Kingdoms of Darkness and Light
As I mentioned in my previous article, I was a game developer for twenty-five years, and during that time I authored a lot of design documents for games that never got created or published. Many of these I still have sitting in a folder on my laptop, most of which date back to the early 1990s.
I realized recently that since I am never actually going to be able to realize these ideas, that I might be able to use them as examples for people interested in the craft of game design.
However, there are two important caveats here. First, what I am about to present is not a complete game design (which would run to many pages), but rather a game proposal document, one that outlines the basic concepts and structure of the game. This is similar to a “pitch document” which would be used to try and sell the concept of the game to a publisher or investor. I will also include commentary on the game design process and the sources of some of the ideas. These comments will be placed (in parentheses).
Second, I don’t claim that this game, if made, would be successful in the market. As Talin’s law of game design states, “there’s no way to know how much fun a game design will be until you’ve built it.”
Kingdoms of Darkness and Light (KDL for short) is a strategy / simulation game inspired by Civilization, Populous, Master of Magic, and other games in the 4X (explore, expand, exploit, and exterminate) or “god game” genre. The distinctive aspect of KDL, however, is that it is more than just a physical simulation of empire building, but a simulation of magical and dramatic elements as well.
(A pitch document will often mention other, similar games that have succeeded in the past. Also, it would include a section on target platform and other marketing details which I will omit.)
(There are some more modern games that have a similar flavor, although not the exact same concept. These include games such as Fallen Enchantress and Age of Wonders.)
If you have ever read The Lord of the Rings, and in particular if you’ve studied the map of Middle-Earth, you may be struck by the fact that there’s something funny about the geography of the land of Mordor. Not only does it seem to violate our ideas about terrain formation, but the geometry of it seems almost intentional — as if it were a fortress crafted by an intelligent designer (which of course it was) rather than being a natural feature. The descriptions in the text only reinforce this sense of “unnaturalness” — the barren desolation of Sauron is described as a volcanic wasteland covered by an ever-present cloud of smoke and gloom.
Contrast this with the beautiful elven forest-cities of Lothlorien and Rivendell. In Tolkien’s world, as with many others in the fantasy genre, the land itself is a reflection of the moral and spiritual character of it’s inhabitants.
What is proposed is a game where this sort of “dramatic / moral character” is simulated, so that the landscape at any given location is determined by the inhabitants of that area as well as the events that take place. In turn, that landscape generates mana (magical energy), much like the “land cards” in Magic: The Gathering, and that mana can be harvested and collected by the player who then utilizes the energy to further his plans.
(The name “Kingdoms of Darkness and Light” is inspired by the novel Creatures of Light and Darkness by Roger Zelazny. I reversed the two words so that the connection wouldn’t be obvious. The name “Darkness and Light” is distinct enough that it could be used as a brand for future spin-off titles, i.e. “Cities of Darkness and Light” and so on. I also like the rhythm of the syllables: DUM-dah-dah DUM-dah-dah DUM. Two triplets and a final beat.)
KDL involves many of the familiar elements of real-time strategy games — moving and controlling units, resource management, building, etc. As in other 4X games, the player controls not only a large number of units, but can also build an empire, reshape the terrain, and construct cities and other structures. However, there are a number of unique aspects of KDL that extend and enhance this popular genre, such as Dramatic Landscapes and Representative Combat.
(The word “unique” here is a bit of salesmanship on my part. There have been a lot of games since I first wrote those words, and the ideas may no longer be as original as I believed them to be.)
The player takes on the role of a powerful wizard seeking to cast the ultimate spell, a “Master Spell” which will grant him access to all of the knowledge of the universe. In order to do this, he will need to collect vast amounts of mana, more than he alone can generate. Since mana is generated by land, he must control a mighty empire in order to generate and collect the energy he needs.
Unfortunately, rival wizards who also want to cast the master spell will be competing for those same lands and peoples. There isn’t enough energy in the world for two magicians to cast the master spell, so competition is inevitable. These other wizards are ruthless and will attempt to destroy your wizard’s infrastructure, replacing it with their own, through quiet subversion or outright war.
(Another version of this design had a different story premise, where the goal of the game was to construct an astral gateway. However, the game mechanic was the same — collect enough mana to win the game.)
At the start of play, each player controls a small village, populated by a single creature type, generally the weakest type for the faction that he controls. The player will also have control of a single “leader” character, who acts as the player’s representative. The player will gradually expand their village, send out parties to found new communities elsewhere, recruit new followers
and generally expand their sphere of influence.
Eventually the player’s representatives will encounter those of other wizards. For a time, there may be peaceful coexistence, but eventually there will be competition for land and resources and conflict will occur.
Part of a successful strategy is to transform as much of the land as possible into a form that is compatible with your minions. For example, as the evil “Dark Mage”, it would be your goal to spread the taint of darkness and death throughout the land. These evil, twisted lands will be blackened with ash, while the only plants that grow there are tangling brambles with stinging
thorns. The land will be inhabited by cruel orcs and other creatures of the darkness. Similarly, you would resist the incursions of other wizards seeking to transform your lands into a form that is incompatible with you — those pesky elves, for example, always planting forests and singing!
Eventually the wizard will have collected enough resources and knowledge to begin construction of the master spell. The first player who casts the spell wins the game.
Dramatic Landscapes — A land with “character”
The visual and behavioral aspects of an area are determined by the inhabitants of that area as well as the events that take place there. When creatures settle in an area, or conquer it by force, the appearance of the landscape is gradually transformed or “morphed” by their presence so that it is thematic accord with them. For example, if a group of sinister orcs were to settle in a pristine forest, the trees would gradually transform into hostile brambles, while artificial structures such as castles and houses would take on a dark and forbidding aspect.
The different kinds of terrain are organized around eight magical aspects. Each aspect represents a particular mood or theme, such as “dark and forbidding” or “enchanted woodland”. Each aspect is also associated with a faction, which represents the kinds of creatures and characters that are tied to that aspect.
This transformation of landscape is one of the primary goals of the game. A player who has chosen to represent a particular aspect will find their powers and abilities increasing as that aspect comes to dominate the landscape. Land which is harmony with the player’s minions will strengthen them, and through them, the player.
Representative Combat — Managing Complexity
One of the most vexing aspects of modern strategy games is that the endgame often becomes tedious due to the large number of units under the player’s control. A typical game of Civilization requires 10 hours or more, and by the end of the game the player is controlling dozens of cities, and even larger numbers of military units.
One solution to this problem is the concept of meta-control, which I first experienced in the game Populous. The idea here is that you don’t control individual units directly; instead you set a general policy which those units will obey intelligently.
In KDL, policy is set through “leader” or “foreman” characters. The player starts out with a single leader, but as their empire increases, more leaders will flock to their cause. Each leader is unique, and has distinct attributes and abilities. Leaders can recruit followers from the surrounding region, which are ordinary citizens (peasants or grunts) or creatures who are attracted to the leader’s charisma. As leaders gain levels and experience, they will attract a larger number of followers. While the number of leaders is limited and rises slowly, the number of followers that those leaders can recruit can increase almost without limit.
(Another game design that I wrote years later was a Sim Mars concept, inspired by Kim Stanley Robinson’s Red Mars trilogy, that incorporated a different kind of meta-control mechanic: the player would place radio transponders on the landscape, which would issue commands to any automated rovers within a given radius — so if the player wanted to harvest minerals at a location, they would plop down a transponder with a “harvest” signal at that point. A line of “come here” transponders would define a road.)
The player directs a leader to perform some activity, and the leader will direct their followers to engage in that activity. However, unlike many other games which feature named “heroes”, not all of the leaders are fighters, and not all of the leader activities are combat-oriented.
For example, one type of leader could be a master blacksmith or mason, who is focused on construction. You can direct the leader to build a smithy or a cathedral, and they will direct their followers to gather materials and assemble them in the designated location.
The landscape will possess an ambient scattering of followers even in areas where there is no leader, however the activities of those followers will be disorganized and unfocused. These followers will still participate in the mana economy and engage in activities beneficial to their own survival. It is only when a leader is in proximity to they hear the siren call of contributing to the larger cause.
Leader abilities are skill-based rather than class-based, with their own skill tree. So a leader who starts out with only combat abilities can eventually become a master blacksmith or magical researcher.
Some possible categories of leader skills:
- Construction — the ability to build new buildings.
- Evangelism — the ability to attract additional followers.
- Forestry — leaders with this skill gain more wood per tree harvested.
- Mining — leaders with this skill gain more ore per unit of time.
- Research — Allows the the discovery of new player spells.
- Scouting — Increases the detection radius of the leader.
- Spell casting — Allows the leader character to cast spells in battle.
- Spying — Allows the leader and followers to slip past an enemy undetected.
- Warfare — Increases the leader’s ability to lead troops in battle.
- Wilderness Travel — increases the speed of any accompanying followers.
As the player’s strength and reputation increase, additional leaders may be acquired. Wandering mercenaries may volunteer to join your forces. Leaders may be “created” by promoting a follower character to a leader using magic. Leaders may be encountered as you explore the territory (for example, an ancient ruin may have a leader character in it who needs to be rescued.) Finally, it is possible to “suborn” your enemies’ leaders through the use of magic and skill.
An interesting fact about leaders is that any leader can be recruited by any faction. A high elf leader could become an agent of the dark lord if they so choose, or a “good” player could have a reformed demon as a leader. This creates an interesting dramatic tension and depth of character, and allows the player to make up an interesting backstory explaining why this particular leader chose to go down this path. The tension increases as the player starts to equip these characters with weapons and artifacts aligned with their faction.
Another common problem in 4X games is that construction in the endgame — when the player has a large number of cities to manage — is fairly mechanical and not very interesting. Typically, every one of the player’s cities has a construction queue, and the game encourages the player to keep building more and more esoteric and advanced structures rather than letting the queue stay empty. (It’s possible in most of these games to have them simply produce gold or population instead of buildings, but this is not very interesting). This means that the experience of the endgame largely involves a lot of mindless clicking whenever the game wants your attention.
A simple solution to this problem is to limit the total number of advanced buildings of each type. That is, every village and hamlet can have a chapel, but you don’t need more than two or three cathedrals in your whole empire. Rather than setting an arbitrary cap, however, in KDL these advanced buildings require some hard-to-find resource that has a limited supply. For example, a cathedral requires an archbishop, and there are a limited number of archbishops. Whether archbishops are actually leaders, or are a special kind of follower, or neither, is still to be determined.
This means that during the end-game, most established smaller towns will have empty construction queues. This is OK.
(Another mechanic that is worth exploring is the idea of city specializations — that is, certain cities specialize in a particular type of activity, and choose to build buildings consistent with that activity. I’ve seen a number of recent 4X games do this, but only in a fairly rudimentary way with a small number of choices and combinations.)
An important “fun factor” in 4X games is option escalation. What this means is that when the game first starts out, the types and variety of actions that the player can perform is limited, and by consequence, easy to learn. However, as the game progresses, more and more kinds of actions are “unlocked” and the game becomes more complex.
This is accomplished in several ways:
- increasing the number of spells in the player’s spell book
- allowing leader characters to learn from experience.
- acquiring new magical artifacts.
The acquisition of new spells is primarily a function of the research activity. Like combat and construction, this activity is leader-driven. So a leader who has a talent for research can build a library and recruit followers to staff it.
Spells can also be discovered through exploration of the world (scrolls contained in ancient archives), or via conquest of independent (non-faction) bands of enemies.
The Faction Wheel
The eight major factions are arranged in a circle, or wheel, so that some factions are adjacent while others are opposed. Factions that are adjacent are potential allies, while factions on the opposite sides are bitter enemies.
(Exactly what is meant by “alliance” was never determined. I assume it means that you could use some spells / mana from an allies resource pool, or recruit a small number of creatures of that type.)
I’ve numbered the factions 1–8. Faction 1 is adjacent to both faction 2 and faction 8 (it wraps around).
Faction 1: Lords of the Dawn
Creatures: Serfs, Halflings, Knights, Crusaders, Priests, and other entities of Chivalrous Civilization.
Terrain: Green hills, Ploughed land, picket fences, orchards, cottages, whitewashed keeps, Arthurian castles.
Faction 2: Forest Guardians
Creatures: Elves, Rangers, Fairies, Sprites, Druids, Unicorns, Elf Lords, and
other enchanted woodland creatures.
Terrain: Enchanted woodland, sylvan glades, tree houses, fairy castles.
Faction 3: Watchers in the Gloom
Creatures: Trolls, Gorgons, Hags, and other creatures of the swamp.
Terrain: Bogs, swamps, marshes, jungles, rude mud huts and vine-covered ruins.
Faction 4: Frostclan
Creatures: Yetis, Ice elves, white wolves, snow dragons.
Terrain: Ice, snow, glacial, frozen lakes.
Faction 5: League of Darkness
Creatures: Orcs, Assassins, Dark elves, Demons, and other creatures of
darkness and evil.
Terrain: Blackened Ash, Brambled forests, dark forbidding towers.
Faction 6: Night Creepers
Creatures: Skeletons, Zombies, Phantoms, Vampires, Liches, and other undead creatures.
Terrain: Lichen-covered rock, petrified forests, graveyards, mausoleums, bone cities.
Faction 7: Blood Raiders
Creatures: Goblins, Barbarians, Nomads, Fire Dragons, and other warlike
denizens of the desert.
Terrain: Desert, Sand dunes, Volcanic mountains, barbarian camps, iron towers.
Faction 8: Stone Masters
Creatures: Dwarves, Gnomes, Giants, Ogres, Mountain Men, Shadow Dragons, and other creatures of the earth.
Terrain: Sandstone flats, Pine trees, rocky plains, mine entrances, keeps and cities constructed from rough stone slabs.
(I’m not completely happy with this design, and it could use a lot of refinement and fleshing out, or even a complete rethink. Also, I originally wanted to call faction 1 “Lords of Light”, but that is the title of a different Roger Zelazny novel. :))
(I’ve also thought about avoiding the standard Elves/Dwarves/Orcs tropes and creating unique and distinct creature races. However, the familiar archetypal races provide a convenient “shorthand”, avoiding the need to explain to the player about the history and culture of this world. The optimal solution might be to create a blend of original and exotic species to be mixed in with the traditional ones.)
The Mana Economy
Any creature that dwells in an area for a length of time leaves a certain thematic “effluvium” which affects the aspect of the land. A larger population of intelligent creatures leaves a more intense residue than do smaller populations of lesser-developed lifeforms.
As mentioned previously, once enough of this residue has collected in a given place, it begins to re-characterize the land to it’s own aspect.
Other actions which affect the aspect of the land might include planting certain types of trees, plowing fields, building cities, magical spells
which, blessings and cursings, and mighty battles.
Most of the terrain would starts out in a ‘neutral’ state, not aligned with any given faction or aspect. This does not mean that the neutral terrain lacks variety — there could be hills and forests, but they would be generic, wild places.
This neutral terrain would still contain hazards, in the form of independent (non-faction) bands of hostile enemies and monsters. This allows the character’s leaders to gain battle experience earlier on in the game before having to face the other factions.
As your creatures settle into new areas, they gradually transform the land, making it more comfortable and compatible with their way of life. The appearance of the land changes, as does its effect upon the creatures that live on it. Creatures who are on or near their own lands will gain strength from that land, and gain morale by it. In addition, once the land has been transformed fully (which is only possible when the population gets relatively dense), it starts to generate magical energy which can be collected by special structures that you can build.
Creatures also generate mana. Large concentrated populations of creatures generate great quantities of mana, and can also “warp” the appearance of the land more intensely. A highly developed city that is entirely populated with creatures of a certain aspect will generate that type of mana in great quantities, and will affect the appearance of that city. For example, a large city
populated entirely by minions of the Dark Wizard will have forbidding ebon walls, sinister-looking towers, and just a general look of evil. A smaller city, however, with a lesser population might just look brutal and harsh with crude stone fortresses and black iron gates.
In any city, there will be a random mix of buildings and styles, because the mana field will never be completely uniform at any point. In addition, other factors, such as wealth and poverty, food supply, population level, and special building types can also affects the appearance of inhabited structures, so there will be a great variety, just as there is in Sim City.
Magical spells are powered by mana. Magical artifacts and spells often require a continuous intake of mana to maintain their operation. Different types of spells require different types of mana. The type of creatures under a wizard’s control will therefore determine the types of spells he can cast.
The ability to cast spells is highly location-dependent. The player has only a relatively small “personal” mana pool, although this does grow over time. Being able to cast more powerful and expensive spells requires a combination of the player’s personal mana and the local mana available on the terrain. This means that the player’s magical power is greatest in or near regions whose terrain is compatible with the player’s chosen aspect.
However, some spells are “long-range” in that they can be cast in one location and have an effect some distance away. A player could cast an ice storm or a “cloud of despair” originating in one city, and target it to a nearby enemy city. But the cost of bombarding a city half a world away from your empire would be prohibitive.
One of the primary challenges in creating a game like this is maintaining game balance. There are several issues that must be addressed:
Symmetry of factions while maintaining distinctiveness: since the player can choose any of the factions, they must all be of roughly equal power — none can be stronger or weaker than the others. At the same time, however, they can’t all be clones of each other. Each faction must have it’s own distinct spells, buildings and creatures with different powers and abilities.
Negative feedback loops: Most of the game mechanics described so far are positive-feedback: the more land you acquire, the more powerful you get, making it easier to gain even more land and more power. This means that if a player gets an early lead, winning becomes a foregone conclusion and the game becomes boring.
There are many potential remedies. For example, it is often the case that more advanced buildings and characters require a greater upkeep cost. Moreover, the loss of a major asset can be devastating, so you have to spend additional resources protecting it. Similar logic applies to defending the empire’s borders — the larger your empire, the more border you have to protect.
Another option is to say that certain powers and abilities depend only on locally available resources, rather than the resources of the entire empire. An example would be the defenses for a city, which cannot rely on help from neighboring cities in time to hold off an invasion.
Note that too many negative feedback terms will produce stasis. For example, if creatures grow weak when away from their homeland, then the defensive advantage is so great that conquest becomes impossible and the game becomes a stalemate.
Limits on mana accumulation: Since, as described above, mana begets mana, we need a way to make sure that we don’t saturate the landscape such that eventually every tile under your control has the maximum amount of mana. We instead want mana and aspects to “clump” so that some areas are more concentrated than others.
Elbow room: in many 4X games the starting point of your empire is located well away from the starting points of your rivals. This gives you room to grow your empire a bit and focus on construction and building before having to shift into a wartime economy in the mid-game.
Graphics and Interaction
I’m not going to talk about this too much, since the state of the art for 4X games is pretty much established. It would be a 3D rendered terrain composed of squares or hexes (like Civilization 6), with the camera able to rotate to any angle as well as zoom in / out. A “fog of war” effect will be used to limit the player’s ability to see areas outside the sensory range of his units.
The primary challenge would be implementing the morphable terrain, which would be fairly artwork-intensive.
I’m not a big fan of real-time games, mainly because I like to think about my moves and contemplate my strategy without being rushed. (Also, I don’t have the reflexes that I had as a teen-ager). However, what a lot of games do is that they are “slow” real-time at the largest scales (where rapid response is not so important), and then switch to a turn-based mode for battles.
Players interact with the units by selecting leader characters on the playfield. There will also be a method for moving the camera to a leader who is off-screen, as well as selecting that leader. The player can either point and click to indicate an action they wish to take, or pop up a context sensitive menu which lists the special actions that a leader character is capable of.
The number of on-screen controls will be kept to a minimum, allowing the 3D display to take up most if not all of the screen area. An overview map will be superimposed on one corner of the display. The number of “game modes” will also be kept to a minimum — for example, whereas in Civilization there is a special “city mode” which is used for the construction of buildings, in this
game all of the construction is done within the main game mode. There will of course be “summary / overview” dashboards to check on the status of the player’s leaders and/or cities.
KDL could certainly be done as a multi-player game. In a turn-based game, you need to put an upper-limit on turn length so that one player doesn’t freeze the game for everyone else. For a real-time game, you could allow players to have a limited supply of “timeouts” that could be used to temporarily pause the game so that they have time to think. The supply of timeouts would be renewed gradually.
There are many design questions that remain unanswered. (There are also a number of details in the original design doc that I have left out for brevity, such as mechanics on city decay.)
Here’s a sample:
The biggest unknown, I think, is that the algorithm for generating mana and its influence on creatures is critical; getting this right will make or break the game. The design as it stands is extremely vague in this regard.
Similarly, the concept of morphing terrain is unproven, both visually, and in terms of whether it makes for a compelling game experience.
Another question is whether it is possible to eliminate a faction from the game, or at least the wizard controlling that faction.
What are the mechanics of settling new cities? For most 4X games there is a specialized “settler” unit which requires resources to generate and is used up in the process, but it would be preferable not to have to introduce a new unit type.
Do the same followers stay with a leader when they move around the map? For example, when a leader engages in a sortie into enemy territory, do the same set of followers go along, and suffer attrition from one battle to the next? Or is the set of followers dynamic, changing with each assigned task?
How do followers operate in battle? I’m assuming battle is a special mode that zooms in on the battlefield and allows the player to make tactical decisions. Does the player control individual followers in this mode? Do followers gain experience? I kind of what to keep followers simple and disposable, to avoid overwhelming the player with complexity, but I don’t know how much fun combat will be without fine-grained commands and tactical options.
A complete game design would include a detailed description of all of the game elements:
- all of the types of creatures, what their powers and abilities are, what they look like, and so on.
- specifications for all of the kinds of buildings available to each faction.
- all of the types of terrain, and what the effects they have.
- all of the various leader abilities and tech trees, as well as any equipment that they might use.
- all of the spells that can be cast by the player.
- a list of all environmental hazards, wandering monsters, artifacts, treasures and rewards.
- anything else I forgot :)
While this is being worked out, there would be a parallel effort to do a quick and dirty prototype of the game. In the prototype, there would be little or no artwork — terrain aspects would be represented as solid colors, and creatures and monsters would be sprites (possibly borrowed from other games) or rectangles with text. The most important result of this phase would be to come up with a viable AI for computer-driven opponents.
After that would be a group of graphics programmers and artists working on coming up with a viable technique for procedurally generating the morphable terrain.
And eventually someone is going to need the patience to play the game many, many times.
Common Misconceptions about Computer Game Design
Although I no longer design computer games for a living, I was a game developer for 25 years, starting at around 1983…
More Game Development Misconceptions
My recent essay, Common Misconceptions about Computer Game Design, turned out to be much more popular than I would have…