Game Design RTS101

Battlefield uncertainty and fog of war

Almost since the inception of games, game designers have known that giving all participants universal and precise access to complete information about a game’s state can be a roadblock to the skill and fun derived from playing. In card games such as War, Rummy, Blackjack and Poker for instance, at least one or more of each player’s cards is known only to that player, thus introducing a level of uncertainty about the relative position each player has in relation to other players. Poker specifically leverages this situational uncertainty to drive tension – a player’s ability to bluff, and to read and predict the reactions of his opponents, is core to being successful at the game – in the terms of modern video games such as StarCraft 2 or Dota, this is typically called a player’s ‘game sense.’

Games both physical and virtual have tried numerous methods for introducing uncertainty regarding complete knowledge of the game state into their mechanics and systems. In the case of tabletop wargames and computer strategy games, this is typically a decision informed by an actual phenomenon. Since the quote provided in the Wikipedia article I just linked is so apropos, I will reproduce it here:

War is the realm of uncertainty; three quarters of the factors on which action in war is based are wrapped in a fog of greater or lesser uncertainty. A sensitive and discriminating judgment is called for; a skilled intelligence to scent out the truth.”

Ultimately, however, it seems that the game designers in real-time strategy games are largely content to refine a model that was conceived over 20 years ago

Uncertainty + Fog of War

th (7)I feel like I need to begin by defining specifically what I mean when I say ‘uncertainty’ in this article. In the context of games, uncertainty tends to signify a randomness of outcome. Dice rolls, card drawing from a pool of face-down cards, et cetera. While interesting, this isn’t what I want to cover right now. I’m specifically speaking about situational uncertainty, about uncertainty related to opposing player’s strengths and actions within a game. Most often, in strategy games, this has typically been achieved through the medium of a ‘fog of war’ enshrouding the battlefield, with units and sometimes abilities able to pierce that fog in a limited area. I do want to talk about the fog of war in its most basic context, however, in an effort to make a couple of points about the mechanic.

Shoot, perhaps one of the best examples of battlefield uncertainty is with the classic board game Battleship, in which players can only see their half of the board.



Why is situational uncertainty considered to be important, specifically in tabletop and virtual war and strategy games? Despite being quite sure that some of you out there might have a better and more complete understanding of the underlying logic, please humor me as I attempt to do the topic justice.

Firstly, wargames and strategy games of all types are designed to be both knowledge and skill based. That is, players of these games are asked to examine complex situations, and navigate them in real time based on their knowledge of the parameters laid down by the game’s systems. Typically, over the course of a match, the player is asked to make a number of meaningful choices that inform their approach to besting their opponent: when to expand their economic operations, when to produce armies of units and where to place them on the game map (force projection), when to upgrade to gain access to new, more powerful unit types etc.

These decisions are, in well-designed games, non-trivial and often come at an extreme opportunity cost: choosing to build a new refinery or command center represents an amount of funds equal to, let’s say, 5-10 units in a game like StarCraft (400 minerals / 50 minerals per marine = 8 marines). Thus, at the time when a player constructs their command center, they are in a position for the opposing player to produce that additional value in units and overcome their defenses, winning them the game, or to take some other action which would put them in an advantageous position.

However, within a title that includes situational uncertainty caused by fog of war, the potential aggressor has to actively examine their opponent’s map area with units to determine their actions. In many cases, this puts an onus on each player to gain through subterfuge or force a knowledge of their opponent’s game state while simultaneously denying that knowledge from their opponent – knowledge then, becomes a resource that can be denied to their opponent or wrested from them.


As with poker, executing actions based on partial knowledge and on an understanding of the psychology of combat becomes a skill in its own right, and playing mind-games on one’s opponent becomes another tool in the player’s toolbox. The dance of learning as much as you can from your opponent while simultaneously denying them relevant knowledge about your own situation, is pretty core to games like StarCraft in particular.

Removing the ‘fog of war’ from a StarCraft or Command and Conquer, where tech trees and production facilities can bottleneck what a player is able to make and where oftentimes seeing a strategy before it can be successfully executed can invalidate it entirely or limit it severely (e.g. Dark Templar rush, Nod bike rush, early Banshee), or from games like Company of Heroes, where flank and ambush attacks are core to gameplay, would drastically change  the gameplay in ways which it’s easy to argue would be detrimental.


th (52)While I’d love to go into detail about the mechanical design of fog of war in various RTS titles throughout the history of the genre, to do so in the context of this article would simply be too much to read. At over 2000 words, it’s a stretch asking any normal human being to parse through all of this, and adding in exhaustive and exhausting details about all of the games I want would defy reason. Perhaps I will cover each of these games in their turn in future articles.

As with most cases in the RTS genre, the Blizzard model for implementing ‘fog of war’ into RTS games is pretty much the industry standard. The earliest implementation that I recall seeing was in Westwood’s Dune 2, in which Fog of War was represented by a total blackout of map features, and was permanently removed once initially revealed. Subsequent titles such as WarCraft: Orcs and Humans innovated on this by introducing what amounts to a dual layer fog of war: an initial ‘opaque’ layer indicating that it has never been explored, and a secondary ‘translucent’ layer which falls after a unit leaves it, leaving behind an after-image of any terrain features and structures revealed by the unit’s sight bubble.

Speaking of which, in almost every case, fog of war is revealed by units in a circle around that ‘model’ so to speak and is cumulatively shared by all units a player controls. So, information revealed by an enemy unit is always actionable by a player’s other units. In some rare cases, as in Company of Heroes 2, developers will attempt to modify this based on reality and have things such as structures, bushes, trees and other large objects obscure line of sight, but in many cases games are content to allow units to ‘see through’ most object’s in the game’s map, leaving special objects such as hills and (more recently) brush or foliage as exceptions to this rule.

Some games attempt to tweak this system slightly. StarCraft 2 gives Terran players a ‘radar’ which shows enemy units moving in the Fog of War. Supreme Commander and Planetary Annihilation provide radar and sonar of various types to determine the locations of enemies that are not visible by their units (which is certainly interesting, as in many cases this radar can be baffled or deceived in various ways. In Company of Heroes 2 and a couple of other games some abilities or conditions exist which disable radar minimaps for a greater or lesser period of time or create temporary zones that block line of sight. But most of the time, everyone seems to stick with some slight variation of the same model.


Far too often, RTS developers seem content to utilize a variant of Blizzard and Westwood’s fog of war in their games. From the very presentation of ‘fog of war’ to the way in which units interact with it, little thought seems to have been given to realizations of either the mechanic itself, or alternative mechanics which might produce similar uncertainty in more interesting and engaging ways. First, let’s look at a couple of existing titles which intentionally challenge the status quo.


There are plenty of other ways to tackle the issue of situational uncertainty in strategic games. While a commonly accepted facet of gameplay, I don’t believe that it’s entirely necessary in all cases to introduce such a system in the first place. In RTS/RTT game Z, players compete on a fairly small map without resources or fog of war. No buildings are constructed, just tanks and robot infantry, which are produced on timers from pre-built factories on the map. As a full-knowledge game, there was no real guessing at a player’s build order or map location, just the tactics of being somewhere they weren’t and utilizing your resources better than they utilized theirs.

Upcoming ‘economic RTS’ Offworld Trading Company also has no fog of war at all, though its large number of resources and interaction mechanics do seem to promise a reasonable degree of uncertainty as to the wisdom of acting against another player, and I presume that predicting and manipulating the game’s market economy is intended to be a poker-like game of baiting and bluffing. Regardless, the gross physical actions of all players are in the open for others to see and consider, though a certain number of important functions and metrics are only available to an individual – again, much like traditional multi-player card games in some ways.


Eugen’s Wargame titles have an interesting system in which units have a line of sight that is not communicated to either player, thereby seeming to show the entire map while hiding its situation. This, coupled with the immense size of maps and the prevalence of terrain objects that can hide combatants leads matches to become quite tense, with intel-gathering operations incredibly critical and very much encouraging flanking attacks on opposing support operations. As with many systems in the Wargame titles, this decision is at once refreshing and frustrating. The lack of feedback is both an interesting, challenging and quite often frustrating aspect of combat, which has profound implications on how combat proceeds in Wargame matches.

Thought Experiments

Imagine a game in which all units had realistic sight cones: where infantry could see for a mile or more, but only in a roughly 160 degree cone to their front, or a game started with both players having perfect vision of the map, with mechanics introduced later to create a shroud that obscured a player’s actions within a specific area. Or, in a historical game, where the very smoke of firing the great weapons of war would itself obscure the battlefield over time. A science fiction game with one faction having a completely different relationship to map visibility than another (e.g. a race that could sense the approximate location of enemy forces through a fog of war or similar). Or, shoot, a mechanic that allowed aircraft in the game to be seen if they’re flying above a cliff where a unit on that cliff would still be obscured.

Even minor variations on the normal model would be refreshing to see: a literal volumetric ‘fog’ of war: for instance, a Dagobah-style swamp planet where players could use lights on units or structures to see further, but with the risk of being seen far beyond a normal unit’s sight range because of this. The ability in such a case to set up decoys within the fog for enemies to waste resources attacking, et cetera.

It seems to do some injustice to the RTS genre as a whole that developers are not innovating within this space beyond incremental improvements like Relic’s True Sight system. Soren and his grew are indeed brave with Offworld Trading Company by bucking the long trend of RTS games that follow in Blizzard’s and Westwood’s (worthy) footsteps. More innovation like this is good for gamers everywhere.

Thanks for reading,



    1. As great as Wargame/Steel Divisions vision/stealth mechanics were I think they might have been a little hardcore for some, particularly when small seemingly unobtrusive terrain features would obstruct vision. Its certainly good inspiration to work from however. I too am a little bored of the classic fog of war.

      Liked by 1 person

  1. I really like this. Especially your last idea of volumetric fog that highly restricts line of sight, but can be improved by light sources. However, the opponent can see the light source from a distance but not know who or how many people are there.
    I will try my best to implement this into my current project.
    I really like your posts. Keep it up!


  2. An old Jowood game, World War 3: Black Gold had a similar “volumetric” fog of war. There was a day-night system and units (being mostly vehicles) had headlights. Turning on headlights during the night would reveal more of the fog of war but also expose you from a distance. Driving around without lights would be risky as you can just bump into enemy defenses but you were basically undetectable.

    I’ve perused some of these posts but I’m asking myself if anyone here played any of the late 90s and early 2000s European strategy games (Sudden Strike, World War 3, etc.)

    Liked by 1 person

Comments are closed.

%d bloggers like this: