Hey folks,

Today I’ll be proposing a breakdown of competitive real time strategy games into two distinct sub-genres – “classic” and “modern”. To do this, I’ll argue that StarCraft II is not a successor to Brood War – it belongs to a different sub-genre.

To be clear – this piece does not take a position on which game is “better” or “worse”.

Special thanks to RAPiD for providing feedback on this piece prior to release. Special thanks to Lalush and AVEX, whose conversations have contributed a lot to my thinking around this topic.

The Two Types of Competitive Real Time Strategy Games

Over the past two decades, a number of real time strategy games were released with either a partial or fully committed vision of a competitive multiplayer scene. Titles progressively coalesced around a common feature set and way of doings: professional playtesters to help balance the game, in-game ladders and automated matchmaking, periodic balance patches, financial and logistical support for competitive tournaments, and so on.

A hallmark of this period was progressive consensus around design paradigms that players did and did not want. One of the most significant was increasing focus on the strategy component of real time strategy and reducing focus on its real time or mechanical component, particularly around designing a convenient and easy to use user interface. Bruce Shelley, one of the lead designers of the Age of Empires series, noted this explicitly when discussing the re-design of the farming mechanic in moving from Age of Empires I to II.

This meant design decisions like streamlining macromanagement (multiple building select, rebindable hotkeys, etc), micromanagement (stronger AI, unit pathing, waypointing, etc), and simplifying economic management to encourage more focus on military strategy. Designers targeted both the casual and competitive scenes – mechanics were progressively simplified, but there was also lots of work done to ensure a good environment for competitive play. There are plenty of examples of this:

• The inclusion of Age of Empires III, Dawn of War, and Command and Conquer in the World Cyber Games in the mid-2000s. The WCG required companies to sponsor their games in order to include them in its tournament. The inclusion of these titles indicates companies were willing to put concrete financial support into their competitive multiplayer scene.
• The support for televised matches of Age of Empires II and the direct engagement and solicitation of feedback from professional players both indicate strong, tangible support for a competitive scene.

My point here is that the design shift toward emphasizing strategy was not an attempt to “casualize” the genre – companies viewed support for a competitive scene as vital to their interests. Here’s another quote from Mr. Shelley regarding the commercial success of Age of Empires:

How do you think Age of Empires ended up with such a large online following?

BCS: I think it sold well overall because it appealed to both online/hardcore gamers and casual gamers. There were a lot of different experiences in the box and players of all tastes could find a game that really suited them. For everyone the game was bright and inviting. For online particularly, I think it was well balanced and challenging, and always a little different thanks to randomly generated maps. We had playtested it for years in multiplayer mode (the artificial intelligence for skirmish mode was finished late) so it was fairly well polished for online play by the time it published.

The progression in design from mechanical to strategic focus was instead considered a modernization of the genre, increasing its accessibility to the mainstream market and making the gameplay experience more enjoyable. However, I’m going to argue that it achieved something completely different – it created a distinct, separate sub-genre of competitive real time strategy games that is fundamentally distinct from the “classic” competitive real time strategy games that preceded it.

To do this, I’ll argue that StarCraft II belongs to this “modern” genre, and is therefore not a true successor to Brood War, which belongs to the “classic” genre. I’ll also include numerous references to Age of Empires 2, which shares a number of design characteristics with Brood War and whose successors took a “modernization” approach similar to StarCraft II.

Design Fork-In-The-Road #1: Mechanical Execution vs. Strategic Execution

Classic real time strategy games emerged in an era where they were viewed as strategy games that were not turn-based. As a result, they placed emphasis on mechanical execution – building units, placing structures, researching technologies, etc – because that was the key value-add for players coming from a turn-based background.

Modern games’ progressive reduction in focus on mechanics in favor of strategy and decision making had a dramatic impact on how games played out. Classic real time strategy games rewarded the player who could build the most units. It was mechanically impossible to maximize production efficiency by the mid- and late-game while simultaneously doing everything else. Players that were more mechanically skilled could win by simply building a lot more stuff than their opponent.

The modern approach changed this paradigm by streamlining the mechanics of production and emphasizing the right units – choosing a composition that effectively countered your opponent and adjusting it as the game went by. The streamlined production mechanics made it more difficult to win by sheer mechanical force, so players placed more emphasis on building the right units.

This extends to economic management as well – the sheer size of mid- and late-game economies and their spread across the map in classic real time strategy games required enormous mechanical skill to effectively manage. The limitations in the user interface contributed to this, such as simplistic or out-right missing waypointing systems, which forced players to do lots of simple mechanical actions. Modern games streamlined user interfaces to make economic management easier, allowing players to focus more on what their economy was trying to achieve rather than whether or not their economy was actually doing anything.

Army control was affected, too. Controlling a maximum supply army in a classic real time strategy game was extraordinarily difficult due to things like poor unit AI, weak pathing, and limitations in the user interface around control groups. This created a comeback mechanic in which superior mechanical control of a smaller group of units by the losing player could defeat a winning player managing an uncontrollably large army. Modern real time strategy games change this. Armies are mechanically easier to control, meaning that a player with a larger army will usually defeat a player with a smaller army. Comeback mechanics instead center around strategic decisions like harassment or tech switches.

Positioning and tactical battle planning changed as well. It was difficult to perfectly control armies in classic real time strategy games – as a result, simply moving units to the battlefield and getting them to engage took precedence over taking the right position and planning out the overall movement pattern of an army. There was also more incentive to fight whenever possible – a higher percentage of engagements were viable because their results were more dependent on control than composition. Modern games reduced this mechanical burden, allowing players to spend more time planning their battles and winning by taking a superior position – in other words, focusing on more strategic tasks. This also means more emphasis on poking, prodding, and re-positioning over direct engagements.

The shift in focus from mechanical to strategic execution creates a shift in how games flow: more focus is placed on the trade-offs in a player’s strategic plan rather than trade-offs in the execution of their existing plan.

Design Fork-In-The-Road #2: Economy as a Playstyle vs. Economy as a Simple Input to Military

Classic real time strategy games generally featured three core playstyles that emphasized different skill sets. Age of Empires called these “boom”, “rush”, and “turtle” – economic focus, military focus, and unit composition focus. The general RTS terms I heard for these playstyles back in the day was “macro”, “micro”, and “turtle”, which represent the same basic concepts.

Modern real time strategy games progressively de-emphasized focus on economic play, viewing it as a burden on a player’s ability to focus on strategy. Economic playstyles of classic real time strategy games were particularly mechanically burdensome; managing numerous bases in Brood War or managing a near-one-hundred villager sized economy in Age of Empires 2 largely consisted of doing endless numbers of simple, mechanical actions in order to keep economies going.

(As an anecdote, competitive players complained about farm auto-queuing in the expansion to Age of Empires II, noting that it would dramatically reduce the game’s skill ceiling. For those who didn’t play the game, farms automatically ran out of food after a certain period of time, and players had to manually click each one and re-seed it as the game progressed. This was pretty challenging, with players having dozens of farms in the late-game and needing to re-seed constantly. The expansion pack allowed players to automatically re-seed farms by queuing future farms at their Mill.)

The modern approach was to instead target economy as a simple input to military units – in other words, simplify and reduce economic mechanics to flatten the economic differences between different playstyles. This encouraged players to focus on their unit composition and tech choices instead of their economy.

This shift in design had a substantial (and probably unintended) consequence: it fundamentally altered the dynamics around cost-efficiency when players chose to turtle.

Economic playstyles in classic real time strategy games centered around the idea that, by having a much larger economy, the economic player could trade units inefficiently against a turtling opponent and still come out ahead. This gave the economic player great incentive to expand across the map (or, in the case of Age of Empires, place lots of town centers), amass a huge number of workers, and constantly attack their opponent to prevent them from building a superior unit composition.

The turtling playstyle responded to this by emphasizing a highly cost efficient unit composition. If the turtling player could survive the constant attacks from their economic opponent, they could slowly move across the map with a significantly superior force.

This was an interesting, action-packed design. It limited the number of games where players played passively while maxing out their supply. The economic player was attacking constantly and building across the entire map, while the turtling player was scrambling to stay alive.

This wasn’t limited to the economic against turtle dynamic, either: the strength of expanding one’s economy, the ability to trade and engage the opponent’s army directly, and the need to spread across the map created action-packed games when one player rushed or when both players opted to play economically. Here’s a few examples.

Modern real time strategy games remove this dynamic by equalizing the differences in economy between different playstyles. Rather than players trading off between a focus on economy and military, every macro style has a similarly strong economy – the difference is the military strategy and unit compositions that are chosen. This places a huge emphasis on cost-efficient trades, which typically favor the defending player. It also means that turtling playstyles produce significantly more passive games because there is no longer an incentive to attack into a more cost-efficient force: players instead harass their opponent’s economy or, more likely, choose to turtle themselves into an equally good or superior late-game unit composition.

This is pretty easy to see if you compare Age of Empires II with its successors – large, “booming” economies are replaced by economies that serve only as an input to the ideal unit composition. Players tried as hard as they could in Age of Mythology and Age of Empires III to limit economic growth whenever possible and play the races with the most cost-efficient characteristics – Set due to his animals in Age of Mythology, Dutch due to their banks in The War Chiefs, and Japanese due to their Shrines and Musketeers in The Asian Dynasties. Here’s a quote from a guide written to help players adjust to the changes in Age of Mythology:

Compared to those in AoK, villagers in AoM work much more efficiently – you won’t need as many as you did in AoK. By the time you reach Heroic Age you shouldn’t need to build any more, unless you have lost a lot of them due to fighting.

It’s no coincidence that the same article also contains this advice:

Avoid fighting enemy soldiers if you can help it, and focus on attacking your opponent’s villagers and economy. Only face an enemy army if you are sure of a victory or if he is threatening your own economy.

I imagine most of the folks reading this blog post come from a StarCraft background, so I’ll go into more detail about the differences between the Brood War and StarCraft II economies.

(I’m not as familiar with Brood War as I am with Age of Empires 2, so I’ll take this opportunity to note that I am incredibly grateful to the numerous folks who gave me feedback on this section. Non-inlined-references are at the bottom of this post.)

Bases in Brood War began to mine more inefficiently when more than one worker was assigned to a single mineral patch. This meant that fifty or sixty workers spread across five or six bases would mine significantly more than fifty or sixty workers assigned to three bases. This is the basic foundation of an “economic” vs. “turtle” dynamic – the economic player can take more bases, have a much better economy as a result, and trade cost-inefficiently with a player sitting on fewer bases – hence the idea of a Protoss player being “one base ahead” in PvT when the Terran plays mech.

StarCraft II changed this dynamic by allowing perfect mining efficiency at two workers per mineral patch rather than one. Because there’s little upside in terms of supply efficiency of going beyond 70 workers, StarCraft II effectively has a “soft-cap” of a three-base economy. This means that a turtling player sitting on three bases will have a similar economy to an “economic” player sitting on more bases.

The design here stems from a desire to reduce the burden associated with economic management, and it’s very intentional – notice the goal of avoiding annoying players in this pre-release interview with Dustin Browder, the lead designer of Wings of Liberty:

Q: That leads us back to the balance between Micro and Macro. Since Blizzcon 2008, you have changed the economy system again. Back then there were already two vespene geysers in each base, but they would shut down for a short time after having collected a certain amount of gas. Therefore you would have to check the status of your source of income frequently, forcing a large amount of Micromanagement. Why this change?

A: Oh dear, we are thinking about how to modify the geysers since forever. We want you to have to manage your economy more. And the geysers would be a perfect start point, since they were quite unspectacular in the past: You sent three workers there, and that’s it. So we decided to change the mechanic, which hasn’t succeeded thus far. It was extremely hard to balance the new system. Had we decided to regulate the gas supply necessarily by hand, to collect the regular amount of resources, we would have severerly disadvantaged the newer players, since they couldn’t afford expensive units like Battle Cruisers and Templar. But just these units have the most appeal to casual players. Therefore, we would have to modify the mechanic in that way, that you still earn enough gas if you leave the geysers to themselves. But then, Micro experts would collect by far more resources and would produce only very mighty units like Carriers and Archons. That would also be unfair. In addition, the constant geyser-checking would become annoying very quickly. We want to reward the players, not annoy them.

(emphasis mine)

This line of thinking should sound familiar – it’s the same approach Age of Empires took to re-designing its farming mechanics. The removal of collection points followed a similar pattern in Age of Empires III.

The effects of the three-base soft-cap are compounded by the way bases are placed on the map. For a multitude of reasons, many stemming from Protoss’s relative inability to effectively defend three bases in the mid-game when their third base is too far away, bases in StarCraft II are much closer to one another than in Brood War. This is another advantage for the turtling playstyle – not only is their three base economy just as good as their opponent’s hypothetical N-base economy, it’s also fairly straightforward to achieve.

The result is that enormous emphasis is placed on cost-efficiency. Trading inefficiently is strongly discouraged because there is no effective way to counter-balance this with a substantially stronger economy. This lends a natural advantage to turtling playstyles that emphasize cost-efficient unit compositions. It also means the designer needs to put extra effort into balancing the game; if it’s reasonably straight-forward to turtle to the game’s “best” composition, then the “best” composition needs to be regularly examined and potentially nerfed. The ability for players to improve their way out of this situation by playing an economic playstyle is substantially hampered.

StarCraft II thus heavily incentivizes players to either rush or turtle. It manages to partly avoid this polarization by enabling a number of powerful timings in both the early and mid-game, allowing macro players to play something in-between the two styles. However, I want to emphasize here that what’s considered “macro” play in StarCraft II is not dissimilar from the “turtling” play we saw in classic real time strategy games – players are rarely, if ever, incentivized to trade inefficiently, meaning they adopt the same fundamental get-the-best-composition mindset of a turtling player. The inclusion of timings just acts to provide exit ramps that enable players to actually attack each other before maxing out their supply.

Before we go further, one point worth raising here is the concept of oversaturation. Brood War allowed players to oversaturate bases with workers to a far larger degree than StarCraft II, which severely punishes mining efficiency after the 17th worker. Furthermore, the large distance between bases meant that additional expansions were difficult to defend and hold. This incentivized players to stay on fewer bases than the economy fundamentals might otherwise suggest, encouraging players to expand ahead of their opponent only when facing a turtling style.

This is likely one of the key reasons that in Brood War we don’t see the same sprawling economic vs. economic styles that we saw in Age of Empires II – expanding in that game meant building town centers, which had built-in worker garrison and self-defense capabilities. Expanding was relatively safer and there was no concept of oversaturation on farms (oversaturation on wood was possible to a large extent while limited and eventually hard-capped on gold and stone), design features which heavily incentivized expanding outward. This also points to the fact that the economy is interdependent with other gameplay systems – the immobility of Terran mech disincentivized expanding too quickly, stronger static defense capabilities in Age of Empires II made expanding safer, and so on.

Design Fork-In-The-Road #3: Basic macro- and micromanagement vs. advanced macro- and micromanagement

I touched on this topic in the mechanics section, but I focused mostly on its impact on how games play out. I also want to examine its effect on user experience. Classic real time strategy games, partly by choice and partly by technical limitation, emphasized very basic macro- and micromanagement. The game engines’ capabilities were fairly limited, and players spent most of their time doing basic actions – selecting production facilities, queuing units, reorganizing their economies, moving units, targeting enemy units, etc.

Modern real time strategy games streamlined these mechanics and shifted focus to more advanced macro- and micromanagement with more strategic implications. Age of Mythology emphasized god powers, Age of Empires III emphasized positioning around cannons, and StarCraft II emphasizes ability mechanics (spellcasting, siege / unsiege, etc).

This approach synergizes with the economy changes mentioned previously. Dumb units in classic real time strategy games were remarkably ineffective without handholding, meaning that fights came down to control more than anything else. Losing units was also less painful because it was more acceptable to trade inefficiently. Smarter units in modern games are much more capable of winning fights on their own – a stronger army will generally defeat a weaker army, regardless of control. This places more emphasis on cost-efficiency and changes how players think about taking engagements. Players actively avoid fights they think they can’t win, because they’re less able to simply control their way to victory.

The difference in moment-to-moment actions also changes how difficult a game feels. Players in classic games spent their time doing things that were very easy and basic. Easy tasks feel easy. Players in modern games spend their time on more advanced and harder tasks like spellcasting. Hard tasks feel hard. I think this partly explains the feedback from Korean professionals that StarCraft II is “too difficult”, even though it is less mechanically challenging in-aggregate than Brood War. Advanced actions, particularly binary mechanics like spellcasting, come across as more challenging than basic tasks like queuing units.

The experience of controlling units on a moment-to-moment basis is also particularly different. Classic real time strategy games featured units that were ineffective without manual intervention. Controlling units meant a focus on achieving something. Units in modern games are not only more capable than their predecessors, they’re also easier to control thanks to user interface niceties like smart casting and stronger waypointing. Unit control has a chess-like feel to it – controlling an Oracle is less about achieving anything than it is about executing the right strategic decisions to get it in the right place at the right time to achieve a specific goal. Once that’s accomplished, the unit can do the rest on its own, at least relative to units in classic games.

How this ties together

These three forks-in-the-road share a common trait: classic games emphasize the real time component of real time strategy games while modern games emphasize the strategy component. Designers take different forks in the road because they synergize well with each other. An emphasis on building more units instead of the right units fits well with a sprawling, hard-to-manage economy that allows for inefficient trades, a streamlined user interface meshes well with smarter units that are easier to handle, and so on.

In the case of Age of Empires, this was very intentional – Bruce Shelley specifically cited simplified micromanagement as an achievement in the progression of the series.

This piece does not argue that one approach is objectively better, though it goes without saying that individual players will usually prefer one over the other. Rather, it emphasizes that they are different approaches which result in fundamentally different gameplay experiences.

This piece also does not argue that mechanics and strategy are mutually exclusive. Classic games featured plenty of tactical battle planning, positioning, and strategic development – observe the progression in the meta-game of Brood War over the past two decades. Similarly, modern games don’t benefit from de-emphasizing mechanics. I discussed both of these ideas in my mechanics piece.

StarCraft II’s design decisively treads the strategic fork in the road. It emphasizes the player’s plan and how they adjust that plan rather than their execution of that plan. It limits sprawling economic playstyles, producing more passive games with less action. Its user interface is strong and heavily featured. It requires more input from its designers to continually rebalance the game because of the strength of turtling into cost-efficient unit compositions (and remember – most “macro” play in StarCraft II is just turtling in disguise).

It’s not a successor to Brood War – instead, it’s arguably the best-to-date version of an entirely different type of game, the “strategic” or “modern” competitive real time strategy game. It’s more streamlined, more passive, feels harder, and is more focused on the right decisions – it emphasizes the strategy in real time strategy. Brood War and Age of Empires II are more mechanical, more active, feel easier, and are more focused on optimal execution – they emphasize the real time in real time strategy.

In the future, I’d like to examine the implications of these design approaches on a game’s viability as an electronic sport. We’ve already gone fairly long today, so I’ll save this for another time.

If you enjoyed this post, please consider following me on Twitter or Facebook to receive regular content updates, or checking out my game-design videos on YouTube and Twitch. Thank you for reading and see you next time.

Additional References

• https://www.reddit.com/r/starcraft/comments/1p38fo/why_adjusting_units_will_not_fix_passive_play/ (nice post about brood war vs. starcraft economy)