Last year I wrote an article encouraging 4X companies to expand their designs in certain areas.  It’s too soon to tell if that article had any impact, but it’s not too soon to ask for more!

Starting Conditions

I’d like to start in an obvious place: Starting Conditions. In classic 4X design, the player (note that any time I say “player” in this piece, I am also including the AI) starts out with one city or one planet and then expands from there. Why though? I understand it’s a compelling story to work your way up from almost nothing into a massive empire, but once you’ve lived that story a few times, wouldn’t it be nice to try something else?

AoW3 lets you start with no cities, but what about 3, instead?

Companies could easily allow players to start with 3, 5, 7, or however many cities they want (depending on map size). This would pose all kinds of neat new problems for players to solve and get past the early game routine that some people don’t like as much. I think starting as an already established kingdom/federation is a fun prospect, especially for players with busy lives who are shorter on time. My experience tells me that the 4X audience does tend to be a little bit older than other game genres like FPS or RTS, so maybe this would be a way to better appeal to that demographic.

Some might argue that any 4X game that starts with more than one city is actually a Grand Strategy game. I don’t agree with that. “4X has always had games that start with one city or planet because it always has,” I’m told. That’s circular logic, and I believe that 4X and Grand Strategy have something to teach one another. Besides that, the most classic of all 4X games, Master of Orion 2, allowed players to start with multiple colonies using the “Advanced“ start option. As our genre develops, it’s natural that it takes on the attributes of others. 4X has already adopted many RPG elements, for instance, such as experience points, leveling, mana systems, and so on. Taking something from grand strategy- especially as an option- shouldn’t be all that controversial.

Speaking of starting cities/planets, I think there could be some innovation here, too. How about a dial that lets you control how many starting buildings, population units (pops), military units/vessels, resources, and so on? For some games, this wouldn’t always work. Endless Legend for instance makes you research most of your buildings, but games like Age of Wonders III and Planar Conquest have all sorts of beginning buildings a city could have at the start. Conversely, a player could dial it down to nothing in order to increase the difficulty, as many are wont to do.

Finally, I’d like to see games implement a way to easily re-randomize the starting terrain/system around a player’s starting city/planet. Some games have this already like the very obscure (and poorly received) Apollo4X. One of the things I actually loved about that game was how easy it was to re-randomize all the planets if you didn’t like the configuration the game came up with. It was a great innovation. Other games should implement this. How many times have you restarted a 4X game when you realized your starting position sucked? I know it’s scummy, but we’ve all done it! A game like Thea: The Awakening could REALLY benefit from that little design nugget. And it’s not like the game would need to regenerate the whole map, just the city’s surrounding tiles. That would be fairly easy to implement, I would think.

The Endgame

I touched on victory conditions last time and I’m going to do that again, as well as discuss the endgame as a whole. First, let’s get victory conditions out of the way. I like the idea of having multiple victory conditions from a design standpoint because it keeps the end in doubt. We all know that snowballing and steamrolling are a problem, and multiple ways to win can help mitigate those issues. If you’re not sure why more than one wincon is good, READ THIS as an example. I think most people feel that players should at least have more than one path to a win.

02 Endgame Art
Endless Legend made a good attempt at multiple victory conditions, but it’s not perfect.

Instead of listing different types of victories design companies could use, I’m going to give some general guidance on how many and what types of victories games might need. 4X games have a lot of subsystems: diplomacy, research, questing, combat, etc. Different people find different subsystems more interesting. I would submit that each major subsystem should have its own victory condition. If you have a diplomatic system, you should have a diplomatic victory. If you have a quest system, you should have a quest victory. Of course, each victory needs an on/off switch in the options so that players can customize their experience or focus on the types of victory they enjoy most. That’s just as important as including multiple wincons to begin with.

There are also ways to make the  endgame more engaging. As mentioned, snowballing – when every win (like taking control of a city, planet, etc.) makes the next win easier to achieve – is a real problem. So why not add a “Desperation” mechanic that boosts productivity from pops for a few turns after each city/planet is lost? Players can get new buildings and military units finished faster and maybe add to their gold or mana stockpiles in order to be in a stronger position to take back their possessions.

The boost from Desperation would be directly proportional to the number of cities lost from the player’s max, so the more cities he/she has lost, the greater the bonus. That would mean that by the time you’re down to your last one, your production and bonuses would be off the charts. It would certainly make taking that last settlement far more interesting in my mind.

Another way to add tension to the endgame is through triggered events. The Antarans from Master of Orion 2 are a good example, as are Distant Worlds’ and Polaris Sector’s endgame threats. Stellaris has something similar, although the current execution might not be the greatest. Anyway, triggered events that add a new danger, resource, difficulty, building, super-technology, or win condition to the game as it appears to be drawing to a close would change the whole dynamic of play. Players would suddenly have to adapt to a whole new environment and maybe even have to pursue a different strategy for victory. As with all these types of suggestions, there would need to be an off switch for it in game setup. There are players out there who just want the game to proceed along predictably, and in some ways, I don’t blame them.

Thematic Content

If you haven’t read my Sorcerer King vs. Thea: the Awakening article yet, I would encourage you to do so. This section assumes you have. I just want to reinforce the idea that many modern players want more out of their 4X games than just strategic and tactical considerations. Plenty of us want to get to know the world and the lore behind the people and places they see.

03 Thematic Art
As the playerbase ages, they want more sophistication in their games.

I think EL started this in the more recent crop of 4X games with its compelling factions and faction quests. That was a great place to begin, but one problem I have with the game is that I don’t know where I (as the player) fit in. Am I the collective consciousness of the people? Am I God? There is no Ghandi or Rjak-type character for me to be.

I think Thea, SK, and Stellaris have also shown that one of the primary selling points for a 4X game can be its storyline. Still, there’s room for improvement. In Thea, I choose a deity, but I don’t get the sense I’m portraying that deity during play. SK doesn’t give me enough of a world to interact with even though I totally love the basic situation in that game. Stellaris promises emergent thematic play, but the still-in-development diplomacy system really hamstrings unexpected story generation.

04 Thundergod Art
Beating Thea means learning the stories of individuals. Nice touch.

Games with narrative/thematic elements give players more context to the conflicts as they play out. That context makes the players feel like their decisions are more consequential and important. Nothing, in my opinion, sucks more in a 4X game than the feeling that your decisions don’t have an impact on the results. Developing the theme and lore of a game goes a long way in making sure those feelings don’t creep up on the player.


Of all the things I’ll suggest in this article, licensing is potentially the most dangerous. 4X is one of the few video game genres with little to no licensing at all. It’s kind of amazing when you think about it. The RTS genre really started off with Dune. FPS and RPGs license almost anything. Electronic card games license real-world card games like Pokémon or other intellectual properties (IP) within the company like Hearthstone. And there are hardly any sports games out there at all that don’t license something (notable exception: Rocket League).

So what’s up with 4X? Why don’t we have any licenses? Well, for one thing, 4X games don’t make a lot of money. If “Civilization” isn’t in your title, you’re going to sell fewer than a million units. That’s not enough to entice a lot of IP holders and probably not enough to even pay for the really big licenses that might be able to carry a title on their own. And a license gamble that doesn’t pay off usually ends in bankruptcy (see THQ). However, that won’t stop me from asking!

05 B5 Art
Easily the best space series from the 90’s. Why so few video games about it?

I think some really great 4X games could be made out of the following IPs:


  • Star Wars
  • Game Of Thrones
  • Deadlands
  • Star Trek TNG/DS9 Era
  • Battletech (Clan War era ~ 3060)
  • Dungeons and Dragons: Birthright
  • Babylon 5
  • Mad Max
  • Magic: The Gathering

Try telling me that some of those titles don’t get your creative juices flowing. Battletech especially piques my interest. Blowing up mechs is always cool and doing that in a turn-based strategy game with EL empire management and AoW3-styled battleboards sounds sweet!

[Side note: there is a 4X game based on Star Wars available on GOG called Star Wars: Rebellion]

Small Items

I have a whole list of small things I’d like to see invented or incorporated into modern 4X games. I’ll just briefly go through them:


A lot of games have espionage (EL, Master of Orion: Conquer the Stars), but not too many games have what I call subterfuge. Subterfuge, for the purposes of this article, is being able to insert units into key strategic places unseen by other players. For instance, being able to hide a ship from the AI’s view in a nebula or hide a stack of ranger units in a forest. Subterfuge also includes setting traps. Players could place something invisible to others that damages or kills units as they move through that tile. Alternatively, why not have cloaking technology that could hide a starbase or camouflaged totems that foment unrest in a nearby enemy city? Increasing the number of ways players can interact with the game world is an excellent way to bring that world alive and make the player feel like he or she has many varied and viable actions to take in order to achieve his or her goals.


Shadows Featured
The Forgotten’s invisible units are a good start, but the idea can be expanded.

Enhanced Questing

Thea has one of the best questing systems I’ve ever seen in a 4X style game. It takes what EL and SK did and moves it to the next level. Other games should follow suit. However, I’d like to see questing taken to an even higher place. Thea’s quests involve many compelling storylines and factions, but once the quest is done, there’s no permanent effect left on the world. An enhanced questing system would have the depth of Thea, but also change the course of future events based on the player’s choices. Work together with the orc tribe to defeat a common foe? From then on, the orcs will be able to expand their empire but also have friendly dealings with you in the future. But in another game, let’s say you screw over those orcs. Then other orc tribes relentlessly assault you because of a blood oath they took to avenge their brothers and sisters. That certainly sounds like a fun game to play, right?

07 Quest Art
I like the tough moral choices in Thea, but there are no long-term consequences.

Customizable GUI Elements

One of the big bugaboos of video game design is tooltips. Any game that doesn’t launch with them is rightfully lambasted for it, and even games that do have lots of them are still met with complaints that there aren’t enough. However, after a while, these tooltips just get in the way. Once you learn where the diplomacy button is, do you really need the tooltip anymore? Tooltips and help screens should have a toggle between Beginner, Veteran, Expert, and Off. For beginners, tooltips would pop up EVERYWHERE and very quickly – even tooltips for basic things like “Close” when the cursor mouse-overs the X on a panel. For a Veteran, the delay on tooltips would be increased and some of the more obvious tooltips would be eliminated. For Experts, tooltips would have an even longer delay and they’d be gone altogether on the main screen. And then obviously Off would eliminate them altogether, giving players a clean interface to use once they are totally comfortable with the game. The same ideas could be applied to help screens and info boxes as well.

End 1UPT

I cannot stand one-unit-per-tile! I know there are people out there that love it, but it messes with the AI, clutters the map, and increases the amount of scrolling I have to do. Civilization V brought the mechanic into the mainstream and it is certainly the most high profile game to use it. Thankfully, though, Civ VI is said to be modifying that somewhat. I think AoW 3’s limited stack sizes is an ideal middle ground. You don’t have to manage huge numbers of units, but you can also use the map more strategically. I hope other 4X game designers follow suit.


Distant Worlds: Universe is the undisputed king of automation in 4X games. Stellaris took a stab at it with Sectors, but for the most part, 4X games don’t offer too many systems for helping players more easily manage the ever-growing complex list of systems and subsystems in games. 4X games need better auto-scouting mechanics, better auto-settling mechanics, ways to carve up an empire into smaller, more manageable chunks, better unit design and deployment systems that automatically address needs in an empire, and finally, smart build queues that automatically update when a tech is unlocked or resource discovered.

08 Automation Art
If it weren’t for automation, there’s no way people could play this.

Dynamic Loot

4X games need to get smarter. Almost all of them have some type of “goodie hut” mechanic that gives the player random loot. Why random, though? I guess if it’s random enough, sooner or later the player will get something useful, but it doesn’t need to be that way. How much extra coding would it take to have the computer look at a player’s current resource levels and then modify the loot table based on that?

Take EL for example. Late in the game, there’s a tech that lets you reset all the goodie huts so you can explore them again. When the player goes back to do that, the game shouldn’t just hand out random crap. It should look at the FIDSI and as well as the strategic and luxury resources and provide rewards based on what the player needs most. Need a few more emeralds to activate that resource? The next goodie hut has a higher chance of giving them to you. Low on dust? The next goodie hut will have a higher likelihood of awarding that to you. It’s always been smart to loot goodie huts, but now that smart play takes on even more importance as the player understands that there’s a great chance he or she will get something crucial. Imagine how this might affect multiplayer. The rewards would also need to scale depending upon when they are found. In the early game, finding a few dozen units of something is great, but by the mid to late game, you’re going to need many times more to make a difference. A good dynamic loot algorithm would take that into account. The race to explore would be of even greater importance and provide the combatants another arena in which they can compete.

Non-Fascist Governments

Most games pay lip service to government types if they address it at all. Even Civ games, whenever you can choose government types, don’t really restrict the player from just doing whatever. Stellaris took a nice stab at making a faction’s dispositions and governmental organization matter, but it’s time to take it to an even deeper level. Stellaris’s bonuses and modifiers are just too small to be felt, and the real in-game consequences are only mildly play-altering. The players need feel the difference.

09 Fascist Art
Hello, I am Ghandi. Would you like to be nuked today?

Most games offer a dictatorial or fascist set up, thanks to our genre’s forefather games like MoO and Master of Magic. One of the few recent games that has a totally different societal structure is Thea. That game introduces the player to a collectivist society where there is no real strongman in charge of the government. That’s a refreshing difference. More games should explore different ideas when it comes to government and society in ways that are more than just window dressing. Part of the fun of strategy games is learning to adapt to new and difficult situations. Mastering the ins and outs of a republic or dealing with strong personalities in an oligarchic government can be part of the fun in a 4X – more than just choosing whatever government type gives you the biggest bonus to production or starship lasers.


Just as before, the purpose of this article is mainly to spark innovation and conversation. The ideas suggested here are merely a sampling of brilliant design gems I’ve come across over the years. I’m sure our gifted community has plenty to add. I believe that 4X is ready to evolve into the next level. There is hope that in the near future, more 4X games, other than the Civilization franchise, will reach mass audiences. By pushing the innovation envelope, we just might get there.


  1. Awesome article Troy. Going to have to read it a couple times again there is so many great ideas in it.

    I know that Thea really has added some great new concepts to the genre as had Endless Legends and many more. While personally I prefer to dominate my way to victory more victory scenarios can’t hurt at all.


  2. Great read! I definitely agree with the majority of your points, aside from stacking units. I hate to see all the units that I’ve built hide in one tile. I enjoy seeing the art teams’ work on the map.

    I’m a little surprised that you didn’t mention super weapons/buildings. That’s an area that I feel the 4x genre can really benefit from.. I’m talking about a multi-hex unique lazar weapon that decimates any foes around it, or a dense forest (that spans multiply hexes as well) that provides a penalty to units around it. I mean, how cool would it be to stumble upon a super dense tropical rain forest that is pain to attack through? Pop a city near it and you’ve got a all-natural defensive layer.

    Civ BE somewhat touched on this area with the biome unique multi-hex buildings that had quests tied to them, but I think they only scratched the surface .


  3. Every major subsystem needs a victory condition? But you have problem then: One player plays for economy, an other for science and the third for diplomacy. Every player plays his own game and never touch the other players then. Boring! It needs conflicts between the players in a game!

    Victory conditions must have close dependencies: You want to win with science? Then you must stop the other players! You want to win with the best economy? Then enslave the others! You want to win with diplomacy? Occupy their lands! IMO is a counter of points not a good game design.


    • What about economic victory being linked to other factions, same being with diplomacy, and the science victory is the hardest to get without others helping out.

      Diplomacy through a cannon is just another version of warfare really. How about diplomacy through resistance. Setting up a coalition to defend against and defeat an aggressor. Same with the economy, but only through some particular achievement.


      • Victory conditions linked to other players are not functional: Why should I help you to win the game?

        And an other problem: It is not funny when the game is suddenly over, because I can’t see your progress (in diplomacy, economy,…). Wargames are popular, one reason is the feeling for your own and your enemys progress.


      • I thin the most important thing is that victory systems should be based around some sort of interaction between empires. Tech victories are the worst IMHO because there is nothing inherently interactive about the way research works. If an opponent is getting close to winning a tech victory, your only real recourse is to attack them and destroy them – at which point you are basically railroaded into a conquest victory anyway.


  4. Great article, Troy!

    Just wanted to point out that there is, in fact, a 4x game set in the Star Trek universe (specifically the TNG/DS9 era): Birth of the Federation, released back in 1999. Despite its age and some obvious flaws, it remains one of my favorite space 4x titles to this day; I continue to entertain the (admittedly vain) hope of a sequel.

    Liked by 1 person

      • You’re welcome! Always happy to bring a 4x title to someone’s attention, especially one that I have a particular fondness for. :)

        Not that BOTF was great, objectively speaking — it was too heavy on micromanagement, the AI was poor, and there were a number of major bugs that never got fixed (including a bad memory leak that causes slow turn processing, even on modern machines). However, it was pretty damn immersive with good atmosphere, really throwing you into the Trek universe. Anyone who’s a fan of both Star Trek and 4x titles who *didn’t* play this game really missed out IMHO.


    • Ah yes, BotF had some very interesting combat mechanics for its time, I really liked it at the time and believe that it is an approach worth considering even for today’s game designs.

      For those who are not familiar with it, in BotF, combat was semi real time: at the beginning of each turn you issued your orders and then those orders were executed for some 30 seconds or so in real time, then another pause to issue orders.

      Things were essentially simultaneous, you had to think a little in advance to try and predict what the enemy will do, but apart from that it still pretty much played like a turn based system. As said, I loved it.

      Liked by 1 person

  5. Interesting thought starter topic, thanks:

    Maybe it is just me, but I find the idea of fixed victory conditions to be antiquated and boring.

    The games I loved the most were always the ones, which – upon seemingly winning – just gave me a bigger goal to pursue and kept on going.

    As such I prefer to look at victory conditions more as milestones in the end game and would love to see more games where the game only really ends if you lose or if you had enough, but apart from that, you can keep on going forward.

    Admittedly, designing such an open ended game is an extreme challenge and making those end game conditions and events meaningful is a lot more work than simply putting up a “you have won, congrats” screen for having researched the “answer to life, the universe and everything”.

    How cool would it be, if instead you would then have to research the question too, and once you have the question, you would need to figure out what to do with it and then some…

    Of course that would require a very sophisticated RNG capable of generating meaningful context sensitive research topics and game events to avoid repetition. While I believe this to be possible in today’s home computing, I agree that it is more work than what a typical game could reasonably expect. Maybe one day as someone’s love child of a project, we might see something along these lines.

    However the above mentioned open ended victory conditions can be made a lot simpler, we could consider them to be achievements with various metrics you can fight to improve. Examples:

    – Can you unite the galaxy in one big alliance? For how long? Will that alliance endure if external forces challenge it?
    – Can you spread you culture to most of the galaxy? Really? How long will it be before it fragments into local variations and what started out to be your culture is turned into something completely different?
    – Can you claim technological superiority over the known universe? For how long? What will you do with it?

    Liked by 1 person

  6. I agree with most of this. Single units per tile don’t bother me as much as doomstacks do, and I do agree that limited units per tile is a good comprimise. I actually live the way EL handled it where you can increase the number of units per tile through the research tree. Although I do think it would be better if certain units took up more ‘space.’ Like maybe you can only have six units per hex, but tanks count as two units, or something like that.

    I disagree with you however on Goodie Huts. To me, immersion is one of the most important aspects of any game. Why would an ancient civilization plan their dungeons/temples/whatever-the-ruins-used-to-be around the hardships of a future occupier? Goodie Huts should remain random, but I think a cool idea would be to have a research node based on Archaeology which would give you a hint towards (or even flat out tell you exactly) what every Hut contained. That would make them remain more relevant mid-to-late game and could even add heightened drama to the game. “I really need Emeralds. This Abandoned Temple over here is lousy with them, but it’s deep in another player’s territory. If I send a small force, they might get picked off, but if I send a bigger force, the enemy might think I’m invading and close borders.”

    Liked by 1 person

    • Greater stacks over research tree like in EL and ‘Fallen Enchantress’ are a bad design descision:
      1. Its overpowered,
      2. everybody must research it,
      3. if you have greater military power you can conquer more citys and will have more research … and greater military power too.


      • That is pretty much a matter of how research was designed to work. If it is CIV like, where research is a function of your total economic output, then sure, more research leads to better troops, which lead to more territory, which lead to even more research, repeat.

        If you decouple research from economy however, then this stops being the case. So I would advise against picking any single design element and claiming it to be bad, just because it does not work in a given context. You are of course right, that in CIV like games, high research often leads to snowballing.


    • Hi Blas,

      I played EL, but I can’t see any advantages of the researching of stacksizes: Everybody must do it asap in EL. But EL has a great combat system: Terrain is the key for a small army and you must think first ;) In one of my battles a 4-stack army beat a 2+6-stack army in a bottleneck. Nice!

      But you are right: Civ has a bad research system and snowballing is one problem.


      • You said it yourself, a 4 stack beat a 2+6 stack. So in that case, and many more if you use the terrain to your advantage, that research node did your opponent no good. How is that overpowered or a must-have? Actually I find that as long as resources are plentiful, researching gear trumps bigger stacks by a mile. My fastest victory ever (Economic) ignored the Army wing of the research tree entirely as a matter of fact.

        You are, of course, entitled to your opinion but I don’t understand how you can call something overpowered and then give an example of how it was worthless.


      • Hello Expalphalog you are right, my example is not easy to understand ;-)
        My example was a game with the “World Generator Full Customization”-mod to make the world rougher and with more mountains. A great mod! It shows us the power of height level terrain (IMO THE great idea of EL!), but it shows us the lack of interesting terrain (the most terrain is flat) too. I hope for more games with height level maps like EL and extensivly usage of it. This make a battle with a small army winable.


  7. Great article. I’d like to add a few corroborating points. For the customized UI, the game “Birth of the Federation” (underrated IMO) presented a differently skinned UI depending on the race chosen. The controls were the same but their appearance changed drastically depending who you played as. This kind of “species specific UI skin” is something I’d like to see more of in future strategy games.

    My second point is about the end game. Nearly all strategy games have little reward for all the hours of effort put into them. The player spends long hours (fun hours, but a lot of them) working to win the game. When (and if) they do win they get a static image with some game stats and a “You Won” message. It feels like “OK, You won, Kay. Thanks. Bye.” Going back to Master of Orion, Master of Magic and Master of Orion 2, winning those gave a custom cut scene depending on the victory condition attained. The ‘custom’ part was just stringing together from a set of clips but it’s far, far better than “here’s your game stats, now go away.”

    My last point is about game balance and starting conditions. A standard trope of strategy games is to have various difficulty levels and then generate a random map to play on. If you get a awful map (as mentioned in the article) an easy game can be rather hard. If you begin with a really great starting position then a hard game isn’t so hard after all. How about this idea: make the major element of setting the game difficulty level control how good a starting position you are given. Want a hard game, expect few resources around you and few if any places to expand to. Set it to easy to get lots of local resources and plenty of places to expand to. This takes the pressure of having to have to tweak the AI to give it crazy bonuses for harder levels. Those bonuses can still be there, they’d get toned down a lot though.

    Thanks for a great article, just wanted to add in a few things.
    -= Far Oceans =-



Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s