Time is an interesting thing in games.
At a very basic level, you usually have the power to pause a game. Some games play with time more explicitly, as the recent Prince of Persia games have. Yet others take time manipulation even further, like Braid‘s suite of time-bending mechanics, or maybe just temporal echoes like The Misadventures of P. B. Winterbottom. Like death, time is one of those immutable things that we face in nature that is ripe for fictional and mechanical treatments in games, both in the narrative and in the play. (That could be an article in itself, though, so I’ll save that for another time.)
On a more specific level, I’ve been looking at time and how it functions mechanically as a game design element, working on my own game design again. I’ve been pulling in ideas from a variety of sources. From the differences between “speeds” of cards in Magic the Gathering or the World of Warcraft TCG (instant, sorcery/ability… and then there’s phasing, delay counters and other oddball mechanics) to warmups and cooldowns, from turn-based RPG notions of speed to real-time games and rate of fire or travel speed, there are a lot of ways that time makes a difference in games. Specifically when considering balance, I tend to look at speed as another “handle” to tweak in order to nudge around valuation of varied game design elements. Mark Rosewater has written about similar things when talking card design in MtG, so before I blather too much more, may I recommend an article or two of his?
Mark Rosewater Q & A
Plenty of Time
You Make the Card FAQ
Equipment to Be (especially the Topic #6 subsection)
That last one is likely where I latched onto the idea of “knobs” to tweak when defining costs on game mechanics. I’m pretty sure he’s written about it elsewhere, but durned if I can’t find those articles. Point being, the cost/benefit ratio for each game design element is an important factor of balance.
In MtG terms, two similar cards might be allotted the same level of power but that power might be expressed in different ways. One might be an “instant” card with a cheap effect, while the other is a slower “sorcery” card with a more powerful effect, but both have an equivalent abstract power level. (And yes, something like a 2W 1/4 creature with an ability might have its power balance figured differently against a 2W 3/2 creature with a different ability. MtG design isn’t always about the card’s speed… but speed is what I’m looking at today.)
So what of turn-based tactical RPG design, since that’s what I’ve been looking at in my other articles (this, a game of my own design, so I can speak with some authority about the thought processes behind it)?
Balance, Part 1: Tao of Picasso
Balance, Part 2: Asymmetry and Art
Balance, Part 3: Systems, Defaults and Munchkins
Balance, Part 4: Triangles, Trinity and Triage
I’ve pulled inspiration from a lot of different sources on this, but four major examples stick out: Final Fantasy Tactics, Band of Bugs (the first game I worked on here at Wahoo/NinjaBee), World of Warcraft and Final Fantasy X. Some thoughts, then…
In physics terms, “speed” is distance per unit of time. Many RPG systems incorporate speed as a core characteristic of units in one form or another. Sometimes it’s “agility” which affects a variety of things, sometimes it’s a simple measure of d/t. Band of Bugs expresses unit speed purely as distance (number of squares on the grid a unit can move) per turn. Each game “turn” consists of a single action (with or without a move) by each allied and enemy unit. Once everyone has had a shot at doing something (even if it’s only defending in place), a “turn” is over. A fast unit can move farther each time it gets to move, while a slow unit can only move a few steps each turn.
BoB units alternate activation. A player gets “action slots” within the turn to move their units, and a turn allocates all of its slots more or less randomly between all of the units on the field, friend or foe. Players can choose to move whichever unit they like in their time slots. The first two slots might be reserved for the player, but then the computer or other player might get the next two, then they may alternate single slots, then back to two. Since it’s not “all my team moves, then all your team” like Disgaea, FFT or Valkyrie Profile, Covenant of the Plume, you’re never sure which unit the other guy will choose to move during their time slots, so it’s a little trickier to plan things out. This rewards a more conservative set of tactics. I’ll admit, I’m a bit annoyed that I can’t predict my opponents, and that I can’t gang up or make a front by using my units all together, but it does make for a more chaotic sort of feel, which again, rewards a different sort of play style. It also makes more chronological sense than “all my guys go, then all yours” as if mortal combat were some sort of square dance with weapons. …which might be sort of interesting in the right context, but I digress.
Point being, speed in BoB has more to do with how far a unit moves than anything else, and this is pretty common among TRPGs. It seems to be largely rooted in the “turn” structure, where the balance comes in giving each unit an action per arbitrary large chunk of time we call a “turn”. True, this may mean an enemy squad that outnumbers the player team 2 to 1 will also get twice as many actions per turn, but that’s just something that designers have to keep in mind. (On the other hand, if a player team of four units against a computer team of eight units simply alternated these “time slots” without the overall turn structure of giving each unit one action each turn, the player units would effectively move twice as frequently as the enemy units… also clearly a weird notion of balance. More on frequency below, though…)
Contrast that with Chess, where each player has the same number of units, and they take turns moving a single unit at a time according to that unit’s rules. It might be said that the Queen has extraordinary speed potential, if framed in distance/move terms. We might call her flexibility “agility”, while we’re talking about RPG terms, thereby differentiating her from the Rook and Bishop. All three have the same speed potential (as far as they can see in a clear line), but different agility. Of course, the game doesn’t track all that well with RPG design; imagine how the dynamics of Chess would change if you could move all of the units on your team once before your opponent got to move anything at all. It would be almost unplayable. (Which could lead to discussions about why TRPG teams tend to be small, and turn order advantage…)
Of course, there are the Fire Emblem games that take speed and use it differently. The FE games use speed to determine whether a unit can attack a foe twice in a combat round, and to determine how likely an opponent is to actually hit you. High speed means better dodge rates, then, and enough of a speed disparity between attacker and defender may well lead to one getting hit twice per exchange. Speed doesn’t affect how often a unit moves, then, or how far (the former is determined by the “all my team, all your team” system, the latter is class-based), it affects combat. It’s probably closer to what most games call “agility”, then, though the two-for-one strike bit does feel like a unit is simply much quicker than its opponent. It’s such a strong tactical effect that it seems to me that many FE players prefer speed over raw power.
So, there are several different directions to take the notion of speed. The direction you go with it will depend mightily on what you want the game to play like. I find I like the effect speed has in the FE games, though I like the tactical utility of letting speed affect motion range. In something like Phantom Brave, where you aren’t constrained to a grid, it can be even more useful to have a big range of motion.
Warmups and Cooldowns
Timing on actions can also be an interesting knob to tweak. Final Fantasy Tactics has a quirky magic system (the Faith mechanic is something I’ll revisit later; I’ve used something like it in a tabletop miniatures game I’m designing) that might be parsed easily by a World of Warcraft veteran. WoW character abilities tend to range from instant cast abilities with a wide variety of cooldown timers (time before you can use them again) through a range of what I’m calling “warmup” abilities; those that require a bit of time between activation and effect, whether it’s a “fire and forget” missile sort of thing or a “stand here and mumble for a while and eventually your effect will happen” thing. There’s even a sort of hybrid; the “channeled” technique where the effect happens while standing there mumbling, and it keeps happening for the duration of the mumble period. There’s also the instant-cast, lasting effect sort of spells. There are a lot of knobs to tweak with timing there, both with the casting and the effects.
By comparison, the FFT system is pretty simple. It’s just a warmup system where magic spells take time to cast, but once cast, they have an immediate effect. The only cooldown is waiting for another turn to move your mage and start a new spell (and refilling magic points, I suppose, but combat doesn’t usually last that long). There is a marked lack of what I’m calling “smart” or adaptive targeting in FFT, though. In WoW, even your spells with a warmup time are still cast at the initial target, wherever they may be. (OK, assume your cast isn’t interrupted, anyway.)
In FFT, you choose a target and a spell strength. Stronger spells take more time to warm up. …more time to allow the targeted space to change. If memory serves, you target a place on the ground, and units may move into or out of said target. (I don’t think you can keep targeting a unit… though if you can, you don’t know where it will move, so you may well hit friendly units with splash damage.) Using powerful spells then naturally changes into an exercise in prediction and spatial control. That’s all well and fine from a “depth of tactics” standpoint, or for the “I want the game to hate me” crowd, but the trouble is one of feedback. It’s been a while since I played the game, but I seem to remember that you can’t tell when the opponent units will move. You know how long your spell should take to cast, but not exactly who will move between casting and the effect. It’s hard to pin down what might happen to that targeted space. That makes it hard to use bigger spells. If you knew who would move and when, you could predict where they might move thanks to blocking and baiting, but absent that knowledge, magic is little better than slow artillery. Carpet bombing locations has some tactical value, but it’s an indelicate, imprecise weapon.
It seems to me that the missing piece of “when” is crucial to making that sort of warmup-based location-targeted magic useful in a turn-based system. In WoW, where everything is real-time, that question largely goes away, since most players can just move anytime (mobility controlling spells aside, of course). Later FFT games (FFTAdvance and FFTA2) have moved to an “instant cast” system that make tactics simpler and easier, for better or worse. (I know, old school is the best school, and all that.) It’s a change I like overall, but I think the predictive potential of delayed warmup magic could be better realized with better feedback on who would move when.
Back to cooldowns, though, WoW tends to use cooldowns to throttle the use of certain abilities. More powerful abilities tend to get long cooldowns. Utility and common abilities get short cooldowns so they can be used more often. Players get the benefit of being able to use something right now if it needs to be used, but they have to wait to use it again. It’s a nice balance between usability and throttling, permitting powerful abilities without them overwhelming the more common ones.
Similarly, the Kingdom Hearts series has fiddled with its magic in each iteration of the IP. Early titles used instant-cast magic point-consuming spells (MP being yet another knob to tweak for balancing things, of course). MP could be refilled by collecting drops from fallen foes, recharging at save points or by using items. Later iterations of the game have dropped MP altogether and have used cooldowns to throttle magic use. It’s an interesting change that sort of smooths out the pacing of magic use. Instead of using a bunch in a flurry then refilling MP, players can use magic spells more regularly over time, but without the temporally concentrated burst of activity they once were capable of. The spells refill automatically, so resource management moves from keeping track of MP and items, and more into keeping track of time.
Time as a Cost
Or, “where it all starts to come together”.
I really like using time as a cost for abilities, ideally with a mix of warmups and cooldowns. To really make things work, I want enemy turn order to be crystal clear, so that prediction can work. When time becomes the coin of the realm for using abilities, as a matter of feedback, it needs to be clear how things change during that time.
That’s where I’m borrowing from Final Fantasy X. It’s one of my favorites in the FF series, partially because of the combat system. It’s a turn-based sort of system in that it’s not “real-time, always running”. The game waits for your input, and your characters and the enemies take turns beating on each other. It’s a different sort of “turn” from the games I’ve written about above, though.
There’s a timeline that all unit actions fall on. Units act when they are ready, with the frequency of their actions depending on their speed rating. Faster units get to act more frequently. Time itself ticks on underneath the turn order. There’s no “each of us gets one hit on each other” ad perpetuum, it’s just characters checking against the underlying clock to see if they get to act again. (I’m not sure what the system does in the case of a tie, but I’m not sure it matters to the user since they know ahead of time and can plan accordingly.)
As for feedback, you always know who takes the next turn and the next handful of turns after that because there is a UI element that shows you the next several actors (not their actions, just who is up to bat next for the next ten actions or so). Further, those actions can be modified by things like Haste spells (higher frequency of actions) or Quick Attacks (low power, shortest cooldown), and the turn order shifts around to reflect that. You can also swap characters in combat and the time layout may shift to reflect changes in the order because of the new character’s speed. You can predict the next few moves very nicely this way and react with much more control.
It’s that sense of “time goes on” that I like, rather than the Red Rover sort of game where teams alternate attacks. Yes, you will always get some of that “wait ’til I hit you” time stutter-step with a non-real-time combat engine, but I think that the FFX system hits a nice compromise.
This time treatment is actually fairly similar to the World of Warcraft Miniatures game, which I like quite a bit. It’s not something that functions to the depth of a WarHammer or WarMachine tabletop game (go check out 6 Inch Move for some great articles on tabletop gaming, by the way), mostly as a matter of scale, but the WoWMinis game is pretty solid for a simplified tactical game. That game also has a ticking time line that progresses all the time (though players track it, not the computer). Each unit has a personal clock, and they get to act when their clock matches the master clock. Moving (or doing nothing, actually) costs a tick, and unit attacks or abilities each have a “tick cost”, effectively a cooldown. Their attacks are balanced against the time cost. There are no magic points, just time costs. (OK, if you use the optional ability bar cards, those effectively have a “ten tick round” potential cooldown, as they can only be used once per ten ticks of the master clock, so they function a little differently, but still, that’s a time-based throttle.)
I played a game with my wife where her Orc Warrior was able to use an ability to get some extra range on one of his turn’s moves and then immediately take another turn simply because everyone else was on cooldown. He subsequently squashed my fragile mage, winning the game for my wife. That sort of tactical situation (effectively a last-ditch two-for-one double turn) doesn’t come up in a game where you just alternate actions. Yes, I lost, but it was awesome that the game allowed that sort of thing.
So, for my game, I want to have a FFX-like time system, where units act according to their own timing measured against the master clock. There will be a display of who will be acting for the next handful of “turns”, allowing for predictive tactics. I love the idea of both warmup and cooldown abilities, and I want to make a system that makes them interesting and fun. I will probably include a UI element that shows when a warmup ability will actually trigger within the master clock’s queue. Unit speed will dictate the frequency of their actions (and other subtle effects we might usually attribute to agility), and actions will be paid for with their time cost.
I want the gameplay focus of the game to be on smart tactics based on solid intel. There’s a place for guesswork, but that will be optional, as players can turn off the timeline UI or introduce fog of war. At its heart, I want my game to err on the side of ease of use (not necessarily ease of conquering the game’s challenges, mind you) so that the challenge comes from the game’s tactical situations, not fighting the system’s limitations.
One More Thing
A few final quirks of time in games: Talking is a free action; time moves at the speed of Plot, or as fast as players need it to.
Read Full Post »