The importance and challenge of 'speed bumps' In RTS economies

I've been playing a lot of Command and Conquer lately. From revisiting the original Command and Conquer Tiberian Dawn and Red Alert via the Remastered Collection, to spending time every week re-learning Kane's Wrath and practicing commentaries on game replays, to dipping my toe back into Red Alert 2: Yuri's Revenge via CNCNet just for the heck of it, Westwood's (and ultimately EA's) storied RTS franchise has been the water that I've been moving through lately. I've even been dabbling with OpenRA's take on C&C, Red Alert, and Dune 2000 and futzing around with RA3 and C&C Generals skirmishes.

I've also been continuing to play Universe at War and Grey Goo as I've grown to know many of the fine people who still love and play these games, and spending time in those communities has become a regular part of my afternoons and evenings.

While there are many things I dearly love about the C&C franchise, one of the things that I've seen time and again in these games is how awkward the pacing of these games can be, particularly in the early game... Whether it's Chinese Dragon Tanks melting base structures early in the game, or massed Nod Buggies destroying ConYards in overwhelming numbers in under 2 minutes; Shadow Strike Teams swooping in to take out infrastructure, and well, the list goes on. There's a tremendous volatility to the game that can be as frustrating to experience as it can be exciting and dynamic.

This is something I feel like Petrolgyph at least attempted to address in both Grey Goo and Universe at War. Both games have at least some C&C DNA and both implement changes to that gameplay formula which, in part, control the pace of the game. That is, both games attempt to put some kind of mechanical or systems guidelines in place to ensure that players don't end up in radically different game states before they're able to interact with each other. Grey Goo, I think, less so than UaW, but that's just me going down a rabbit hole.

And that got me thinking about why that is, and how C&C lacks what I'll call "speed bumps" or "speed limits" that are seen in other RTS, and how those speed bumps affect gameplay outcomes in RTS games.

That line of thinking kind of snowballed out as I looked at the larger implications of attempts by game designers to control or guide the pace of various RTS games: from Upkeep in WarCraft 3 or Company of Heroes to Match Phases in Dawn of War 3 or Steel Division, and the various pros and cons of these systems.

To try to summarize my point: in an effort to maximize fun and minimize the worst aspects of RTS gameplay, it tends to be better if game systems provide certain "speed bumps" which work on a systematic level to set a more methodical rate of army and economical growth.

However, designing mechanical "speed limits" which feel artificial or overly restrictive is a thorn in the side of players and can sour them on a game.

Let's try to flesh out the above 2 paragraphs a bit.

Speed Bumps and Speed Limits

I'm sure there are better names for these systems. They probably already have official designations somewhere that I've either forgotten or never learned. Please let me know in the comments if you know about them!

In RTS games, players are asked to build up an economy (of course!).

Typically, they are producing workers to harvest some sort of consumable resource from fixed spots on the game map (mineral lines, berry bushes, gold, lumber, et cetera) or, there is some sort of point or location on the map that is captured or built upon, as in Supreme Commander or Company of Heroes, which passively generates resources over time.

These resources are spent either to continue to expand the player's economic output, to expand the player's ability to produce military units - e.g. creating factories, to expand on what we call the "tech tree" - the power of the units and tools the player can bring to bear, or in the actual production of military units.

The speed at which the player is able to expand their economy and produce units and go up their tech tree etc is mediated by a number of things. Namely: they have a lot of expenditures they want to make but a limited pool of resources (money) with which to purchase things.

All strategy games have various systems by which they try to guide the player to a greater or lesser extent. This is in fact one hallmark of newer RTS - often, they try to minimize false choices and keep the games' tech trees more clear with branching options down which the player is expected to progress.

So, the player's economy and tech tree themselves provide "speed limits" on how quickly the player can progress: they can only spend the money they've earned, and their ability to expand their economy is limited by the money they spend on workers or refineries or Command Centers, et cetera.

What I'm really talking about here is the rate at which the player is able to change the state of the game. The faster they can grow their economy, the more options they have to alter the state of the game. This is controlled, internally, by player-controllable outcomes. I call these "speed bumps" - houses in WarCraft or Age of Empires, population cap upgrades constrained by the Quanta resource in Ashes of the Singularity, purchasing additional Barracks to produce troops faster (in a variety of RTS) - these are all "speed bumps."

They're ah... the cost of doing business? The cost in time and money of expanding your army; the cost of gaining access to more potent units, the cost of increasing your income rate. Et cetera.

With a speed bump, the game dictates these limitations or requirements on the player, but the player themselves determines how to approach the requirements; I know I need to buy a new house for every 10 guys; how do I fit that into the cadence of my gameplay? I know if I have 3 bases I can support production off of 9 Barracks; how do I build up so that I end up with that many Barracks and that many bases as fast as I can without allowing my opponent the opportunity to come crush me?

So, to me, a speed bump is a game system which gives the player an obstacle to progression that they have control over their approach to overcoming

Things like population limits with structures or upgrades to increase them provide a kind of cadence to gameplay in games like Age of Empires or StarCraft that isn't really as present in Command and Conquer. You can see this in those 2-4 minute C&C matches where one player decides to sneak an extra refinery and gets steamrolled by what feels like eleven-thousand Nod Bikes and Buggies. Or where one player tries to tech up and gets steamrolled by eleven thousand Nod Bikes and Buggies.

Command and Conquer games tend to go much lighter on these "speed bump" requirements, which can make gameplay more volatile. These games tend to settle into a meta where most games have a kind of agreed-upon cadence. We see this in Command and Conquer Remastered, where matches now tend to have a "safer" way to play and you see less zany/risky play from those who take the game seriously (tournaments, high ladder rank, etc). In C&C3: Kane's Wrath, too, it's common to see a kind of 'standard accepted build' that minimizes risk and establishes kind of an agreed upon cadence to the game.

Contrast this to Age of Empires, where aging up requires considerable economic investment for each Age you increase, and each Age opens up new tools to take the fight to the enemy: cavalry for harassment, artillery for destruction of buildings, et cetera. These speed bumps take the form of forcing the player to expand their economy in order to afford the purchase, and at times to actually shift their income priorities from, say, food and wood to a greater focus on gold and stone, for example.

Blizzard's games have taken various approaches to this sort of thing. In StarCraft and StarCraft 2, speed bumps are almost a 'sweet spot' between Age of Empires' regimented Age system (and plethora of upgrades) and Command and Conquer's cost of expanding economy or production (Refineries, tech, and factories are somewhat expensive).

WarCraft 3 however, introduces what I will call a "speed limit" in addition to the "speed bumps" of standard BlizzCraftRTS tropes like expensive economic expansion to new bases: the Upkeep system. This "speed limit" reduces Gold income by a fixed % once the player's population reaches a certain level. It's a system that the player has only limited ability to "game out" in that once they exceed the population threshold, their economy is automatically limited.

Dawn of War 3 also has a "speed limit" via its system of match phases, where player's economies are influenced by the match phase system in a way they have no control over.

Thus, a "speed limit" is a constraint placed upon players by a game that alters or restricts how they play without their input, control, or influence.

Both "speed bumps" and "speed limits" are a way to control or influence the pace or progress of RTS gameplay. Typically, "speed bumps" feel better to players because it allows them to take affirmative actions towards addressing the system.

You could call income itself a sort of speed limit as well, I suppose. For practical purposes though, players are in control over how quickly they go about expanding their economy, which makes it more of a speed bump type system in the terms I've defined above.

(Why) Are Speed Bumps Important?

Let me take a step back and talk about why this is important to me.

I've talked about resilience before in terms of homeostasis or equilibrium, or in terms of time-based disadvantages. Now, I'm trying to tackle another aspect this topic in terms of pacing or more specifically, in terms of systems that are designed to control pacing.

Let's look at the design of Homeworld: Deserts of Kharak. In the earliest stage of the game, the players face a very delicate balance between unlocking new units, unit production, and unit upgrades. Mis-timing an upgrade could leave the player very vulnerable to having their army crushed with no real hope of reprisal and therefore basically out of the match. As I mention above, this same sort of... fragility also is in evidence in many C&C titles: if I lose the wrong structure, or get behind in tank numbers, or pick a poor tech window (thus putting me behind in tank numbers) it could mean the end of my base and/or the match itself for me.

So anyway, to summarize: resilience to me is how easy it is to keep players in the "fun" or engaging parts of the gameplay and how easy it is for players to find themselves in the worst parts of the game's possibility space, whether winning or losing.

This is where I have to be really careful though. When I talk about 'resilience' or 'equilibrium' or write that I want the early game to be more 'resilient' or that I want the game state to be harder to shift too far in one direction all at once, it can easily sound like I'm advocating for a game space that's boring or a game design that lacks dynamic elements. That's not it at all.

What I'm trying to advocate for in a lot of my more recent writing is a design that reduces how often the worst moments occur in RTS, without the somewhat painful or annoying contrivance of 'speed limits'.

Because I'm going to be honest. To me, building Supply Depots or researching +1 damage is not inherently enjoyable. To the extent that it's necessary, I see these sorts of thing as a necessary evil. But, playing Age of Empires 4 recently got me thinking about the difference in terms of how easy it is to achieve poor gameplay dynamics in C&C games, and I think at least part of it has to do with the lack of pacing controls.

There's always going to be that moment in Company of Heroes 2 when you or your team get dramatically out-played and you're trapped in your base watching the enemy shell you as you struggle to get 1 or 2 squads out, with no real hope of turning things around. There's always going to be the moment in StarCraft 2 where you're crippled by a cannon rush or bamboozled by Dark Templar or struck down by a Medivac drop. That's going to happen.

People are going to be out-played and out-maneuvered. That's point of the whole damn thing after all. What I'm trying to explore right now are ways to decrease the likelihood of the worst game states from occurring. To make it more likely that more times the game is played, more players spend the most amount of time possible in the 'fun' parts of the game.

Again, an illustration may be useful.

I spent quite a lot of time the other weekend in the Age of Empires 4 Technical Stress Test. It's the first I've really interacted with an Age of Empires game in a meaningful way since maybe 2011 or 2012, when I spent quite a lot of time in Age of Empires Online. I've never been a huge Age player myself - to me, it emphasizes things like upgrades and the construction of large, fortress-like bases, while my caveman brain is more pleased by the clash of armies: fast, and often, if possible!

That being said, one thing I've observed in Age of Empires games is the pacing. It starts the player off with largely slow units, opening up into cavalry, and finally delivering the player siege weapons and the suchlike in the later stages of the game. Rushes are still possible, and players are in constant danger of having villagers killed in large enough numbers to have an impact on their ability to close out the game.

But, in an Age of Empires game, the early stage of the game is still somewhat 'resilient' in that the player is given ways to protect their villagers (garrisoning in structures, town hall arrows, wall-ins with buildings and actual walls). Into the midgame, more sure harassment options are provided, and into the later game, players are given more game-ending ability in the form of units designed to efficiently take out structures.

This sort of progression, this sort of pacing if you will is a good example of what I'm talking about, especially as contrasted against my examples above regarding Command and Conquer, where in the Remastered Collection, a 2-4 minute match length is far from uncommon.

Company of Heroes has a similar sort of pacing to it that keeps players from straying too far outside of the current rough game state. It does have some rough edges, such as the moments when one player gets out the first vehicle, when the first tank comes out, and then in the later game, when tanks are more common and the heaviest vehicles start to hit the field. These power spikes can feel really frustrating to players on the receiving end of them.

Speed Bumps and Speed Limits - What Can We Learn?

Manpower in Company of Heroes games is a 'speed limit' as well: as the number of units a player fields increases, the rate of income generated decreases. This helps to create kind of a natural cadence to the game, offset by things like units' move speed, the rate of capture of points on the map, et cetera. I think this upkeep system feels better than the one in WarCraft 3 because it is gradual and granular, not firmly banded.

Also, income generation in COH2 is already passive and a matter of holding ground instead of building workers and creating expensive base expansions. It's just something that kind of happens in the background.

Dawn of War 3's game phase system has a pretty negative effect on the game, sadly. It was, I think, an attempt to smooth out the pacing of the game, but in practice it doesn't seem to do that very well at all. The economy scaling feels too much into the late game, while the early game is fairly fragile, a fact which isn't helped by resources from dead units being recouped by the player who lost the unit.

Age of Empires - high cost to tech to new age, has a good impact on game pace in terms of making the early game a back and forth thing: villagers can be protected with house-walls, or hide in structures like the Town Hall; it takes a while for good harass options to come out, and longer for structure damage to come into play. There's a lot of picking around at the edges in Age of Empires, a lot of it driven by speed bump moderated systems: Aging up, a plethora of upgrades, plentiful minor bumps in population supply via houses, all of that.

In Command and Conquer Remastered - the APC tech had to be moved due to the game degenerating into constant APC rushes - which bad for player experience.

And yet, to me, Age of Empires games have a tendency to feel constrained by their abundance of speed bump systems. These games take a long time to get going, historically, and to me at least tend to feel like there's a lot of buildup before any sort of reward is realized.

So, clearly, there's kind of a 'sweet spot' - or, a variety of sweet spots perhaps - where RTS titles can introduce speed bumps and pacing that enables more balanced interactions without degenerating into something that feels too onerous a burden on players.

Walking the Tightrope

As I said before, I don't, really, like speed bumps in RTS design. Building houses, researching upgrades, it kind of feels like it's getting in the way of the "fun stuff" - that is, smashing armies against each other in a battle to see who will come out on top. But, it seems like these sorts of "housekeeping" systems can be quite helpful to create a better experience for players.

Would Command and Conquer be better if it had houses or some other metaphor for expanding player supply? ... Possibly? I doubt many of the people who love C&C would be thrilled with that idea, though such a system might help reduce weird edge cases in the early game in particular. But it might also reduce the fun of the game.

Though of course it's particularly hard to talk about modifying an existing beloved franchise in a meaningful way that wouldn't upset some or many of that franchise's core fans and players.

I think if I were to try to draw a rule or conclusion from my above ramble, it might be something like this: speed bumps are necessary. There's going to be spots in an RTS where players need to be throttled a bit. And upgrades, houses/supply limiters, economy expansion systems... These are some tested methods to generate this sort of speed bump.

But, while there might be such a thing as too little throttling in terms of gameplay, I would also argue it's possible to do too much as well. So, for me at least, it seems clear that such systems need to be used sparingly. As little as necessary to reach the desired outcome, in other words.

And player autonomy and time in particular, needs to be respected. Hard cutoffs are easy to design, like WarCraft 3's Upkeep, or Dawn of War 3's match phases. But they chafe players in ways that building pylons don't since the player feels like they lack control over outcomes involving these systems, and like they're being punished for playing well. Where a speed bump might allow the player to play well by managing its particular challenge.

Speed bumps alone don't seem to be enough,in a way. They need to work in conjunction with other systems to create the sort of fun that RTS players demand and deserve.

Universe at War has better pacing and gameplay resilience in many ways than Deserts of Kharak, which has more speed bumps in the way of upgrades, supply cap increases, and those sorts of system than Petroglyph's 2007 RTS.

This comes down to other elements of each game's pacing: UaW allows players to peak out their economies pretty fast, putting them on relatively even footing early. Also in UaW, buildings can take a beating against many troop types while also being built pretty quickly, and defensive structures are quite powerful for their cost. Faction and army design also plays into this, with players able to do things like summon temporary units, stun enemies with mines, crush enemy units, mind control them, erase them from existence, and a variety of other pretty high impact effects, some present in the early game.

Powerful defenses is one thing you see in both Age of Empires and some C&C titles to help even out the pace of combat a bit, which introduces a tech and army size check of sorts, and is one reason C&C introduces super weapons to end late game stalemate, while Age of Empires 4 has both Wonders and Sacred Sites... but I'm getting a bit off topic I suppose.

This has been a bit of a tough topic for me, since my instincts run a bit to the contrary of what experience is telling me is best for the genre. It's a circle I'm having to square a bit.

This isn't to say that every game needs the specific speed bumps seen in things like Age of Empires or StarCraft: there could be systems out there that work better for specific games. The Gaalsien in Deserts of Kharak use upgrades on their Carrier to increase population capacity, for example, as do the factions in Ashes of the Singularity with their population upgrade system that runs off the game's 3rd Quanta resource. These upgrades do prevent population cap from being harassed once enabled, which would need to be accounted for in future games that look to use systems of that nature.

The resource cap in C&C games and in Grey Goo is supposed to work as a sort of speed bump, I think, and has a lot of interesting interactions (stealing silos with Engineers to steal resources, or killing them to reduce enemy resource storage) but in practice this doesn't work as well as it might due to the very high initial resource cap, which is seldom exceeded by players in the first place. Forcing some technologies or structures to cost more than easily attainable would be a good way to actually have this system work as intended. You want to buy Tier 3? Better get some Silos and save up your resources! Otherwise, they don't really serve a purpose unless the player is not spending resources the way they should.

Anyway, I think I've rambled my way around the topic enough. Thanks for reading. Looking forward to seeing your thoughts.