and an ambush only works once before the other player knows exactly where and how big it is. The element of surprise is almost toally removed, and you can send reinforcements to a unit that no real commander would know was in trouble.

Sci-Fi games like starcraft can pass it off with cameras on every marine's suit, but a profound tactical advantage could come from jamming that signal.

Rock, paper, scissors, boxing glove, blinfold.

[edited by - Iron Chef Carnage on September 12, 2003 1:01:16 PM]

also worth mentioning,

Age of Empire, Total Annihilation, and Starcraft are 3 very different playing games ...

The makers of warcraft and starcraft are against defensive play styles, and for that reason, there is no defense that is even remotely effective against a perfectly planned and executed counter attack - and the defenses are expensive, so that even in the early game, choices like cannons or dragoons, bunkers and towers or more siege tanks are very tough to make. These games seem to favor a definate benifit from building a FEW defense, with some sort of logarithmic decline as you add more defenses ... so that a valid defense is always possible only by having your static and mobile units both present. And since these games have relatively low unit caps, it is not possible to have effective defense squads at each base you build ... this leads to a strong value in the axiom "information is ammunition" and the winning player is usually the one who gets wise to his enemies plans earliest ...

Age of Empires is a defensive game, heavily so ... the early towers are all but unbeatable by the early mobile units, and the ability to rush your peons into the building to defend really makes a big difference ... not mainly for their damage, but for their lack of dying (it''s like giving them a 50 fold health increase). So the reason age of empires games are not all turtle like stalemates, is because of the necessary expansion required to increase resource gathering ... without a big expanded base, the defender would always win ... but you cannot build town centers across your entire base, and spread out towers cease to be effective killers once the game have siege weopns and or healing units ... so in this game, the players must scout the enemy early, and usually a good player attacks right at an enemies point of expansion, or on the opposite side, if the enemy has moved his mobile troops to defend the expansion.

Total Annihilation is somewhere between these two, because it is NOT a really defensive game, early on, because of resources limits and the need to grow toward more and more metal deposits, constantly. But, the strength of the defenese and primarily their LACK OF LIMITS - combined with a resource model that never expires, makes spreading more and more defense throughout your base a valid way to create a stalemate inducing position. And then there is the transition to nuclear power, which is the point in the game where most victories are achieved, either by the person who did not advance, and built in force, or if not attacked quickly, by the person who gets nuclear power or more advanced units built earlier ...

so all the games play as a tech OR unit massing race ... but they each allow for different moments of fear / confrontation ... starcraft has a VERY effective rush, so no-one can afford to ignore early units and defenses, whereas AoE has more early defense, so people may rush either for units, or simply for tech and resource gathering, TA has a significant segrigation between normal and advanced buildings and units, and so a defender may be unbeatable, until you take the advanced, or he may spread to thin and fall easily to your mass of level 1 units ...