Game Design Games Written by Wayward

Food, Gold, and Beyond: Exploring Multiple Means of Generating Resources in RTS Games

Thanks very much to my Patrons, who supported this article.

Right now, Age of Empires IV is a big topic of conversation in the RTS space. It’s among the largest modern RTS launches, and while there has been some spirited discussion of the state of the launch, the game’s balance and the feature set with which it released, the window around Age IV’s debut compares favorably to that of Company of Heroes 2, Halo Wars 2, Grey Goo… I could go on. Regardless what you think of the game personally, I think it’s fair to say this RTS is a pretty big deal at the moment.

While it remains to be seen how (or if?) the game will grow over time and how fan and player concerns will be addressed, it’s definitely jogged a ton of conversation in RTS spaces online. From art style to camera zoom to expected launch feature sets; ranked multiplayer, faction design, and more. There’s a lot of conversation on Discord, Twitter, forums, and elsewhere, all inspired by the development and launch of this new strategy game from Relic, Microsoft, and World’s Edge.

Among my very favorite elements of the Age of Empires series are the complexities and optimizations present around its food and gold resources. While many of the systems specific to food are hidden (a topic for another day perhaps?) and difficult for the average player to determine without outside assistance, each source of food comes with its own pros and cons, and best sources at various stages of the game.

On top of this, various civilizations have their own benefits and drawbacks to food collection: Delhi and Abbasid players, for instance, gain extra benefits from Berry bushes but cannot hunt Boar; the English have superior Farms, and China’s farmed crop is rice, which has different regrowth rates etc.

All of these various things give the player a lot to think about and a variety of efficiency levels for generating food income. Now, there is a clear hierarchy of which food sources are best to use, which can be a bit of a challenge when trying to create a system like this, but Age of Empires actually bakes that into the equation. I talk about that below.

See the source image

Gold in Age of Empires IV is generated in an even wider variety of ways: Traders make trips to neutral towns and back and generate gold based on the distance they travel; players can store captured Relics for gold income, the Chinese have an Imperial Officer which generates gold based on visiting player structures to gather taxes, the Rus have a Hunting Cabin that generates gold based on the number of nearby trees, holding Sacred Sites on the map generates passive gold income, and more (most civilizations have some special feature related to gold income, some of which I mention above).

Food and gold are 2 important resources for players to generate in Age of Empires: gold is the game’s “tech” resource and is used extensively by high tier units and many upgrades, while food is the primary resource for population and army growth. These two resources are particularly bad for the player in terms of short and long-term impact on the game if their income is interrupted, and their generation gains tremendous benefits from a gameplay perspective by having so many ways to acquire them across the course of a match.

Wood and Stone are important too, of course, but I can see why from a design perspective their harvesting is left somewhat more simple and straightforward: wood is spent mostly on production and economy, and stone mostly on defenses. The expenditure of gold and food more often impact the game in more direct and dynamic ways, and their manipulation is often felt keenly by players.

What’s the Point?

See the source image

Hello, I like Money!

A couple of years ago, I wrote an article about resource system design. I’m hopeful that this piece will compliment that one and serve as a good companion piece, as opposed to rehashing the same ideas. In this article, I’m specifically considering the benefits and challenges of adding multiple discrete and largely or entirely unconnected ways to generate one or more of the game’s resources.

To be clear: this article is not an exploration of the topic of how many resources the game itself boasts. Whether it has 1 like Command and Conquer, or 4 like Age of Empires IV, or umpteen like Offworld Trading Company, is really beside the point. Instead I’m looking at how many methods are available to a player to generate any given resource, and the what impacts are on an RTS of having more than one way to gain that resource.

And the first question to be answered is, why would a game want to provide multiple ways of generating one or more of its resources in the first place?

First off, to show where I’m coming from so that we’re clear: I’m very much in favor of RTS systems design which attempts to require a player to win multiple engagements, over time, before the game is definitively decided and one player is unable to continue to meaningfully participate in the game.

I have looked at this in the past in terms of time based disadvantages, equilibrium, or speed bumps which would keep the player who’s ahead from continuing to accelerate towards victory. I’m attempting to keep this particular article a little more agnostic since I feel I harp on that idea too much, but I do think that providing multiple means of acquiring resources can assist in making any 1 act of harassment or disruption to lead down a slippery slope, which is one reason I’ve been looking more into such systems lately.

I should also say that, on the converse side, it is likely to provide more opportunities for enemy disruption and harassment, increasing the dynamic and back-and-forth feel of multiplayer games as there’s more ‘surface area’ for players to mess with each other. I actually see this as a positive: firstly, I believe quite strongly that optimal RTS design makes it almost impossible for a player to engage with the game in an ideal state. Giving the players multiple income streams, but allowing each income stream to be riddled with inefficiencies and harassment/disruption, reduces the players’ ability to mentally model the entire game state and reduces the game’s ability to become ‘solved.’

More ways to generate income can lessen the impact of any single method of income loss while increasing each player’s ability to impact the overall resource income of their opponent via harassment or disruption. This is especially true if the resource generation methods are at least partially independent of one another.

See the source image

One good example of this is Act of Aggression.

In AoA (the base game, not the reboot edition), the player’s primary method of income generation is a refinery which is built adjacent to resources deposit of either Oil (yellow) or Aluminum (blue) or Rare Earth Elements (red). Additionally, injured infantry or the pilots of destroyed vehicles can be captured in various ways and sent to prisons, which can sell the prisoners for various resources. Lastly, Banks can be garrisoned by infantry in order to generate passive income over time – the more units garrisoned, the more income they generate.

While in this case infantry, or units in general, are kind of an important component of the system, once set up, combat alone is sufficient to garner potential prisoners, while banks are ripe for capture or can just sit alone and generate income over time.

So, while AoA does have its issues, it does serve as a fairly good example of partially independent systems of resource generation.

The prisoner system contains inherent counter-play, as do banks, and the main system of income is vulnerable to harassment. However, I found the system to be more than a little fiddly when actually playing the game: either collecting enemy POWs or saving your own is a bit of a chore and could have done with, in my opinion, some serious polish. Any sort of system around this to make either retreating or capturing less of a struggle would have greatly improved the feel of this system, which I feel overall has a lot of promise.

Perhaps it might be worthwhile to do a comprehensive overview of AOA’s strengths and weaknesses at some point. I’ll have to give that a think.

Creating more variables increases the difficulty of optimizing systems, especially if some system(s) have built in inefficiencies. RTS games run the risk of their players optimizing and streamlining to the point where the depth of their systems is stripped away: too great a focus on pure DPS output makes combat more flat, for instance; this is a similar principle. Don’t put all of your game’s economic complexity in one basket: give players various tradeoffs to manage in their pursuit of their critical path through the game’s systems.

Additionally, multiple partial points of failure expands the space to mess with one’s opponent while also reducing the ability for any one act of disruption to cause a killing blow. Consider Company of Heroes and Supreme Commanders ‘resource points’ – the player wants to own as many as they can, and stealing them from another player is a common part of both games. This, however, diverts the player from dealing immediate, crippling damage to their opponent, and each theft (particularly in SupCom’s case) only has a marginal fractional impact on either player’s economy.

This is akin to how I think of objective-based design. It blunts the all-or-nothing feel of some RTS systems in favor of incremental changes which can be reinforced or reversed over the course of the match, and diverts army strength and player attention into multiple, smaller tributaries. It’s also kind of opt-in complexity in many cases, where the core gameplay doesn’t necessitate players engage with alternate income methods, but presents these options for more competitive and experienced players to leverage advantages.

This is, I think, a very important lesson from Age of Empires in particular (Northgard does this too; I cover this later in the artile).

For gold, the resource is so important to the player that they’ll always want to find as many ways to generate it as possible: mining is kind of just the baseline. Relics, Sacred Sites, Traders, Chinese Officials, et cetera – Gold quickly becomes the main constraining factor in the mid and late game and the player is only rewarded for getting it in as many ways as they can.

Gold, however, relies on the fact that each method of generation is an important and separate part of the greater whole. Just mining, or trading, or storing Relics, only takes the player so far. Changing the balance of these sources of generation against one another could easily change players’ mental calculus such that one or more sources of gold would be easy to ignore. Which brings us to food, and how that resources efficiencies shape the game.

Food, meanwhile, runs out in every case aside from the cost-inefficient method of acquiring it: farming. Sheep, berry bushes, fish, and hunts get used up over time, and the player must always be looking to find the next most convenient source of food lest this important resource run out.

There are levels of efficiency to food, and a hierarchy of importance for gathering it, but regardless the player knows that in most cases they’ll eventually resort to some farming. But, holding off on spending all of the wood the player will need to switch from other food sources to farming is pretty essential: common wisdom among competitive players is, building a farm too early can lose you the game.

Thus, the player is highly incentivized to find multiple ways to generate food, and constantly plan ahead for the next convenient source(s) of it, a much different sort of dynamic to gold, all told.

And, importantly, the player is forced to eventually use less efficient methods than they’d prefer, due to the more efficient methods being used up over time. Therefore, the game’s core design requires the player to utilize all or most of its gathering systems anyway. This is a very important note, I feel like: Age of Empires systems design leans into its own inefficiencies and gives them a purpose and a strategic impact beyond just punishing new players: it causes a wide and varied engagement that is resilient against over-optimization.

Definitely a good trick for a designer to keep in their back pocket.

Challenge and (Attempted) Response

While there are some downsides to adding multiple discrete methods of resource generation, I think that in most cases, the benefits outweigh the issues. Let’s take a look though, at the downsides (or potential downsides) so we can see what we’re dealing with.

See the source image

First and foremost, complex systems don’t necessarily create depth. This is probably more than a little obvious, but it bears addressing right out the gate. The various method of generating the same resource have to be created to be complementary, and to have meaningful benefits and drawbacks that aren’t clear cut in every circumstance. The concern, of course, is that with an obviously better choice to make, the player will be able to ignore the apparent depth in favor of only one method of resource generation, creating what are effectively false choices instead of providing the player with an array of methods to keep filling their coffers, so to speak.

RTS players and communities can be powerful optimizers, and systems which present clear efficiencies, and/or which are much easier to benefit from, will mostly be chosen over those which are less optimal. In fact, less optimal systems in that case become ‘newb traps’ which just serve to disadvantage less ‘in the know’ players, without clear methods for educating them into the efficiencies around which the game is really played.

Balance is important – trite to say, difficult to achieve, true either way.

However, we do have a couple of methods we can use to offset this particular difficulty. As mentioned above, in Age of Empires, you’re going to end up using those less efficient sources of food because of how quickly it’s used up and how important it is. So, resource volatility or consumption rate can force players to use less efficient methods of generating a source of income, and various skills such as planning ahead, aggression (stealing from the enemy) and the like serving to allow the player to continue to gathering more effective resources for as long as possible.

Degenerative outcomes is another problem with this sort of system. In Iron Harvest, players are able to gather Iron and Oil via pickups on the map. There are also pickups for other things that aren’t resources: weapon swapping and crewable weapons are available too. These pickups, while a fun idea at first, actually have a bit of a negative effect on the early minutes of the game.

Iron Harvest’s resource pickups are absolutely vital to securing a player’s early economy (we won’t talk about weapon pickups, which have their own issues, but at least provide another early target for a huge military boost in the early game) and falling behind on securing them is as much a tactical as a strategic problem. Allow the enemy too easy access to these insta-grab resources, and you’re looking at having mechs knocking at your door before you’re ready. Or worse, one of the game’s very cost effective hero units.

Which… is not what you’re hoping for in the game.

See the source image

Halo Wars 2 has a better pickup system, where units drain the resource cache a little at a time, forcing the player to commit the unit for an extended period of time, and allowing for some degree of counter play as players can chase an opponent off a cache and steal part of it for themselves, or to sneak fast units into the enemy area of the map in order to sneak some of their opponent’s resource caches (without necessarily consuming the whole resource value). However, even this gameplay strongly favors fast and hard aggression and can really ramp up a game’s early game and unlock game ending tech a little faster than might be ideal.

Typically when designing a resource bonus system, what I’d look to do is provide granularity and delay: pickups are a very binary outcome and are front-loaded, which can introduce a ton of randomness to gameplay outcomes that isn’t necessarily ideal. I cover this a bit more in depth later in the article, but Command and Conquer’s Tiberium Spikes and Oil Derricks are I think a very good example of this system which account for both aggressive action and defense over time.

It almost feels like it would go without saying, but the added complexity of multiple income systems is something to consider. Age of Empires 3 and 4 both have fairly complex maps and minimaps with everything they have to include: neutral towns for trading, deposits of all of the games’ resources in all their various on-map varieties, This puts an additional burden on the player in terms of knowledge and in the map-maker in terms of composition. It’s not for nothing that Age of Empires 3 can feel a bit daunting to get into (though I will say a significant part of that is its shipments system)

As a better example, perhaps: if you compare Warcraft 3’s gameplay to that of StarCraft 2 at a macroeconomic level, you start to see a bit of this picture. StarCraft 2’s gameplay is a lot more immediate and approachable since the player doesn’t have to worry about hero experience, creep timings, on-map structures, item management, and more, and I’m not sure about the validity of the argument that Warcraft 3 is meaningfully deeper than StarCraft 2, especially if the depth/complexity ratio is specifically considered. It seems to me that the added complexity of Warcraft 3 hasn’t bought commensurate depth in its gameplay systems and outcomes, if you follow what I’m saying.

This isn’t to throw stones at Warcraft 3. It’s a phenomenal game in a lot of respects. It’s just an easy way to demonstrate, relative to another Blizzard title, the comparative increase in complexity added by its creep camps et cetera. It’s good to keep in mind.

To summarize: RTS with multiple ways of gathering 1 resource can lead to ‘traps’ where many of the methods can be useless. There are potential solutions to this. Additionally, such systems could result in degenerative gameplay outcomes – long term recurring benefits tend to be better than one-shot single pickups, particularly in the early game. And lastly, the additional complexity burden of multiple resource systems needs to be accounted for.

Some Examples

See the source image

Halo Wars 2 has some other interesting examples of multiple ways of gathering resources. First and foremost, all resources in HW2 are generated primarily from build slots in bases. So, the more bases you have, the more ability you have to acquire resources in this way. Also, in the early game, players must rush to gather resources from pickups spread around the map as a free and quick way to get an early game boost. Finally, Power is generated not only from structures placed on build slots but also from capturable areas in the map, providing other points of contention aside from bases and allowing the player to attempt to save build slots by capturing these points in the middle. It seems to work fairly well in practice in this game.

In Command and Conquer games, players often sell off buildings they don’t need any more (in some cases, units can also be sold off as with the Soviet Crusher Crane in Red Alert 3 or the Grinder in Yuri’s Revenge) as a way to recoup build costs. Also, Oil Derricks or Tiberium Spikes also serve as alternate income methods. Generals perhaps does the best at this out of all of the C&C games, with Chinese Hackers, Allied Supply Drops, etc, allowing for late game income after on map resources run low.

I think that C&C might not do quite enough in this regard on average: the income is barely supplemental at best in most cases. Generals is a step in the right direction here, I think, with its various methods of late-game resource production that pick up as the main income is consumed. This is, of course, on top of Oil Derricks. But in C&C, the % of steady income provided by Oil Derricks or Tiberium Spikes feels a little low as a source of supplemental income.

Command and Conquer 3: Kane’s Wrath has the GDI MARV, which passively harvests Tiberium it moves across, and the Scrin’s Eradicator Hexapod, which passively gains resources when enemies die nearby, as alternate methods for late-game income. These systems, while interesting in their own right, might actually be more interesting if not placed on a pivotal, potentially game-ending unit.

The idea of Scrin being able to gain resources from dying enemies, or the GDF being able to send units to mow through Tiberium fields, has a lot of implications that it might be interesting to see in a slightly different context than attached to something as monolithic as the MARV or Hexapod…

Iron Harvest has an interesting (as mentioned above, this system has some issues) spin on this, with significant early game resources held in the form of on-map pickups that players must rush to capture and acquire lest they lose out on the race to have powerful units available earlier than their opponent.

Interestingly, Tiberium Spikes and Oil Derricks are kind of the middle point I was talking about in my Iron Harvest example above: the first time they’re captured, they provide an initial burst of resources (helping the player to grab them first and benefitting quick reaction time) and then provide a steady trickle to whomever owns them thereafter. Possibly, it might be interesting to have that resource boon periodically reset itself or be able to be activated in some way, but overall I think it’s a solid take on ‘long term income’ though I would argue they might be under-utilized in C&C…

See the source image

Northgard is another example of multiple routes of resource generation, with at least as much interaction depth as we see in Age of Empires. Gold in Northgard is generated via trade routes from the trading post, direct selling of goods via market, workers on trading posts and merchants, workers on docks, and exploring ruins. Food, meanwhile, is obtained in a cyclical manner – the game contains a summer and winter season, and food income types are variable across these seasons. As with Age of Empires 4, different clans in Northgard may also have unique ways of generating one or more resources.

Farming is good in summer bad in winter, allowing the player boom and stockpile for winter; hunting is middling with average income between all seasons; fishing is better in winter, which means the player’s food economy is more resilient in winter, but this method is overall not as good at booming. Herbalists are dual purpose units: they’re healers who, when idle, also generate food. This is a great way to increase tile food generation density but they do not collect food when healing, so over-reliance on them runs the risk of famine during wartimes.

This sort of unit-based passive income is a great source of utility and depth; the Age of Empires Priests are actually somewhat similar, with their utility regarding Relics and Sacred Sites as well as in combat – the Relic itself is a tool that can be used in a variety of situations and not just useful for generating passive gold, and its utility, too is something that I think is quite clever.

Of course, as stated above, the Relic must have all of its various utilities balanced against one another in order for it not to be too obvious or easy to choose how to use it, but in the case of Age of Empires 2 and 4 at least, it seems to have hit its mark.

Tying it Off, Finally…

See the source image

Majesty is another game which has multiple methods of resource generation

I suppose we come full circle to the question: “what’s the point?”

For me, this article has mostly been a mental exercise and categorization: Having not been a big Age of Empires player for many years now, I find myself looking at its systems anew and appreciating their effects and possibilities. Thinking about how food and gold are acquired and play out in various circumstances, I started drawing connections to how other games handled their resource generation systems, and saw a number of threads carried across numerous titles.

To me, the idea of multiple avenues of generating a game’s resources is interesting and exciting almost precisely because this can be used to create systems which are more granular and difficult to disrupt wholesale, which can lead to the sort of resilient gameplay I have been chasing in my design and writing for a while now.

Thanks for reading.


  1. Nice writeup, Brandon! You did touch that point, but I’d suggest going a bit deeper (probably in a future article) about late game resource Reinforcement (unit-based passive income as you call it, mostly “in base”) vs Infinite Economy, which seems to be at the root of most of the design issues to be solved in RTSes.


  2. Good article. I think you make a good point about making sure each way to gather a resource should have a place in the economy. I’d like to see more on how Age of Empires balances many different ways to gather resources. One thing I thing the Age series does well is allow you to upgrade different methods of resource gathering, allowing players to specialize how they get a resource. Sup Com seems to lack this, where you can’t upgrade one or another method of gaining mass or power, which, I think limits the economic optimization game. Also I think the article ended a little abruptly, and would be interested in a sequel where you go more in depth on AoE4.

    Liked by 1 person

    1. The on-way conversion (energy to metal/mass) used by Total Annihilation and Supreme Commander are justifiable for the games’ reliance on metal as the only “real”, map-based resource, since energy can be built anywhere and is as infinite as the main resource. That said, upgrading the energy and harvest devices would make sense, exactly to promote holding the territory and saving up space (for energy structures) on smaller maps, it’s a feature we’re actively using in our game, TAP.


Comments are closed.

%d bloggers like this: