In games such as Starcraft Brood War, players at the start of the game don't have as much to do compared to later on when they have to manage multiple battles and multiple bases at once. This means at the start of the game, the gameplay is fairly dull, whereas later in the game it becomes so frantic that only extremely high APM players can play competently.

This is especially a problem in Total Annihilation where being able to quickly react to incoming enemy bombers allows for the minimization of damage from the bombing run, which means a player who can react faster has an advantage over a player with a slower reaction time.

In general, for any RTS game with a fog of war such that when enemy units are discovered, fast reactions would grant an advantage, this is a problem.

I want to create a game which not only removes micro but also removes any requirements for fast reactions throughout the game. In other words the gameplay experience should be "flat" without any spikes in activity, and there should be no surprises possible.

Players should not be sitting around waiting for something to happen and then have to frantically try to click as fast as possible once that something happens. In other words the game should feel more like a slow and steady tug of war than paintball. There are such games in other genres, such as fighting games where, ignoring the power-up bar, gameplay does not change much at all over time.

In my opinion, this means that all changes perceived by the players must be a constant stream throughout the game (to avoid spikes in gameplay) as well as that players must have plenty of time to react to each change, and that reacting faster to the change gives no advantage.

How can I go about designing this kind of game? I can't even think of any RTS game that comes remotely close to this.

  • 2
    \$\begingroup\$ I recommend taking a look at AI War: Fleet Command. Chris and Keith have spent years refining the game to try and reduce micro and make the whole game have a consistent curve. It isn't perfect, but it's a good attempt. They're working on a fully moddable sequel right now. Full disclosure: while I'm not involved in this game, I have done work for them before. \$\endgroup\$ Commented Aug 9, 2017 at 21:14
  • 27
    \$\begingroup\$ any real-time game will hinge on the reaction time of the players. pretty much by definition. The only way to eliminate reaction time as a factor is to suspend the real-time from the gameplay and go turn based, either cyclic or simultaneous. \$\endgroup\$
    – Baldrickk
    Commented Aug 10, 2017 at 7:59
  • 4
    \$\begingroup\$ Smart casting AI for spellcasting units would be of great benefit to reduce the load on a player. It always annoyed the hell out of me in Warcraft 3 that I had to tell my units when and where to cast their spells. I seriously hope You will finish Your game, I've been waiting for something like this since like... forever. \$\endgroup\$
    – Maciej
    Commented Aug 10, 2017 at 9:58
  • 4
    \$\begingroup\$ Have you played Gratitious Space Battles? Mixing GSB style pre-planned combat with RTS-style base and unit construction might offer something close to what you suggest? \$\endgroup\$ Commented Aug 10, 2017 at 11:06
  • 3
    \$\begingroup\$ Games like Baldur's Gate, Ai War or Crusader King allow you to pause the game whenever you want and give orders while the game is paused, but the actions are resolved in real time. Like that you have all the time needed to make your decisions. Obviously it only really work for one player games. \$\endgroup\$
    – Kolaru
    Commented Aug 10, 2017 at 13:22

17 Answers 17


I would like to challenge the framing of the primary thrust of this question: the notion that "flat" design is a good idea.

Game design is really not that different from plot structure in narrative entertainment (movies, TV shows, books, etc). In a plot, things start off simple. Then, a complication occurs. Then a couple more. More conflict is added. This builds and builds until finally you reach a climactic encounter, followed by the resolution.

How fast such dynamics in plot structure change over the course of a story represents the "pacing" of a story. In a fast-paced story, things change quickly and dynamically, with new complications happening sometimes faster than you can react to. In a slower-paced story, complications are added more deliberately.

But in pretty much every work of fiction, the plot gets more complex with time, not less. Or at the very least, the complexity is frequently changing. New characters are added, other characters are removed. New villains crop up, or new plans are revealed. Etc.

And this works in game design as well. At the start of a game, you have a few tools to tackle challenges. Over the course of the game, you either gain new abilities, or you learn how to use your existing toolset to deal with greater and more complex challenges. And so forth and so on, until it builds to a climactic encounter that (presumably) tests a great many of your acquired/understood skills.

A game of StarCraft works much the same way. You start off small and slow. Over the course of the game, you gain new abilities and powers, and you choose to employ them in a variety of ways. You have skirmishes, then major fights for territory. You're forced to expand, lest you fall behind. At some point, you have a decisive encounter which decides who wins.

This is how it works, and this is a good thing. Having a game play out in a well-paced way allows players to follow through on that narrative progression. A slow build over the course of play, gradually testing their abilities until at last there is the climax.

StarCraft is fast paced, such that it's essentially impossible to feel comfortable about anything. At all times, there is something more you could be doing, or something you ought to be worried about. That "drinking from a firehose" effect is exactly how SC is supposed to work, and that is due to the game's pacing.

Even board games like Chess and Go have pacing. People often talk about phases of the game, because as the game progresses, it becomes a very different game. And that's a good thing; that's one of the reasons why these games have been around for as long as they have.

Making a game "flat" would be analogous to a plot that doesn't change. Where it's the same plot from the beginning of the movie/book/TV show to the end. There's no build-up, no increasing complexity, no climactic encounter. It's just two people looking at each other and maybe swinging a sword every now and then.

It would get incredibly boring. You have to switch things up, change the dynamics, move from one idea to the next. Your plot/game design must advance or it is stagnant. And advancement means adding complexity, changing and reinterpreting the dynamics of play.

Change is just as necessary for game design as it is for plot structure. To throw that away, to discard pacing as a game design tool, is to effectively throw away what people are interested in.

There are such games in other genres, such as fighting games where, ignoring the power-up bar, gameplay does not change much at all over time.

This seems to show that what you're not interested in is not "flat" play, but something else. Fighting games very much do change over time. They have dynamic evolution of play. Positioning matters in a fighting game. If you're in the corner, you lose the ability to control distance effectively. If you're knocked down, you have to deal with all kinds of stuff your opponent can do to you.

The gameplay of fighting games very much does change. What fighting games do not have that RTS's do have is this: permanence.

In an RTS game, you win by denying your opponent the ability to play the game anymore. And I don't mean in the sense that the game declares you the winner. I mean that your every minor victory denies something to your opponent. You destroy some of their units, so the time spent building them is lost. You destroy a base of theirs, so the time and resources spent on them is lost.


You may be able to rebuild an army you lost, but it won't be that particular army that you just lost. Much like Chess; you may be able to promote a pawn, but they still lost whatever thing they replaced. In an RTS, you are fighting for your fundamental ability to affect the game.

In a fighting game, you're fighting for points. A round ends when one side scores enough "points" to defeat the other.

Because of this, every advantage in fighting games is entirely transitory, of the moment. Vortex setups on wake up can still be beaten if you guess right. Being pinned in the corner is a situation you can fight out of. And once you do get out of the bad positioning, you're back to essentially a neutral game (mechanically speaking. Mentally speaking, you're in a very different place).

Worst case, you lose a round. Yet another "point" you've lost, but it's just another point. You restart the next round back at neutral positions, with (more or less) the same situation you started the last round in.

Even damaging your opponent is a temporary victory. They go into hitstun and move backwards, giving you the potential to combo them. But after that, they're either still standing or on their back, a state that they'll recover from quickly enough.

In a fighting game, advantages are temporary. That's what gives you the sense that their gameplay is "flat", that it doesn't change. It does change; it simply resets quickly.

But the reason for this difference is that RTS games are primarily and mechanically about growth. You build stuff, whether its units, buildings, or both. You move out onto the map and gain territory. And so forth. You lose the game when your opponent destroys your ability to grow anymore (more or less).

If you want to make an RTS more like a fighting game, you have to make it less like... an RTS. You have to take away growth. You have to be able to make advantages transitory. You have to make victory work based on a more abstract system, rather than some battlefield condition of still being able to participate.

  • 8
    \$\begingroup\$ While I see the merit in this answer, it is is still a "don't do X" answer to a "how do I do X" question. As far as I understand, the OP wants to avoid the problem of a more casual player being overwhelmed by sheer APM requirements. In SC style of games it is not like there was an up-and-down of "plot"/action, but it is consistently only ever "up". So most SC games seem mostly about who hits their APM limit first. Avoiding that seems to be the question. \$\endgroup\$
    – AnoE
    Commented Aug 10, 2017 at 21:14
  • 3
    \$\begingroup\$ @AnoE: On the RPG.SE site, there's a thing called a "frame challenge". That's essentially what I'm doing here. "In SC style of games it is not like there was an up-and-down of "plot"/action, but it is consistently only ever "up"." Nonsense. Typically, after an engagement, the primary action changes to micro battling to macroing. Or to building a retaliatory strike. Or expanding (which you probably should have been doing while attacking). There are many directions to move in; the game simply forces movement. \$\endgroup\$ Commented Aug 10, 2017 at 21:40
  • 2
    \$\begingroup\$ @NicolBolas, as I said, you present your case well. Nut I'm not convinced there is a need to change the frame here - I feel like the OP has pretty much stated what he wants to achieve, and it is a good goal for a particular type of player. On the SC issue: I meant the online, human-vs-human version, where I clearly recall a strictly increasing stress level on my side, for every game. You are 100% correct for games against the A.I., but against humans, I always felt that way. I wasn't particularly slow, but every game felt like an APM check. \$\endgroup\$
    – AnoE
    Commented Aug 10, 2017 at 22:27
  • 1
    \$\begingroup\$ Does anyone have any data to confirm whether APM increases over your average game of SC2, etc? This analysis suggests that there's a clear stratification of player APM by SC2 League (the higher ranked you are on average the better your APM). But I haven't got anything to confirm how APM varies or does not within a game. \$\endgroup\$
    – user83633
    Commented Aug 11, 2017 at 9:47
  • 1
    \$\begingroup\$ @PeterCordes: But that doesn't fit with the fighting game analogy. You need lots more concentration when you're under pressure in the corner than when you're at neutral positions. \$\endgroup\$ Commented Aug 13, 2017 at 18:11

Very good advice in comments (which got moved into chat). I'll just add, since I'm in the middle of implementing such "flattening" into my RTS (Knights Province) right now.

Classic strategies are unlike many other kinds of games:
They deliver too big of a change from 1st to last minute. In other games the main “gameplay element” is evenly spread along all the game with slight variations. In each session you essentially play the same thing - driving a car, running and fighting mobs, jumping on platforms, etc. Essentially you can restart and dive into the same kind of action within a dozen of seconds. Unlike that, in RTS you slowly build up and need to pass through several stages to get to the “fun” late-game (where you are reaping products of your earlier development successes). If you hit “restart” you are back to square one (collecting ore with a single peon, waiting).

On the other hand, the game needs to give player some time to get up to speed (which is why race cars take time to accelerate. FPS games start in “safe” rooms, etc.) Typically that’s ~10 of seconds (depending on game’s complexity).

Two general pathways are:
- Allow to quickly get past the boring stage (e.g. by prebuilding core town or speeding up the game dynamically). Actually, it is already scratched with giving more serf and builders at the start. Having more of them greatly speeds up the early game! I don’t want to invest into this approach any deeper though. This is a “wider instead of deeper” solution. It sounds good for multiplayer - we can let players engage in MP activity faster. Yet it’s a half measure. Second options sounds much more like fun.
- Enrich initial boring stages with new gameplay elements. It is more favorable for single-player gaming, where player can control himself amount of time to engage with the game. Make those elements go away towards the late-game, to be replaced with expanded town-management and PvP fighting.

Player needs to engage with something. Early-game lacks enemies. Some wild nature elements, enemies that player can fight on his “conditions”, that will prepare him for mid-late-game brawls. Can be seen as guarded spots or natural obstacles. Wolves, bears, boars, etc. These could be single animals or packs. Could be even spawning points (dens, etc.)

Proposed solution is to add more mid-late-game elements into early-game. Give players some basic force (army), some wildlife enemies, bonuses for exploring terrain, control points on terrain (encounters, prizes, fights, etc.). All of this will naturally decline towards late-game as no unexplored/unconquered terrain is left (wildlife destroyed or players get superior fighting forces or something else). Thus flattening the experience.

Also take note on keeping late-game less overwhelming.

  • \$\begingroup\$ In order to keep the threat the same, wouldn't the wildlife have to automatically attack the player at random times? And in order to be comparable to a player, the attack needs to be oftentimes a sustained "push" into the player territory as well, right? \$\endgroup\$ Commented Aug 9, 2017 at 12:08
  • 1
    \$\begingroup\$ That largely depends on the game, as RTS is quite broad genre. In my case - that's a Settlers-like RTS where late-game is not quite frantic as e.g. Starctaft BW. \$\endgroup\$
    – Kromster
    Commented Aug 9, 2017 at 12:23
  • 2
    \$\begingroup\$ @user2108462 As the game designer, you make the rules: wildlife animals could be too scared to attack when the player has enough army or there is enough activity in their base, and flee the territory. Or the incoming attacks from the enemies include "cleaning the wildlife" because it's a nuisance for their troops. \$\endgroup\$
    – Vaillancourt
    Commented Aug 9, 2017 at 12:24
  • 11
    \$\begingroup\$ As Civ has already been mentioned elsewhere, I'd like to mention that this generally describes the barbarian tribes that can be found in many of the Civilization games. They are very weak, but at the beginning of the game when your resources are more limited and weaker, they can pose a real threat - similarly, the reward for defeating them is small enough that it wouldn't matter later in the game, but early on it's quite lucrative. They don't respawn, so by the time they'd be pointless they've (usually) all been wiped out. I'd never appreciated how they help to engage player early on - thanks! \$\endgroup\$
    – A C
    Commented Aug 9, 2017 at 14:16
  • 9
    \$\begingroup\$ Warcraft 3 has "creeps" to keep you busy (gaining xp and items) on times you don't feel like attacking your actual enemy. Creep camps have varying difficulty so they are very important map element throughout the match, naturally more difficult ones give more xp and better items. \$\endgroup\$
    – NikoNyrh
    Commented Aug 9, 2017 at 14:18

Ideas for removing micro from the game :

  • No units at all (at least physically present in the game) : you win by spreading your control over areas like in the Go game.
  • You have units but you can't control them, they automatically go attack some critical point of the enemy. This could work like the minions in League of Legends by following a line or as in Starcraft when you do an Attack move (it will find its way to the critical point and attack automatically every enemy unit it crosses by). If you really wish to control them you could make it only possible to choose the building to attack.

Ideas for making progression "flat" :

  • You can't gain new abilities during a game (like a new building to build, or the ability to launch a nuke) so that you will always have to deal with the same actions throughout the game and make the best of them.
  • Some kind of evolutive gameplay where your actions change at some point of the game (think of it as a new age in Age of Empire) so that the gameplay evolves but not the number of actions available.
  • If you still want to control units you could create a maximum unit number and make it pretty low so that you can reach it very soon and it won't change in late game.
  • 2
    \$\begingroup\$ I was going to suggest this - your gameplay can evolve into higher levels of abstraction, such as in Spore; as you progress in the game, the early-game concepts are abstracted away to give way to later-game concepts \$\endgroup\$
    – Joe
    Commented Aug 9, 2017 at 16:12
  • 6
    \$\begingroup\$ Remember NetStorm? All attack/defend units are buildings. You can't move them, you can't control their attack. You win by expanding your control over enemy territory. \$\endgroup\$
    – Exerion
    Commented Aug 11, 2017 at 8:48
  • \$\begingroup\$ @Exerion Yes it is exactly the idea ! Great example \$\endgroup\$
    – Shashimee
    Commented Aug 11, 2017 at 8:53
  • \$\begingroup\$ Holy cow, I hadn't thought about NetStorm in forever. Need to see about getting that action running again. It was great and a perfect tug-of-war example. However, actions-per-minute pretty much won that game. \$\endgroup\$
    – Patrick M
    Commented Aug 14, 2017 at 22:53

Fair warning: if you don't find this sort of game play in the RTS world, that means there's a good chance that your market of gamers don't find value in this sort of game. Always be careful with the "if you build it they will come" argument.

What makes a non-twitchy RTS tricky is the limited interface that we have to our computer. It's very good at making well timed decisive actions. It's not so good at providing nuanced feedback. If you have a non-twitchy game, it's going to have to have nuances or it won't keep the attention of anyone. Nobody wants to play a murky game of tic-tac-toe at a slow pace.

One question might be to ponder whether you actually want to make an RTS at all. The "real time" part of the game encourages tactics to appear beneath the strategy, and tactics are always time critical. You might want to make a turn based game instead.

Or perhaps you want to make a hybrid. Perhaps you can make a slow strategic game in which armies enter "battles" instead of actually fighting each other. Then, on their own time, players join those battles and fight in them RTS style. This would feel similar to how RPGs often have an overworld which then enters a "fighting mode" when you run into a monster.

I'd also point out that Poker meets most of your criteria, especially when played face-to-face. You can't gain an advantage from betting faster than your opponent, so you have all the time in the world to study their face and try to figure out what they have without revealing what you have.

  • 1
    \$\begingroup\$ The Total War series features a hybrid RTS style similar to what you mention. You control buildings and armies from a turn based overworld, and fight battles in a RTS mode. A good strategist would enter the realtime part with an army advantage, but a good tactician might be able to win without it. \$\endgroup\$
    – Kys
    Commented Aug 9, 2017 at 20:59
  • 2
    \$\begingroup\$ Downvoting for the initial comment about "your market of gamers" not finding value in that sort of game. Only if there were examples of that style game and they were market failures would this be a viable argument, and even then I'd argue those games could always have been made better in some other way. Game design is a pretty subjective art, but I would argue that many games fail with great design and also that there are many that succeed with poor design. \$\endgroup\$
    – livingtech
    Commented Aug 10, 2017 at 17:40
  • 1
    \$\begingroup\$ @livingtech Oh, believe you me. I used the phrase "there's a good chance" instead of saying "there wont be a market" because there's always the possibility of the roll of the dice. Go out there and roll the dice! However, I have found it's pretty helpful to remember that lots of people do this for a living, and if you can't find a single example of the gameplay you want to have to serve as inspiration, that should serve as a warning sign that the dice are probably loaded. Of course, if your goal is "go big or go home," that may be a perfect opportunity to try market-making on. \$\endgroup\$
    – Cort Ammon
    Commented Aug 10, 2017 at 17:44
  • 3
    \$\begingroup\$ If one does want to pursue those roads, it's best not to do it in a vacuum. Rather than blindly assuming there's no issues with your game idea, ask yourself "why haven't I found this game on the shelves yet?" Learn from the environment. Make a real judgement call: is this an opportunity that everybody else missed out on, or is it a quagmire that everyone else ran away from. \$\endgroup\$
    – Cort Ammon
    Commented Aug 10, 2017 at 18:25

Have an action budget

A straightforward approach to reducing the value of having lots of action per minute is to directly limit the number of actions one can take per minute.

For example, a player may have an energy meter that constantly refills (possibly at a rate depending on how empty it is), and each type of action like constructing a building or directing a unit group or manually controlling a single spellcaster consumes energy.

So that it doesn't feel artificial, the game will likely need to be designed around it — e.g. arrange things so that:

  • You are able to burn through all of your energy as soon as you get it to get a meaningful economic advantage over someone who keeps it more topped off
  • You can burn energy to make an attack more effective
  • You are less able to respond effectively to attack with low energy

so that you get a metagame around budgeting your energy

Have more high level commands

A big reason for APM being important in a game like Starcraft is that you can't issue orders like "My marines should form a nice firing arc before engaging" or "My high templar should psi storm groups of mutalisks" — in fact, Starcraft's spellcaster system is practically designed around rewarding reaction times and APM.

Instead, you want to design combat and abilities so that they can take effective actions with few inputs, so that there is simply less reason to constantly micromanage everything. And ideally, so that you can't gain a significant advantage through large numbers of low-level inputs. (e.g. the energy budget would help with that)

Of course, you still need to make sure that you still have enough depth that the player choices are still significant and meaningful.

Make preparation trump manual execution

Arrange combat so that most of the attacker's effort is spent in preparing an attack, rather than directly overseeing combat and taking constant action. This would have effects such as:

  • It would mean that fast reaction times are less important for carrying out an attack
  • You could design in a delay between when the attacker is done with his preparations and when the attack actually strikes, allowing you to alert the defenders

I would say the same about preparing a defense, but that sounds like a harder thing to achieve.


Most RTS games provide you with automation tools to automate some menial tasks, but in many cases these are implemented poorly, as to give you a disadvantage. If you made sure that these systems don't penalize someone for using them then that would go a long way to alleviate your problem.

Problem 1: Queues should not consume resources. Most RTS games allow you to queue somewhere between 5-20 commands given to a unit or building. But resource costs are immediately applied. Therefore you have a significant advantage if you keep your queues low, yet your buildings productive. In other words, you need precise timing. If you allow some slack by only charging the cost when a unit actually starts production, it helps a lot.

Warlords Battlecry is an example of an RTS game that did this when you turned on continuous production.

Doing so can result in some problems when a player doesn't have enough resources to initiate production. Of course, as resource income is near-continuous (happens in small chunks), such a thing can only happen if you build too many production buildings. Some games with useful queues also resolve this issue in varying ways:

One solution employed by Total Annihilation is to deduct resources not when starting to build a unit, but in a continuous fashion while building a unit. E.g. while buliding a 50 cost unit over a timespan of 25 seconds, 2 would be deducted every second. If not enough resources are available, the unit construction 'skips' a tick: the unit takes longer to build. The end result is that the inefficiency from having slightly more production than your resource income is very small; units are delayed by a fraction of their production time. Compare this to the 'plain' queue where every unit queued is a unit you could have had a full build time earlier.

A second method is to simply cancel the production, and leave a message exactly -what- got cancelled.

A third method sometimes employed is that the start of the production is delayed until the player has enough resources, sometimes also leaving a message or indicating it by flashing the resource icon in the player's display of resource totals.

Problem 2: Queuing should be available when a command can be macroed by a macro keyboard; there's no decision making involved. If an ability is essentially always automatically casted, then pre-queuing should always be possible. E.g. Queens in starcraft are a notorious example of something that breaks this rule.

Problem 3: Low line of sight. A lot of the problems could be resolved if you reduce the surprise factor. In most RTS, fog of war is so low that units can fire from it to hit yours. That means hyperactively constantly tabbing through units becomes the norm. If you make sure warning systems are cheap and plentiful enough you can increase the "lead time" a surprised player has: do you have some time where acting quicker has little impact, or does every 16 milliseconds count? Other small details that can help are:

  1. Make enemy units provide line of sight to them and appear on the minimap when they use an offensive ability.
  2. Make long range projectiles slow.
  3. Make sure line of sight is available beyond projectile range. Total Annihilation Kingdoms is an example game that had many flying scout units available that could give you plenty of warning.

Ken Bourassa's answer also illustrates an important point: fight duration. If a fight is decided in a second (like in starcraft) the faster player just plain wins unless outnumbered 2:1. Slowing down the pace of combat, reducing the number of manual abilities or having sensible automatic casting of them, having units that can deal near-optimal damage without too much manual control helps a lot too. Age of Empires 2's Mangonels is a great example of a unit that fares really poorly without manual control. It has to be manually fired or the opponent can easily exploit it to kill your own army.

  • 2
    \$\begingroup\$ Making unit queues 'free' has it's own issue - what happens when you are low on resources? What happens is that really important things like workers get delayed because you didn't have 50 minerals on hand when the previous worker rolled off the production line. This is better than TA/Supcom where you can have a dozen units stall halfway through production, but you still trade one balancing act for another. Now you are trying to simultaneously keep your resources as low as possible while always having 50 minerals ready. \$\endgroup\$ Commented Aug 10, 2017 at 18:37
  • 2
    \$\begingroup\$ @James Hollis: Good question, I'll update my answer with how various games handle this. It's a bit too long for a comment. \$\endgroup\$
    – aphid
    Commented Aug 11, 2017 at 7:15

Tactics win battles. Strategy wins wars.

Either remove, or disincentivise the ability or need to micromanage tactics, whilst allowing and incentivising the player to focus on strategy.

The fact that your units stand under a bomber's path in TA unless you tell them not to is a flaw in the game. They should dodge even if you don't tell them to.

I started making a game a while ago (never finished it because I am lazy/terrible at programming), which was a deliberate attempt to make an a.i. that was much better at the tactics part of the game than a human would be, in virtually all situations.

It was a space naval battle game. While I built the ability for the player to directly control the engines, guns etc, I also had (or at least started to develop) a.i. routines to micromanage those things for you. The idea was the player was the fleet admiral. They shouldn't be firing the guns, they should be issuing orders down the chain of command.

Rather than making micromanaging easier, I intended to discourage the player from micromanaging by making it very hard. Like, impossible.

Firstly, the number of ships to be used was to be very large.

Secondly, I used actual Newtonian physics. Engines provide linear acceleration and rotation based on force x distance from center of the ship. There were no speed limits. Bullets were fired off with a muzzle velocity relative to the velocity of the ship they were fired from. Firing guns forwards propels your ship backwards. Getting hit by a bullet on the front right of your ship both moves and rotates your ship.

There's a reason no-one made space games like this. It's so difficult to control. Even just catching up to a ship and matching speed with them is difficult. So the player would simply click the 'match speed with target at optimum firing range and arc' button and let the helmsman take it from there. The helmsman a.i. would attempt the maneuver, and would report back if they failed "sir, the enemy ship is retreating and they're too fast for us to catch up. Should we abandon the chase? y/n" or "they're too manueverable. I can try to get us a firing arc but we'll be outside of our optimum range y/n."

There would also be default a.i. routines to allow them to react to surprises without input from the player. If they spot an enemy, the a.i. would decide to fight or flee, accelerate to a speed that allows evasive manuevers etc.

I believe this game would have met your criteria:

  • At the start, with only a small number of ships, the player would take direct control over more. As their fleet grew in number, they would delegate more. There is just as much micro at the start of the game as the end.
  • People who are good at micromanaging would have little or no advantage over people who are poor at it. The a.i. is likely to be better than either person and the game decided by who is the best at strategy.
  • Surprises should be handled competently by the a.i. at least in the short term. This then gives the player time to decide whether to go aid the ship that's been ambushed or if it's worth losing the pawn to focus on the checkmate.


There are a few approaches to minimizing the impact of actions-per-minute on the outcome of the battle. Other answers have touched on these:

  1. Removing micromanagement by only allowing you to issue macro commands. You issue objective-based orders and the units fulfill them. The downside to this is that AI is often remarkably dumb and frustrating for the players and very difficult to get right for the game maker.

  2. Reduce the overall scope of the battle. Several people have suggested limiting the number of units allowed in the game and indeed many games do this (Starcraft supply limits with variable supply cost per troops, Warcraft upkeep and supply limits, Total Annihilation unit limits).

    Several answers go over how to maximize player control by grouping e.g. your production buildings together.

    What hasn't been covered so far is self-organizing units. Say you build up 20 infantry. In a micro intense game like Starcraft, they behave as a mob, clustering up when pathing, running into each other etc. There is a benefit for treating them as individuals. Instead of the classic ctrl + 1 makes a simplistic group, you could allow the player to organize their forces into platoons, battalions. An infantry brigade with tank support will attack smarter and take fewer casualties. Assigning fighter escorts to a bomber wing automates them to act cohesively and multiply their effectiveness.

    Several games do this sort of thing automatically at the unit level, such as the Total War series. Several games allow you to set unit formations such as a column for assault or spread for taking ranged fire. These are macro tools to enhance your armies effectiveness.

  3. Penalize micromanagement. In reality, no commander has real-time control over all troops. Orders take time to be delivered – force players to get it right the first time. Sending many orders at once is going to confuse and paralyze your soldiers.


The idea of gathering resources and building your army in the field is a conceit of the RTS genre. No military operation in history has trained and equipped soldiers on the field of battle. (Admittedly, history has little to do with game design...) Resources in RTS's are an element of conflict, and you can't remove this conceit without morphing the game from a self-contained battle to a long form campaign.

However, this start-from-nothing approach is the chief culprit to the slow-start portion of the pacing you're talking about.

There are a few successful games that work around this. Age of Empire III was about colonial wars had an element of receiving supplies from your home country. Several long-form campaign games have automated reinforcements where your backup arrives on an timeline that's roughly known before the game begins.

There are also games that have made a go of the pre-game point-buy, where you receive a budget to spend as you see fit. Let your players define their army up front. To prevent it from feeling like an all-in royal rumble, you can allow players to purchase reserves at a discount that scales with the length of time before the reinforcements arrive.

Another option is to put players on a fixed resource income. One reason RTS's get so hectic in the late game is that resource collection has scaled up to two or three times what it was in the early game. If you control the resource allocation and keep it even for both players, that enforces more of a tug-of-war dynamic. This also eliminates one thing to micromanage while the action is occurring.

Tug-of-War Inspiration

One recent game that has found success by combining these elements – fixed resource income, no micromanagement, pre-game army selection – is Clash Royale. However, the game does have a phase of peak mental concentration. The matches are 3 minutes long with a 1-3 minute sudden death overtime: in the last minute of the regular time, the game provides double resources. This forces you to be twice as active for the rest of the match.

Clash Royale has developed a competitive E-Sports scene, helped generously along by the developer putting together cash money tournaments. People who play or comment on the game at a competitive level describe the matches in terms of 'pushes' – a player builds up a bevy of units on their side and the opponent will defend and counter-push with their surviving units.

It's a freemium mobile game, so critics complain it doesn't feel so much like a tug-of-war as a wallet-warrior, rock-paper-scissors metagame. But it's undeniably a success, undeniably real-time and arguably an RTS.

But for the best overall example of tug-of-war mechanics, I strongly recommend you look at Ultimate General: Gettysburg and its sequel Ultimate General: Civil War. It has most of the elements I mentioned in the answer: cohesive units at the brigade and battalion level, macro commands that rely on realistic unit-level AI behaviors, delayed implementation of player orders reflecting the realism of 19th century communication limits, scheduled-but-slightly random reinforcement arrivals, and serious penalties for trying to micromanage everything a unit does. It definitely accomplishes a strategic run-up to a battle climax without demanding peaks in actions-per-minute. Every battle I played felt like there was simultaneously too much and not enough I can do to steer the fight to victory.

  • 1
    \$\begingroup\$ Note that even Clash Royale intentionally causes the game to become more hectic at the end of the game by providing double elixir for the last minute. There is something charming about not having linear mental exertion in a game. \$\endgroup\$ Commented Aug 11, 2017 at 19:03
  • 1
    \$\begingroup\$ You're right Thane, I completely glossed over that. Editing it in. \$\endgroup\$
    – Patrick M
    Commented Aug 14, 2017 at 22:30

Making an answer of my comment.

You could remove the parts where the user has to create their base and their army. So when the player starts the game, the base and the army they have is all that they'll get for the rest of their play session.

There could be a way to regenerate these units automatically (a clone bay, in space) and to auto-repair the buildings. And there could be an automatic upgrade program which upgrades the units/building as time go by.

Automate all this stuff.

  • 2
    \$\begingroup\$ That would ensure a smoother level of gameplay, but I'm not sure what the draw would be. Remember, even though "fun" in in the eye of the beholder, there are some basic tenets to making a playable game that is also entertaining. If virtually everything is automated, what is the actual gameplay mechanic? Clicker Hero RTS/TBS? \$\endgroup\$ Commented Aug 9, 2017 at 14:22
  • 2
    \$\begingroup\$ @JesseWilliams I get your point. The OP could try to add automation at different degrees and see what fits their vision of what "should feel more like a slow and steady tug of war". \$\endgroup\$
    – Vaillancourt
    Commented Aug 9, 2017 at 15:43
  • \$\begingroup\$ For examples of this, check out the Blitzkrieg and Sudden Strike series of games. \$\endgroup\$
    – Erik
    Commented Aug 9, 2017 at 16:36
  • \$\begingroup\$ World In Conflict uses a technique where reinforcements and other abilities can be deployed without Base Building. \$\endgroup\$
    – DrMcCleod
    Commented Aug 11, 2017 at 20:23

Regarding an RTS that comes remotely close to this, I recommend experimenting with StarCraft II, archon mode: 2 players control a single side.

If you got a couple friends to experiment with, you can try to figure out which of the tasks you want to offload to the second player to get the kind of steady experience you're looking for, and then for your game you keep only the elements that remain.

In general, to fight micro overload you can implement micro elements that feature diminishing returns with larger army sizes. An example is Stalker Blink micro (teleporting low HP stalkers to the back of the line one by one): In a small army this micro makes your army significantly stronger, while in a large army, where any focused unit dies instantly, it's negligible.

You also want to make macro easier as the game progresses. Allow players to group all factories on a single hotkey instead of requiring them to click through every single one of them. Have a saturation point where you no longer need to build miners and depots.

Another example is the Starcraft II arcade mode: Most versions of Nexus Wars are rather chill, as they feature macro exclusively. A similar concept is used for some competitive tower defense games.

  • 2
    \$\begingroup\$ Nexus Wars, Clash Royale, Tower Defense and WC3 AoS (non-dota) maps are indeed example of games which limit micro to large extent. Still - placing important tower precisely between waves, or timing several building to push in Nexus Wars - micro gives a lot of advantage and thus smarter/faster/pro players will always win over dumber/slower/newbier ones. Otherwise it becomes a game of luck rather than game of skill. \$\endgroup\$
    – c69
    Commented Aug 10, 2017 at 18:58

You might wish to look at games like Galcon and Auralux, in which players direct units from one node to another.

They largely remove the reaction time requirement by removing fog of war and slowing the pace of the game, giving you plenty of time to react to changes in the opponent's strategy.

There's also no great mechanical skill requirement, because the actions are relatively simple (select units, select destination) and the units move slowly (you don't need to issue directions very often).

  • 2
    \$\begingroup\$ Eufloria is another RTS that went this route. \$\endgroup\$
    – DMGregory
    Commented Aug 11, 2017 at 4:52
  • 1
    \$\begingroup\$ @DMGregory And Kingmania precedes it by a couple of years. And there was another 3D *Waterworld*/dieselpunk-esque game, and a space-based 2D one about robots and corporations... Does this genre have a name yet? \$\endgroup\$
    – Dragomok
    Commented Aug 13, 2017 at 10:51

In any RTS, reaction time and APM will always yield an advantage.

Games like Starcraft 2 are notoriously biased toward the "RT" part, and less toward the "S" part of their genre. I believe it's not really possible to really make the experience flat. I don't think it would "work" in a RTS. On the other hand, there are some things you CAN do to shift the bias from "RT" to "S".

One such mechanic you can adjust is the DPS to HP ratio (dps = damage per second, HP = Hit point). If instead of losing half your army in 5 seconds, your units would only lose 10-20% of their HP, you can reposition, bring nearby reinforcement to flank your enemies, etc. Positionning and strategy become much more important and reaction time becomes a lesser component.

That's the main aspect I can think of at the moment... I'll come edit if any other comes to mind.


I'm going to attempt to address the lack of information due to the fog of war, and ways to mitigate that.

Perhaps taking a look at the assumptions around the fog of war. In my experiences with RTS games, the fog of war exists immediately at the start of the game. But perhaps instead, put it to the players to create the fog of war.

I can't recall or find it at the moment, but I believe one of the Red Alert games had a structure that created a black out on the other players map. This was a counter to players being able to launch something like a GPS satellite and have full map vision.

But if you start with the assumption of full map vision, and make it part of the early game to decide whether to build obstructions to that vision or build buildings and units that would be visible to the opponent. You could then cut down on the lack of knowledge in the early game, and provide some more meaningful options as well.

Another assumption in your question seems to be that players will have no idea what the other player is doing. However, in games like Starcraft, some units exist with the intention of being able to scout for information. Or there are some other means of revealing information like scanning from a Terran Command Center.

The problem is that with so much going on, many inexperienced and even experienced players may forget or miss information this way. So another approach might be to make that aspect of the game more approachable for novice players. Perhaps a easier to use scanning ability, or some form of automated spying.


While many answers focused on unit tactics, and others focused on army size one of the primary reasons that the mid-late game in SC II is so intense is because you have to use army tactics and macro. If you want the game to have a more even tempo, then the most obvious thing to do is force the user to manage a non-trivial amount of macro from the beginning. So either give them a large pre-defined base with one or two expansions, or give them enough resources to build out an entire base + expo right at the start.

Another big difference is that the first minutes are spent scouting. Again, you can remove this early-game task by, say, making the fog of war transparent to building placement, so that both sides can see where they are placing buildings, at least at first.

To reduce reaction time, you need to reduce surprise. That means you need to make the visibility range for buildings and units much larger than normal. You should only be able to surprise the enemy on the strategic level, not the tactical one. And, you need to give units/squads the ability to automatically retreat in the face of superior force, so that an attacker needs to actually pinch/trap a force to annihilate them. So you could switch between "Stand Ground" and "Retreat to Nearest Base".

As others have said, to reduce micro, and thus, APM, you need to make the units much smarter, so that humans will generally not be able to out-micro the AI. Whatever micro techniques that humans use (like kiting, walling, etc.), the AI needs to be able to do on its own.


There might be some value (whether for you, or others...) in taking your root question (How can I go about designing this kind of game?) in more of a process-oriented direction. I'll reluctantly lean on the tradition of the war/battle-oriented RTS for simplicity.

  1. Start with a clean slate. I.e., don't default to any traditional RTS idioms. Don't add things like structure construction, unit production, resource gathering, minimaps, fog of war, etc. until you can make a compelling mechanical case for how they contribute to the game experience you want to achieve.

  2. Speaking of the gameplay experience--map out the role you want strategy and/or tactics to play in that experience. When and how can players make strategic and tactical decisions? Think about "when" not just in terms of blocking out any "phases" you might want to see in a match, but also in relation to other game events. To float a few of many possible questions:

    • Is the emphasis on extensive strategic planning before players directly engage, with relatively little ability to correct for oversights once the action heats up?
    • Do you want the majority of the match to be a tense series of strategic posturing/probing/positioning moves followed by a relatively decisive direct engagement?
    • Should the players engage multiple distinct times throughout the match with an emphasis on how their strategies and/or tactics evolve in response to each other?
    • Will most matches require players to adjust their strategy or tactics in response to random game events?
    • Do you want players to hone a fairly distinct individual style over many matches, or do you want every match to require experimentation and triangulation to arrive at appropriate strategy/tactics?
  3. Brainstorm about the kinds of crude conditions/causality from which you might be able to compose a stream of changes or a tug of war that requires what you want from your players. I'll try to model what I mean by brainstorming about conditions that might support the mentioned strategic posturing dance kind of gameplay. You might be able to come up with more than one set, or some mix-and-match options.

    • Big situational advantages/disadvantages (composition, terrain, weather, leadership, morale, supplies, technology, flanking, etc.) that can make engagements very one-sided
    • Significant uncertainty about that situation which can only be resolved by extensive research/spying/scouting/skirmishing (I don't just mean fog-of-war; even a reliable intuitive grasp of how long it takes to research and build sufficient anti-air defenses adds up to a lot of certainty)
    • No or minimal ability to disengage/recover from a poorly-considered engagement
    • Through changing game conditions (weather, resource exhaustion, etc.) and/or the availability of viable strategic/tactical counters, limit the formation of boring/unassailable equilibria.
  4. When you feel like you have outlined some promising conditions and causal relationships that might lead to the gameplay you want, I think you've got enough of a foundation to try to come up with a few different takes on what content and mechanics might make a good fit (and which are likewise incompatible). At this point you've probably got enough information to evaluate whether common RTS idioms are going to support or undermine your goals, but there might also be some value in making yourself come up with a take or two that use none of those tropes.


As other people have mentioned, most people want the pace of a game to change during gameplay. I think what you're really looking for is a way to have an RTS game where APM doesn't matter.

Combat action delays

It's possible to change the micro so that it doesn't favour the high-APM player without removing it entirely. You could introduce a five second delay for the attacker between sending orders to a unit and the unit responding, to give the defender time to respond. That way, the defender has a bit of time to respond to anything the attacker does. The attacker can't respond quickly to the defender's tactics, no matter what their APM is. For the attacker, the game becomes about predicting the defender's tactics and developing tactics that don't depend on fast reactions.

This will encourage hit-and-run strikes against an invading army while it is in your territory. You'll need good AI to let the attacking army resist that, and you'll also want low-range, quick-moving anti-raider units that can stay on the edge of an invading army, but run back towards the bigger units if there's a threat. You should also create holograms of new units being trained, so that the attacker can target them. Make sure that any counter units take more than five seconds to train, or else accept that some units have a few seconds' immunity to attacks.

You could divide the entire map between the two teams from the start, and put a command hub in each map. A unit is attacking when it is moving into enemy territory or attacking an enemy unit which isn't already attacking. That determines whether it gets a delay or not. This could lead to weird fights near territory boundaries, so don't put anything strategically useful near the territory boundaries. Either player should gain a tactical advantage by pulling their units away from the territory boundary before attacking, without losing anything.

When someone orders their unit to move into their opponent's territory, show their opponent a blue hologram of where the unit is moving to, and show the path the unit will take from its current position to the hologram. Show a red hologram of the unit when it's told to attack, as well as a path. In-universe, you could explain this knowledge through analysts with precognitive powers, or delays due to radio intercepts, or oracles or anything. You could explain the delay as communication delays or as time for the soldiers to think about and respond to the order.

Research queues

Another reason that APM is important in RTS games is that responding to a battle can mean you aren't researching upgrades as fast as possible. Interesting decisions about resource allocation are important to RTS games, but you don't want decisions about research to be time-critical or they would provide an advantage to high-APM players who can make research decisions during battles. So let players queue their research from the beginning of the game, and make it easy to change the research queue or put all research on hold.


That RTS's increasing need for concentration is unnecessary, or even that the begining of a game is devoid of it is your assumption.

RTS are like board games like chess or go, except that your opponent's moves are usually hidden from your sight.

Like reactives and products concentrations in chemical reactions, board games have two composants that evolve as time goes :

  • Your possibilities
  • Your planning

At the begining of the game, you have almost nothing, so the variety of your moves is at its minimum of the entire game. Meanwhile, since the game just started and you haven't built anything yet, the variety of moves you can plan is at its maximum. Your strategies aren't yet limited by the options you would have sacrified by prefering others and not knowing what your opponent did - since he hasn't done anything either - don't guide you toward a path in particular.

As time goes, you will make moves, these moves will give you opportunities. The more concrete means of interacting with the game you have, the less you will have room for planning, since each choice you made closed some of those that you didn't choose. Take the example of your first base : Once you've built it, you won't be able to build your first base again. You will certainly build others, but hopefully you will take the positions of your assets into account while deciding where to expand.

The balance in RTS come from that basic relationship, which itself has been exploited for millenias in various pre-electrical strategic games.

If you are interested in that aspect of games, look at Game Theory, that's the name of the mathematical field that study the behaviour of rational agents being able to interact with a system in various ways, which apply to actual games as well as more abstract situations like negociations, conflicts or parking your car.

To answer your question, you could perfectly design a game which flatens the player's need for concentration simply by flatening the moves the player has : If the complexity of the player's opportunities is constant through the game, it's concentration will reach its maximum right after the begining of the game, which will also be its average. That's how FPS work, by the way, and MOBA in a lesser extent.

However, it may be quite hard to apply that to RTS : it would imply that players are already met with various and complicated moves available right at the begining of the game, and that this complexity doesn't change through the game. Strategic games are about setting and using assets on a field in order to reach your objective against one or more opponents with their own objectives that are (often) mutually exclusive with yours.

Reshaping the battlefield to your own purpose so that you get the edge and win is the basis of strategic games. Remove that mechanic and you are stuck with something that looks like a RTS but isn't, like dummy food.

That's, I believe, why you can't think of a single game that works like it, or even how you would do it yourself.

I strongly advise you to reconsider it and ask yourself the good questions, like : Why do you want this, what's the reason ? and Considering that reason, would there be another way of reaching what you want without breaking the game ?

And, by the way, concentration is flat throughout the game. It's just that you focus less and less on planning and more and more on using your assets, but the focus doesn't increase, it just shifts.

  • 1
    \$\begingroup\$ "you could perfectly design a game which flatens the player's need for concentration simply by flatening the moves the player has. That's how FPS work." That's not strictly true. Shooting at a large barn does not demand the same level of concentration as shooting at flying bird-demons who are throwing fireballs at you while dodging their fireballs, even though in both situations the set of moves available to the player are the same. \$\endgroup\$ Commented Aug 11, 2017 at 11:27
  • 1
    \$\begingroup\$ "RTS are like board games like chess or go, except that your opponent's moves are usually hidden from your sight" I have to disagree. The main difference is Real-Time component. Not the fullness of information. \$\endgroup\$
    – Kromster
    Commented Aug 11, 2017 at 11:31
  • 1
    \$\begingroup\$ @Kromster You are right, I think I was comparing with strategic games in general. On the other hand, both are linked. The point of alternating turns doesn't make a lot of sense if you don't have a full knowledge of the field, especially of the opponent's moves. \$\endgroup\$
    – Sarkouille
    Commented Aug 11, 2017 at 12:29
  • 1
    \$\begingroup\$ I have to disagree again. Alternating turns without knowing full game`s state worked just fine in TBS games (e.g. Heroes of Might and Magic). \$\endgroup\$
    – Kromster
    Commented Aug 11, 2017 at 12:38
  • \$\begingroup\$ I don't say it doesn't work, but it's doesn't work well with opponents. \$\endgroup\$
    – Sarkouille
    Commented Aug 11, 2017 at 12:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .