Generals 2

By now you've probably figured out that on Generals2.org, we love Generals & Zero Hour. In fact we love them so much that the prospect of their sequel is as frightening as it is exciting. While Generals/ZH are built on some excellent gameplay mechanics, it doesn't mean that even a similarly designed Generals 2 is guaranteed to deliver the same addicting experience of the originals. Below, in no particular order, is a list of not-so-obvious ways EA could blow its opportunity with Generals 2.

1. Smart-casting

What is smart-casting? Smart-casting is when the game intelligently augments a player's ability to activate and use abilities. For example, in CCG/ZH, if you select several Missile Defenders, and then use their Laser Lock ability against a target, all of the Missile Defenders in the group will fire at that one target. If you want to Laser Lock multiple targets, you have to split that group of Missile Defenders into smaller groups (i.e. put them into pairs). Then you can select each pair or sub-group, and Laser Lock a different target with it.

If Laser Lock was a smart-cast ability, it would automatically prevent all Missile Defenders from Laser Locking at once, and instead engage the ability only one Missile Defender at a time. Thus you could simply have all Missile Defenders in one selection group, hit the 'L' hotkey once to engage Laser Lock mode for all Missile Defenders in that group, and then each time you click would "cast" exactly one Laser Lock onto a target of your choice. Thus if you have three enemy targets, and six Missile Defenders, you would have them all selected, hit 'L' once, and then click twice on the first target, then click twice on the second target, and twice on the third target. This will distribute Laser Lock evenly across all three targets. Without smart-casting, you would end up applying all six Laser Locks to ONE target (which is obviously overkill, and inefficient).

If you're not accustomed to micromanagement and control in an RTS, you might think smart-casting sounds like a great idea. You don't have to worry about splitting up your group of six Missile Defenders into three separate pairs. Smart-casting would make it easier to Laser Lock multiple targets (though it would make it harder to Laser Lock a single target with multiple Missile Defenders - a slight trade-off), without overkilling a single target. Of course, that's the whole problem with it - it makes it easier, meaning anyone can do it.

One of the defining features of CCG/ZH gameplay is the focus on combat and unit control. Unlike StarCraft 2 which requires players to spend a lot of time managing their base, economy, and production, CCG/ZH instead emphasizes combat and micromanagement. Thus in order for the game to offer sufficient skill contrast between players, it has to be delivered through micromanagement and unit control. If CCG/ZH were to take steps to assist players with demanding micromanagement tasks, it would remove key skill differentiators from the most important part of its gameplay. Currently, it's impressive when you see a player Laser Lock multiple targets with multiple groups of Missile Defenders because you know that it's not easy to do. It ceases to be amazing when anyone can do it with minimal effort. Here is a skill pyramid to illustrate the effect the lack of smart-casting has on Laser Locking.



Now let's look at what happens when you introduce smart-casting:



Assuming Generals 2 has a similar emphasis on combat and micromanagement, the addition of smart-casting will make it too easy for too many players to play with optimum efficiency, lowering the overall skill ceiling and making the game too easy to master.

2. Pay-as-you-go Resource Spending

The above example is related to micromanagement of units, but there are opportunities for Bioware to spoil the game on the macromanagement side of things as well. One way in particular would be adopting the pay-as-you-go resource spending model of Command and Conquer 3/Red Alert 3, instead of the pay-up-front spending model of Generals. What is the difference between those two spending models you ask? Well that depends on who is giving the answer. Some people claim that pay-as-you-go is a purely a user interface tool designed to make the game easier to manage. Others will say that pay-as-you-go defeats its own purpose by making the availability of resources ambiguous and unintuitive, resulting in unpredictable build times of units and buildings.

With a pay-up-front system, you cannot purchase a unit until you have its total cost available to you. If a unit costs $1000, you cannot start to build it until you have $1000 available. In a pay-as-you-go system, you can start building the unit at any point, and then it simply extracts the $1000 from your resource pool over the unit's build time. At first glance this sounds like it will be better for the game, but it suffers from three major problems.

The first problem, as mentioned earlier, shifts economy management from being a simple binary "either-I-have-it-or-I-dont" glance at your income, to a far less intuitive and more convoluted system of managing income rates vs production rates against your income reserve. For example, if a unit costs $2000 and takes 30s seconds to build, it puts a resource drain of $66.66/s on your income, which translates to $4,000/minute. You then have to know what your income rate is, which may vary slightly depending on how the economy is designed. In CCG/ZH for instance, the planned income rate per supply depot was $3000, but it could vary by a few hundred dollars depending on how you placed your resource collection. Sometimes you deliberately built a less than optimal resource setup just so you could build something else faster, or perhaps you were slowly building up your resources as in the case with GLA, which had an optimum of about 5-6 workers. So using a pay-as-you-go model, if you wanted your $2000 unit to build in the 30s it was supposed to build in, you would have to know that it creates a $4000/minute drain on your income, you'd have to know what your income rate actually is, you'd have to know what else might be draining your income and by how much, and then weigh all of that against how much cash you have in your reserve to know if you'll have enough to complete the unit in time. Even when a game gives you a bunch of UI tools to help you see your income rate vs your expenditure rate like Supreme Commander does, it's still not as simple as comparing $1900 available with the $2000 necessary to build the unit, and knowing instantly that you can't afford to buy it yet.

So why is it such a problem that rate flows can affect the reliability of the build time of the unit? The reason is because in competitive play, EXACTING predictability of build times is vital to advanced timing strategies and tactics. If units and buildings take an exact amount of time to build, you can plan strategies around that. Moreover, if a unit is in your base, and you are in the process of building a unit to counter it, sometimes a split second makes the difference between a win or loss. If a unit stops building at 99% you could lose the game. To the player trying to get that unit built, it makes the game ENORMOUSLY frustrating (far more frustrating than waiting for money to come in!). Simply put, competitive play is better when a unit which has a 30s build time, actually builds in 30s (loss of power not withstanding).

The second major problem, is that in order to make income rate management more intuitive, you need to keep income rate drains for units consistent. Gavin Simon was one of the developers of Command and Conquer 3, and he was adamant about making all vehicles/buildings create a drain of $100/s, and all infantry a drain of $50/s. That way if you knew what income rate you were getting from your harvesters, you knew how many production queues you could sustain. The problem? Locking units in to exact $100/s or $50/s cost/build time ratios tied the balance designer's hands. Sometimes it made sense for units/upgrades/buildings to have low costs, but long build times. Being locked into $100/s or $50/s limited the ability to fine tune production balance.

The third major problem, and one I'm sure many will disagree with, is that pay-as-you-go eliminates a skill differentiator. With a pay-up-front system, you cannot queue up more units than your cash reserves allow. If you have $3000, you can't queue up more than three $1000 units. What this means is that if you're busy micromanaging or intensely focusing on some task, you will start accumulating cash (known as "cash floating"). Cash not spent is time wasted, thus your goal is to keep your cash reserves as close to $0 as possible so that you're maximizing the efficiency of your income (unless you're saving up for something). Doing this is an important skill in the game - one which separates bad players from good players, and good players from great players. With a pay-up-front system, keeping your cash reserves near zero is harder because you are required to queue up units more frequently. In a pay-as-you-go system, you can queue up dozens of units so that you almost never have to think about building them. Sounds good right? The problem is that it makes efficient production management too easy, the way smart-casting makes micromanagement too easy. If a player has the ability to both micromanage AND produce units, they will have an advantage over someone who can't multitask well enough to do both.

Suppose player A and player B have equally impressive micro during a battle, but only player A is building reinforcements while he is fighting player B. By the time both players are done with the battle, player A will have reinforcements on the way while player B has only just started rebuilding them. Player A can then keep up the pressure while player B starts to fall behind. If you were to switch to a pay-as-you-go system, player B doesn't need to be good at multitasking, since the game essentially automates unit production for him. So despite player A having superior multitasking skills, he is actually handicapped by the fact that the game is helping his opponent play better than his opponent can otherwise actually play.

3. Going Through The Motions

There is a difference between bolting a gameplay mechanic onto a game for the sake of having it, and actually integrating it into the gameplay. For the sake of argument, let's say BioWare Victory wants to use a worker/builder mechanic since that's what was used in CCG/ZH. Simply slapping a worker/builder mechanic onto the game without really thinking about ways in which it can affect the gameplay won't accomplish much. We saw this with Battle For Middle Earth 2. In BFME2, EA listened to the fans' complaints about the over-simplification of build plots, and decided to implement a worker/builder mechanic for constructing buildings. In reality though, it offered very little depth over the build plots. Sure the workers could be intercepted and harassed by the enemy, adding a small layer of harassment that BFME did not have, but in practice there was almost no interaction or depth afforded by the implementation of the worker/builder. The worker just disappeared into the building, and workers couldn't harass or interact with each other. It was about as basic as it got.

In CCG/ZH, workers/builders played a vital role beyond their ability to simply construct things. A dozer was a weapon that could run over infantry, which in a pinch, could nullify the effectiveness of sloppy Chinook drops, or crush a GLA worker building a tunnel. Workers could occupy garrisons to keep themselves alive, build demo traps near other GLA workers to stop them from building something etc. Dozers also had high target priorities, meaning you could use them to distract Gattling Tanks and Quad Cannons to buy your Missile Defenders or Tank Hunters some survivability. Dozers could also physically block incoming units, giving you vital time to move some counters into position to intercept. Dozers could also be sniped or neutron mined, and then captured, allowing for some limited tech stealing in the game.

The point of the above is that putting workers/builders into the game simply for the sake of having them, won't result in any interesting gameplay unless time is taken to think about the interactions and relationships those builders have with the rest of the game. The same can be said for nearly everything: the economy, units, upgrades etc. Putting a checklist of features into a game without thinking about each feature's relationship to the gameplay won't give the game any sort of long term depth. Bioware Victory needs to do more than go through the motions.

4. Prioritizing DLC over Competitive Multiplayer

In the context of an RTS game, or any game where players compete against one another, the concept DLC is at odds with the concept of fair play. EA wants to make money off Generals 2, especially in the long run. In order for them to do that they have to make the purchase DLC worthwhile, which ultimately means competitive multiplayer is not in their best interest. Aside from fairly unimportant things such as custom unit skins, there's no place for DLC in ladder play. You can't give some players access to more powerful units, nor can you give them access to new factions (which in itself causes problems because there is a conflict of interest between making DLC factions balanced, and making them enticing to purchase (e.g. an advantage)). In the end this means that competitive PvP multiplayer is hard to monetize, and it is likely that EA will dedicate more resources to polishing the DLC ecosystem instead.

How could prioritizing the DLC features affect competitive multiplayer, and how could that negatively affect the game? A good competitive multiplayer experience is fairly time consuming to get right. There are a lot of details which must be paid attention to, and often simple features require a lot of engineering bandwidth to implement. For instance, a ladder system should be smart enough to avoid matching the same two players together over and over again if one player quits early (e.g. doesn't want to play his opponent) or if both players have connection problems. In Zero Hour, this was a chronic problem - two players would often get matched together multiple times in a row even if they couldn't connect to each other.

The above example is just one of dozens upon dozens of "little things" the game needs to get right in order for the competitive multiplayer experience to feel polished. If a developer prioritizes DLC over competitive multiplayer, then many of those "little things" won't find the engineering bandwidth they need to make it into the game. A weak competitive multiplayer experience will hurt the long-term size and sustainability of the community substantially, which ironically hurts the potential market for DLC.

StarCraft and DoTA have shown that competitively focused communities last longer and stay healthier than communities with no such structure. Competitive play gives players reason and incentive to keep playing the game, and regular events/tournaments/leagues provides the same kind of structured activities that any good community building plan would aim for. While it's tempting to think that putting a bunch of DLC content out there and letting the community run with it will keep the game healthy and active, it will do the opposite depending on how that DLC works. If the DLC makes some players incompatible with others, or provides too many gameplay modes, it will fragment the community. Moreover, DLC still doesn't address the problem of lack of structure. Structure is the glue which keeps a community together and active, and nothing provides better structure than the competitive scene.

StarCraft 2 sold as well as it did because the StarCraft competitive scene is not only healthy and active, it's the competitive scene. The StarCraft community was already primed for SC2 thanks to the longevity of its community. StarCraft 2 also sold as well as it did because Blizzard put a lot of time and effort into making the game acceptable as an eSport, which meant lots of third party sponsors invested money into the scene, spreading brand awareness for the game. The scene is so active and popular that its competitive esports news makes it to the front pages of Digg and Reddit now and again, which is something that does not happen with Command and Conquer games.

Since the Blizzard side of RTS has an absolutely enormous amount of players, and those players are "aligned" with competitive play, sacrificing a focus on competitive play in Generals 2 will ruin a chance to tap into a much larger pool of potential customers, which again, ruins the chance of some DLC purchases down the line. Someone who doesn't buy the game because it's not polished enough for competitive play, by definition, cannot buy DLC.

5. Big armies / too many resources

Unequivocally, beyond any shadow of a doubt, CCG/ZH's defining characteristic was the small average army size. Much of what made those games so skillful and tactically diverse depended on the game's restricted economy and subsequently small armies. In small army games, every unit has a high implicit value, meaning keeping it alive and maximizing its cost effectiveness is a key priority to winning the game. It is this very priority that pushed players in CCG/ZH to develop and utilize the advanced micromanagement techniques we see in those games today.

The actual design and capabilities of the units in CCG/ZH were part of the picture, but not the whole picture. The fact that the limited economy made it harder to replace units had two important effects on the game. The first is that you weren't punished too severely for focusing on micromanagement, and neglecting production. In larger army games, the benefits of better unit control and micromanagement are often counteracted by slipping behind in your production, thus there's almost no point to micromanagement (and by that I mean truly attention occupying micromanagement that you see in ZH, not casual "mass micro" you see in other games). Of course this problem can be "solved" by implementing pay-as-you-go, but in reality that makes the game even worse. For example in StarCraft 2, kite micro with Reapers or Hellions early on in the game is intensely demanding if you want to truly maximize the damage potential (e.g. turning around and firing every ~1000ms that those units take to reload). But in StarCraft 2, you will fall severely behind in your production and when you do eventually lose those units or retreat them, you won't have given yourself an advantage at all. You will be behind in your economy and/or your army. So while you may have gained a cost efficiency advantage over your opponent, you've lost time efficiency.

The second effect is that because units cannot be easily replaced, micromanagement buys you a significant time and resource advantage over your opponent. Compare CCG/ZH with C&C 3. In CCG/ZH, a normal economy could barely sustain production from two war factories. If you are losing units due to poor micro, it's not as if you can replace them from four, five, or even six war factories like you can in C&C 3. The economy itself restricts the speed with which you can replace those units, regardless of how many production buildings you have. And because you can't sustain production from more than two factories, it's rare that you ever have more than two anyway. So even if you're floating enough cash to build five or six units simultaneously, you don't have the production capacity to do that. So as a byproduct of the weak economy, your ability to replace units was not only limited by your income rate, but by your production capacity even if you were floating cash.

Doing some math, we know that C&C 3 has about three times the economic scale of CCG/ZH (average midgame income on a "normal" economy in CCG/ZH is $6,000/min, while in C&C 3 it's between $18,000 - $20,000). However, unit costs and build times between the two games are similar, meaning in C&C 3 you can replace units three times faster with comparable production capacity. Or another way to look at it, one unit in Zero Hour is equivalent to three units in C&C 3. This means that replacing a tank in Zero Hour with a build time of 10s and a cost of $1000 should be equivalent to building a $3000 tank with a 30 second build time in C&C 3. However, because you can sustain triple the production capacity, you're able to build that $3000 "tank" in just 10 seconds. Imagine in CCG/ZH if tanks built in 3.33 seconds? Better micromanagement wouldn't net you much of a time advantage because the tank you micromanaged to kill, will be replaced in just 3.33 seconds. And because in C&C 3, that single tank is actually three tanks, it means you can partially replace that $3000 loss in even less time. In CCG/ZH, it would be like being able to replace 2/3rds of the tank in just 2.22 seconds, or 1/3rd of the tank in 1.11 seconds.

In order for C&C 3 to make micromanagement as rewarding as in Zero Hour, the maximum micromanagement effectiveness of units in C&C 3 needs to result in cost-effectiveness ratios THREE times higher than in Zero Hour, which already has cost-effectiveness ratios between 5:1 and 10:1 (significantly higher than C&C 3). When it comes to RTS game design, ratios that high are pushing the limit of what feels balanced as it is, so to go three times higher than that would just feel broken (e.g. units would feel uncounterable, as quite a few already do in ZH). However if you don't go higher than that in a big army game, then micromanagement does not feel as rewarding, and players cannot build the same kind of advantage through micromanagement that they can in small army games.

If Bioware Victory were to go in the opposite direction of Command and Conquer 4, and follow through on their marketing line of "massive armies", then they will turn Generals 2 into a slugging match instead of something that requires the awe-inspiring finesse and precision of martial arts.

Overzealous bug fixing

Some of the best gameplay in any RTS game comes from bugs. While some bugs severely upset game balance and/or break the intended flow of a game, other bugs enhance the depth of the game by providing new opportunities for players to use units creatively. Take for example mutalisks in the original StarCraft. Ordinarily, mutalisks would spread out so that they are not all bunched together. However when their default behavior spreads them, they are unwieldy and cannot be moved precisely (i.e. you cannot engage all mutalisks in combat without unnecessarily over-risking many of them). At some point during the game's life cycle, players discovered that they can group a single larva in with a bunch of mutalisks, and due to the game's path finding, those mutalisks would all bunch together and literally "stack", occupying the same space. This meant you could control 20 mutalisks as if that group of 20 was just a single mutalisk. This "bug" spawned a staple of Zerg micromanagement, in which proper mutalisk control and stacking became an important tool in any Zerg player's arsenal. Without this bug, mutalisks would be far less effective at harassment, and likely would have needed some kind of buff to compensate. Moreover, the actual handling and control of the mutalisks would not be as impressive or interesting to watch since there would be no point to dancing them around the way players do when they are stacked.

What's more is this behavior had a natural counter-balance to it: area effect weapons became significantly more lethal against tightly stacked groups of mutalisks. Casting irradiate on a mutalisk in a stack meant all of the other mutalisks would take damage until that mutalisk was separated from the group and flown away. Without mutalisk stacking, Irradiate would not be as effective, and it would be trivial for a Zerg player to remove the affected mutalisk from the group.

Zero Hour has its own behavioral bugs which enhance gameplay depth without affecting balance. Technicals, for example, ordinarily have momentum when they stop or start. So if you run through a line of infantry and give the technical an order to turn around and run over another infantry unit, it first has to slow down, turn around, and then accelerate again. This takes time during which the technical is incredibly vulnerable. But there's a trick to overcome this momentum. If your technical is driving at full speed, and if you click precisely to the side of it, it will stop dead and turn 90 degrees in almost an instant. So if you run over an infantry unit, you can quickly click to the left or right of the technical to spin it around and then easily run over another infantry unit.

Unfortunately, many players and many developers have a rigid mindset that anything which was not explicitly planned by the designers is a bug, and must be removed from the game. Bugs aid in the emergent behavior of a game, allowing the players of that game to discover new ways to play and interact with it. The proper way to handle a bug is to first study its effects on the game before deciding whether it breaks the game or causes a balance issue. But even if it does cause a balance issue, if the bug adds depth or skillful nuanced behavior, then balance should be adjusted around the bug to compensate.

A bug should only be removed if:

A. It creates a shortcut that makes it easier for a player to achieve the same results from the otherwise planned behavior of an element. In otherwords, it acts like a skill crutch.
B. It catastrophically upsets balance or breaks the game (tunnel bug, scud bug).
C. It crashes the game.

Hopefully the developers at Bioware Victory are tolerant of bugs, and are willing to see if they make the game better before outright dismissing them as problems. Serendipity is a designer's best friend.

Summary

There are of course a lot of obvious ways that Bioware can ruin Generals 2. It can leave out a ladder or competitive player-vs-player mode entirely, it can integrate DLC and persistence into the ladder, and it can fail to balance/patch the game. However, this list was meant to explore more subtle ways the game could be ruined. I also wanted to avoid simply repeating the items from the Fingerprint of Generals article as much as possible, but if I were to extend this list any, I would also include the implementation of MCVs instead of workers as an excellent way to kill the game's potential.

Generals is a game whose relative success can be attributed to its competitive-oriented design. If Bioware fails to truly analyze what is required to make Generals 2 a competitive game, then Generals 2 will be as easily forgotten as the many other strategy titles which attempted to simplify the gaming experience in the interest of accessibility.

Great article, I agree with 99.9% of that and hope that Bioware takes a close look at the nuances of what made games like ZH and SC awesome competitive games. About the bug-fixing point though, I do think immersion should be given at least some consideration. In some cases where a serendipitous bug caused a unit to behave in a manner different than what the designers explicitly intended, I agree that it could be left alone if it isn't causing technical or design problems.

Since Generals takes place in the "real world", though, I think bugs that defy physics or logic should still be fixed, even if they make for more interesting gameplay. In the example you gave of a technical being able to instantly stop dead in its tracks, that'd look bizarre and immersion-breaking to a casual player who accidentally triggered the bug. The designers could still take note of why that particular bug's effects were fun though, and try to intentionally work that into the game in the future.

I'm more just saying that the overall 'health' of the game has to be taken into account in addition to the competitive e-sports aspect of it. It's true that e-sports are what form the core of the playerbase, and what give a game longevity. At the same time though, many many people buy RTSes and only play the campaign or casually LAN with their buddies. So a game's design can't necessarily be boiled down purely to its mechanics. You could have a brilliantly designed and perfectly balanced RTS made up of polygons fighting each other, and that wouldn't be much fun to most people. Setting and immersion still matter a lot, and any bugs that detract from that should be fixed in an aggressive and timely way.

white titles on white background hurts eyes, good article though. reminds me of cracked.

Very well written AGM. I agree a lot with the pay up front system. DOW2 has it and it really made it hard to just rush to t3 and get a super unit out because it took a long time to gather the resources to purchase one. Purchasing systems like this tend to eliminate winning by luck of quickly getting t3 units out, its a huge risk to save and wait to tier up while your opponent has a superior army.

wtf? don't tell a company how to make their game, let the game developers do their job

btw RESOURCES, don't fuck up like they did with red alert 3, lets wall off the resource node, every single time just make the resource gatherers more bulky -.-;; don't make us manually wall off the thing every single time

kanes wrath / sc2 are perfect with economy, well kw was a little skeptical but both games let you be greedy, all in, or go the middle of the path to playing standard, both games allow you to scout(easily and efficiently)

except in sc2, units die waaaay too fast, kanes wrath the pace was great players could judge exactly what a unit could do and couldn't do

You talk as if developers are gods or something. Chances are, the people who actually play RTS games know more about good RTS game design than most developers do.

If we just let developers make games without giving them feedback, we'd wind up with eco/army spam games like C&C3/KW that have no real micro, and you bludgeon your opponent to death by having a bigger economy/army than he does. Boring.

The laziness in me kinda wants smart casting so I can laser lock like the best ZH elites but I must be strong and resist