Up On The Cross

Have I mentioned I’m working on an indie RPG? Yes, I have.

One of these days this site is going to morph into a more traditional dev blog where I talk about the mechanics of my own project. That transition begins now with a focus on the games that are heavily influencing the design of mine, which means first and foremost Chrono Cross. I’m going to be taking an extra-close look at that game’s systems in the posts to come (which I intend to write more frequently (ahahaha)), starting today with an overview of the high-level decisions CC’s designers made in the service of streamlining the RPG genre back in 2000, and what I’m taking away from them.

No experience

To a genre veteran this is the most glaring difference: the concept of experience points is totally absent. Rather than gaining a level after meeting some numerical threshold of XP, Chrono Cross characters only level up after defeating a boss. After each new growth level is earned, the next few battles will earn the party small stat increases, but soon there is no benefit to grinding other than the items dropped by enemies. This choice allows the designers to know the exact power level of the party at every point of the game, so if a boss seems particularly difficult, we can assume that it is intentionally so.

This amounts to a rejection of one of the genre’s fundamental assumptions, that the player can overcome any obstacle by investing more time into the game. Grinding – fighting “unnecessary” battles for the sole purpose of gaining levels – allows players to adjust the game’s difficulty to compensate for differences in skill. Individual players’ preferences differ as to how much grinding the ideal RPG requires, and CC is clearly catering to the subset whose preference is “none.” There is an enormous difference, however, between a game that doesn’t require grinding and a game that doesn’t allow grinding.

Am I preserving this idea: No. CC is an overwhelmingly story-driven game and I’m making something more in the vein of a dungeon-crawler. I’ll be carefully calibrating how much combat is “necessary” for the critical path, but there has to be a reward for pursuing further battles.

No random battles

These days random encounters are more the exception than the rule unless the game you’re playing is by Atlus, but avoiding them was fairly special when Chrono Cross was released. The closest thing to a battle that you don’t initiate is an enemy dropping on you from the ceiling as you walk down a corridor.

Am I preserving this idea: Yes. I am not Atlus.

Left side: Characters in Chrono Cross briefly earn small stat bonuses from normal encounters, but overall growth is strictly gated by boss battles. Right side: "free" healing with the battle resources you didn't use is one of the most player-friendly features of the game.

Left side: Characters in Chrono Cross briefly earn small stat bonuses from normal encounters, but overall growth is strictly gated by boss battles. Right side: “free” healing with the battle resources you didn’t use is one of the most player-friendly features of the game.

Guaranteed escape

You can escape every single battle in Chrono Cross, guaranteed, with no penalty other than whatever health was lost in the course of battle. You can’t necessarily run off to a different area – boss fights usually begin again immediately after allowing you time to heal and reorganize your party – but even the final boss can be aborted and restarted as many times as desired.

Few other RPGs are this player-friendly when it comes to fleeing combat – in fact a fair number of games don’t give you the option at all – but that decision makes complete sense in light of the importance of tailoring your party composition and Element (spell) loadout to each new area and boss encounter. Having mostly blue Elements equipped and discovering that the boss you just started fighting is blue should prompt you to immediately back out and retool with more red Elements, and maybe a red party member for better overall damage.

Guaranteed escape is essentially of a piece with the decision to remove experience points: if you get into a battle that’s over your head, your only recourse is to change your strategy, not to become more powerful by grinding. Without the ability to freely retry battles with a new party configuration, the game’s strict adherence to a single difficulty curve would be much too punishing to less skilled players.

Am I preserving this idea: Yes. Not just for the reasons that CC allows it, which will all (hopefully) be reflected in my game, but also because the world will probably be more open to exploration without going through story bottlenecks, which creates the possibility of being surprised by encounters far above your level.

Automated healing after battles

If it would have been possible for you to spend a few extra turns casting healing magic before delivering the final blow of a battle, Chrono Cross gives you the option to do it after the fact. Because Elements are only usable in combat, this allows for the most efficient healing possible, reducing your dependence on consumable items to survive in a dungeon. There’s no downside to this player-friendly affordance, as it’s simply automating a process that would be entirely within the rules to do manually, but also entirely tedious.

CC in general is not interested in attrition over time, which is a mainstay of RPGs. Whether it’s called MP or SP, most games have some resource which will run out if you go too long without returning to a safe zone, putting a leash on your range of exploration that can be extended over time through leveling up. As with the absence of grinding, CC doesn’t intend for you to repeat content, so you’re likely to make it through each dungeon without having to back out and resupply.

Am I preserving this idea: Maybe. Probably? Yeah probably. There’s other ways of using “magic” outside of combat, so it’s not quite as vital, but I may end up carrying this feature over just in order to disincentive the tedious alternative behavior. Remember: If the most efficient way to play your game isn’t fun, your game isn’t fun.

Managing 40+ Element slots on each character (of which there are dozens) adds enormous friction to a game that is apparently trying to remove as much friction as possible from the genre. Image via gamersyndrome.com

Eventually, managing 40+ Element slots on each character (of which there are dozens) adds enormous friction to a game that is apparently trying to remove as much friction as possible from the genre. Image via gamersyndrome.com

One-use-per-battle magic

This is where Chrono Cross starts getting into trouble. The reason that the automated healing shortcut exists is the rules governing the use of magic: each Element can be used once per battle, by spending points earned from successful physical attacks. This means you start every battle fresh – no attrition as you explore – but that also means that over time, as battles become longer, the only way your party can have enough magic available is for the number of Elements you carry to increase more or less linearly with your level. By the end of the game each character has a few dozen Element slots.

As far as combat goes, this is fine. The Element management interface outside of combat, however, is an unmitigated disaster – by far the game’s worst flaw (putting aside whether the narrative stylings of Masato Kato are to your personal taste). Replacing just one of these Elements when you have pages and pages to scroll through is painful enough; swapping one party member for another, requipping all their Elements one by one, will kill you just a little bit each time – and as mentioned before, you’ll want to swap party members pretty regularly for the best color matchups against enemies. The option to fill a character’s loadout automatically doesn’t pass muster, as you will almost certainly want to go in and change half of the suggested choices anyway.

Am I preserving this idea: HELL no. For this and other reasons that aren’t derived from CC, party members will have no more than about ten “spells” equipped at most by the end of the game. Spells will be reusable within the same battle after a brief cooldown – not the most original game mechanic, but CC was a little bit too original in this area.

Turning Points

After 5 years of having a proper job, I’m working on a new indie RPG. As you know from reading my (apparently annual) posts on this thing, I’m interested in older JRPG combat systems, especially from series that have been dormant for 10-15 years. I recently replayed Xenosaga Episode I to refresh my memory of the combat mechanics, which hold up a lot better than its wannabe-intellectual space opera story. There are a lot of crunchy systems in Xenosaga, but I’m particularly looking at the Event Slot, a predictable schedule of effects that change from turn to turn, and the Boost Gauge, a tool that lets the player (and enemies) steal an extra turn to take advantage of those effects. Both of those concepts are also central to the combat in Falcom’s relatively newer game Legend of Heroes: Trails in the Sky and I think it’ll be instructive to compare how the two games execute similar ideas. Of course there are plenty of turn-based RPGs where you can manipulate the turn order by various means (off the top of my head, Shadow Hearts From the New World, Radiant Historia, Bravely Default) but I’m interested in how that idea combines with an effect that distinguishes one turn from another, so that it’s not just about me getting more turns than the enemy, but getting this particular turn that will give me an advantage.

The mechanics

Xenosaga: There are four possible effects in the Event Slot that rotate in order: None, Critical Up (higher chance of critical attacks), Boost Up (earn more Boost Gauge when attacking), and Points Up (increased rewards from defeating enemies). This order never changes. The Boost Gauge is a meter that increases any time you use a normal or Tech attack; it does not increase when casting spells, using items or blocking. Normal attacks add about 1/8 of a boost, Tech attacks 1/4, and the amounts double on a Boost Up turn. A character can have up to a maximum of 3 boosts’ worth of meter stocked. However, a character cannot boost if they are currently acting or their turn is scheduled in the immediate future as depicted in the queue in the bottom right of the screen. Finally, enemies can also boost. They will sometimes boost as a counter attack, which stacks on top of any previous boost command.

Trails: There are a variety of effects, and the types of effect and their frequency vary both between battles and over the course of a single battle. They include guaranteed critical attack, increased damage, bonus loot drops for hitting (not necessarily defeating) enemies, and automatic healing of three character resources – HP, EP (cost of magic), CP (cost of skills). Some random encounters have no turn effects, but boss battles almost always have at least some increased damage & guaranteed critical turns. The schedule is randomized, but the upcoming 8-10 turns are shown on the side of the screen. Characters gain CP when dealing or receiving damage. A character with at least 100 CP can spend all their CP to execute a special skill (“S-Break”) on the next turn. Characters can have up to 200 CP; if they have exactly 200, the effect of the skill is enhanced. There is no restriction on when a character can use an S-Break as long as they have the required CP.

CP is a resource that will frequently be spent on other skills besides S-Breaks, while the Boost Gauge can only be spent on boosting. If multiple Xenosaga characters have more than one boost stocked, it is possible to chain boosts one after the other, preventing the enemy from acting for a long time (unless they also boost!) whereas a Trails character will need several turns to bank the CP required to execute their next S-Break. Finally, the most significant difference: a boosted turn in Xenosaga is no different from a normal turn, but in Trails, only S-Breaks can be used to interrupt the scheduled turn order.

Effects coming up in the next few turns of Trails in the Sky SC (left side): Heal 10%, 10% extra damage, guaranteed critical hit. All four characters are ready to use an S-Break.

The incentivized behaviors

I’ve written before about the risk-reward dynamic of the Points Up slot in Xenosaga, which incentivizes delaying the killing blow in order to get the most points out of every encounter, at the risk of allowing the enemy extra turns. The Boost system gives the player a tool to reduce that risk by blocking the enemy from taking the Points Up turn, but there are good reasons to use boost to take the Critical & Boost Up effect slots as well. Since one of the Event Slot turns increases the amount of boost earned from attacking, it is often possible for a character to boost and then earn back 75% of the BG just spent; in shorter battles this is enough of a return on investment to make it worthwhile.

Because of the randomization, the turn effect schedule in Trails tends to produce situations that require reaction from the player, rather than allowing the player to build a strategy in advance. Interrupts are harder to accomplish so you must choose carefully when in the course of the battle it is most advantageous to steal a turn. In boss battles, most of the time I would save my CP until the boss was about to get a guaranteed critical, since the large majority of S-Breaks are powerful attack skills; gaining a critical and denying one to the enemy is the most bang for your CP buck.

In both games some of the effects are beneficial only to the player – for example enemies gain no advantage from earning extra upgrade points or loot. This provides some variety in the relative value of stealing different turn effects.

The problems

Although in theory enemies earn Boost Gauge according to similar rules to players, and so should be limited in their boosting if you keep them away from the Boost Up turn, this system is entirely opaque, and frequently enemies will boost on the first turn of a battle. When they boost as a non-counter attack it will happen the moment a turn begins, before you can attempt to boost, which I suppose is better than having to try to mash buttons to beat them to it. Nevertheless, when you plan to boost on the next turn and are denied that by an enemy, it feels effectively random and therefore, to me, frustrating as hell. Although the triggers for enemy counter attacks have to be learned, they are consistent, and since they can override the player’s boost input, I think they would provide enough challenge on their own.

Also frustrating: earning Boost Gauge only from executing attacks means that taking a defensive posture is problematic. Whereas taking a big hit in Trails might push you over 100 CP, giving you an opportunity to swing momentum in your favor – very much in the spirit of Final Fantasy’s Limit Break – in Xenosaga a character who has to spend every turn healing or defending will never gain traction. There are some hideous difficulty spikes in Xenosaga, with regular enemy groups able to deal severe damage on multiple extra turns, because of the positive feedback loop of not being able to boost when you’re at a disadvantage.

Me getting my ass kicked by a random encounter 30 hours into Xenosaga Episode I. Shion has to heal the party right now instead of using the critical effect seen in the lower right. I’ll be dead in a few turns anyway.

The biggest issue I have with Trails’ S-Break system is not with the mechanics but the interface. Once you input the interrupt command you are committed to executing the skill, and can only choose a target. Many skills have a limited range, and it’s possible to completely whiff because you are just out of reach. There’s no apparent mechanical reason not to allow the player to cancel out of the attack and let the next turn proceed as scheduled. I’m always annoyed by games with grid-based movement that don’t let you test what you can do from a given position before committing (hi, FF Tactics!) and the movement undo system was the first thing I prioritized back when I made Demon Thesis.

My takeaways

In principle I prefer Xenosaga’s consistent, repeating cycle of effects more than the scrambled version in Trails in the Sky, but there’s no question that Trails executes its idea more successfully. Xenosaga would be improved by either taking away the ability for enemies to boost or at least making the rules for when they are able to boost more transparent – simply showing that they accrue Boost Gauge the same as players, and are limited by how often they can fill the meter, would go a long way to making the system feel less arbitrary. Another idea that I’m exploring in my prototype is the ability to manipulate the schedule of turn effects directly, without changing the order in which characters act. The tactical maneuver of changing an advantageous turn effect into a lesser one, or vice versa, doesn’t need to always be coupled directly to gaining an extra turn, which is an extremely powerful advantage in pretty much any combat system.

Imma Let Ryudo Finish But

Grandia 2 Anniversary Edition is out today on Steam and GOG. Hooray! This is the best combat system in JRPG history, bar none, now available without the questionable English voice acting. Grandia’s combat is much beloved among JRPG enthusiasts, so much so that in the absence of any official sequel from GameArts since 2005, multiple developers have decided to take its fundamental mechanics and build new RPGs around them. This is a look at a few of them and how, while they are great games in their own right, there is no comparison with the original.

Child of Light comes from Ubisoft and is built on the 2D engine used in their modern Rayman games, which finally justifies the existence of that series. A very, very barebones RPG with stunning art, Child of Light’s combat focuses exclusively on Grandia’s central idea of interrupting enemy actions. There is a timeline showing how close each character and enemy is to acting. There is a delay between selecting an action and executing it. If a character/enemy is attacked during that delay, their turn is cancelled and they are thrown back to near the start of the timeline. That’s all there is to it, and it’s enough to deliver a satisfying 15-hour game.

Do I have time for the slow powerful spell that will make the next 5 turns much easier, or will the enemy choose a quick attack that interrupts me? Do I have time to interrupt the enemy’s attack, or should I play it safe and defend? Even in normal encounters you’re faced with interesting choices on a regular basis. The timeline being continuous rather than discrete means the possibility space is effectively infinite, and a slight bit of randomization at the beginning of combat keeps can result in a totally different series of choices to make by the time five or six turns have elapsed.

Jumping back a few years, Zeboyd’s two sequels to Penny Arcade’s On The Rainslick Precipice Of Darkness were the first in the wave of Grandialikes, and while they also streamline and simplify, they preserve a few more of Grandia’s wrinkles. In Child of Light, every attack is capable of interrupting; in Grandia and RSPD, only specific types of attack have this property, and generally speaking you give something up to get that advantage, usually damage. In fact, the most inexpensive interrupting abilities in RSPD do no damage at all. Now the question is not just whether you can beat the enemy to the end of the timeline, but also whether it’s in your interest to sacrifice an attack to cancel theirs, or simply trade damage.

The extent to which you can take advantage of interrupts also depends on your party configuration, and in that area RSPD 3 and 4 borrow heavily from Final Fantasies 5 and 6 respectively. Certain jobs/classes/whatever are more adept at interrupts than others, you can spend an equipment slot on an accessory that gives all your attacks the interrupt property, etc. The situational usefulness of interrupts is brought more to the forefront in RSPD compared to Child of Light.

What sets Grandia apart from these worthy imitators is its use of space in coordination with time. Grandia is about as far removed as you can get from the Final Fantasy model where the party and the monsters are lined up on opposite sides of the screen and step forward one at a time to throw a punch. In Grandia, everyone is running around in a frenzy, all the time – and this has real consequences for gameplay! Melee attacks don’t just happen when the character reaches the end of the timeline; they have to run over to the target, and they’re still vulnerable to interrupts during that time. In fact, in the video above, you can actually see an enemy combo’d to death while attempting to reach its target – an entirely different solution to the interrupt problem. Now in addition to all the other mental math you’ve been doing – am I far enough ahead on the timeline? Is the damage I’m preventing to myself worth the damage I’m not dealing to them? – you must factor in your distance from every enemy on the battlefield and how quickly you can cross it.

Although there is a lot of automatic running around (which does as much to keep battles lively as the butt-kickin’ soundtrack) you can use a turn to move manually as well. This lets you not only set up cancels on the next turn, but also avoid bunching your party up where a single area-of-effect spell can hit them all. And need I tell you what a difference it makes when you gain a character whose cancel attack is ranged? Movement hugely expands the possibility space beyond what Child of Light and RSPD can offer – and since Grandia 2 is a full size 2000-era RPG with a 40-hour storyline, it needs to.

Most of all, what the combination of movement and timing accomplishes is to make what is still fundamentally a turn-based game feel as exciting and visceral as an action game. Running across the map to land a cancel with a fraction of a second to spare feels like scoring a touchdown. Compared to Active Time Battle, which also attempted to liven up turn-based battle, but – well, you know how I feel by now – Grandia not only succeeds but does so while still giving the player time to agonize over those close calls.

Unfortunately, after Grandia 2 there was nowhere to go but down for GameArts; Grandia Xtreme and Grandia 3 gave us more of the same great combat, but with the trite anime teen dialogue becoming increasingly hard to swallow. After they were purchased by mobile giant GungHo it seemed like that would be the end of the story, but with the surprising re-release of Grandia 2 – and the announcement of Star Ocean 5 for consoles by the similarly-positioned studio tri-Ace – perhaps there is still a chance for GameArts to surpass themselves.

RNG Crush: Why I Hope Persona 5’s Main Character Is Allowed To Die

JRPGs rely on randomness to alleviate the fact that you spend a lot of time doing the exact same thing. Random damage values, random enemy behavior, random success or failure of special abilities all ideally help the game to create different situations for the player to react to, starting from the same base components. That’s the good kind of randomness – “randomness you can make an informed decision about,” as my friend James Lantz put it. Conversely, there’s the bad kind of randomness, the kind that causes the player to feel like the game got away with something, and the criminal kind of randomness that causes the player to lose an hour of progress without their having made a mistake.

Atlus’ Shin Megami Tensei franchise and its spinoff series Persona are some of my favorite JRPGs, but I have a bone to pick with a mechanic introduced in the PS2 era that makes them susceptible to the terrible horrible no good very bad kind of randomness. In games like Nocturne, Strange Journey, and Persona 3 & 4, you lose when the Main Character reaches 0 HP – even if other party members are still standing, even if they could easily revive you, because the Main Character is Special For Some Reason. SMT games are more punishing than the average JRPG in the first place; when entering a new area, a random encounter against an enemy whose weaknesses are unknown can easily kill off a party member. Even relatively easy enemies can manage to knock out one character in a surprise attack where they get a free turn. Reviving a single party member is a negligible cost in the long run, but if that party member happens to be the MC, too bad.

I'm not the most creative when it comes to naming silent protagonists

I’m not the most creative when it comes to naming silent protagonists

This mechanic wouldn’t be frustrating if you had any control over the likelihood of the MC being targeted, but it’s pretty much entirely at the whim of the AI’s random number generator. Thankfully, Atlus’ designers are capable of balancing their games so that this design decision, questionable though it may be, doesn’t ruin everything. This is not the case with another game with a unrevivable protagonist, Okage: Shadow King, a very early PS2 game from Zener Works. So far as I can tell, Okage is Zener’s only RPG, only console game, and only game released outside Japan. It has a terrifically funny script, a charming and unique art style, and a broken-ass combat system that sucks every drop of fun out of the experience.

There are two kinds of battles in Okage: trivially easy, and you’re-dead-in-30-seconds hard. Boss fights comprise the majority of the latter; each boss is fairly powerful on its own, and also comes with several monster buddies whose job is to all attack the same character simultaneously and kill them before you can act. Time works in an odd way in Okage; there’s an Active Time Battle-y pause before a character can act, and while time stops while you navigate the menu (yes, somehow this game gets right the one thing Final Fantasy gets wrong) multiple combatants can act simultaneously, which means any number of characters can attack the same enemy at once, or vice versa. That vice versa is a problem. Okage’s party is only three members compared to SMT/Persona’s four, and the AI does nothing to compensate for how likely it is that one character will get ganged up on, and that 33% of the time that character will be the protagonist Ari. In fact, Ari is the only character with the ability to force enemies to target him. Zener. What. Are. You. Doing.



It’s interesting to observe how the Persona series has slowly added more and more workarounds to mitigate the bullshititude of this mechanic without getting rid of it entirely. Since instant kill spells are a staple of the franchise, wielded both by party members and enemy Shadows, in Persona 3 you can collect Homonculi, items that absorb those spells to keep the MC alive. Persona 4 gives your other party members the ability to each absorb one attack that would otherwise kill the MC and end the game – but only if the MC is the sole target of the attack. Unfortunately Atlus’ final bosses almost always have some effectively unavoidable random instant kill combo. For example, the final boss of Persona 4 has spells with a chance to inflict status ailments on the whole party, a spell that instakills any character with a status ailment, and two actions per turn. None of the mitigating effects mentioned earlier apply to that instant kill. Good luck.

Finally, in the recent spinoff-of-a-spinoff Persona Q, you have to specifically opt in to this mechanic by choosing the “Risky” difficulty setting (separate from “Hard”). If you voluntarily subject yourself to this nonsense, I don’t know what to say to you, except maybe you’d also enjoy life more if you rolled a die every morning and slammed your fingers in a drawer on a 1. Like the menu selection quick-time events of Active Time Battle, the unrevivable protagonist is a bad way of making the game harder. Meanwhile, SMT4 abandoned the gimmick entirely and was still really hard and nobody missed it, and I hope that continues to be true in Persona 5, or at least keeps it optional as in Q.

Taking one for the team will end your game, so in Persona 4 sometimes the team takes one for you

Taking one for the team will end your game, so in Persona 4 sometimes the team takes one for you

Of course, randomness can go wrong in the player’s favor as well. The instant kill spells mentioned earlier are normally a high risk/high reward option. Persona Q makes the player’s instant kill abilities far too powerful; not only can they be used repeatedly without the expected long term drawback (concisely described here by Robert Boyd of Zeboyd Games – who by the way makes very smart decisions with regard to random elements in his modern JRPGs) but for a good stretch of the game they are effectively guaranteed to work unless an enemy blocks that type of attack completely. In the entire fourth dungeon of the game I saw no more than five instakillable enemies survive my first attack, even if they supposedly were “resistant” to it. The success rate went back down somewhat in the final dungeon, but using instakill as much as possible remained the best strategy.

In other SMT games I consider instakill too unreliable to use regularly, but if I’m about to lose a fight I’ll take a gamble and hope to be pleasantly surprised. In Persona Q it became the only rational thing to do every time, even though it spoiled the fun, and I’m sure I wasn’t alone in making that choice. Commenting on Destiny’s loot cave exploit last fall, Andrew Vestal quoted Diablo’s designers: “Players will always play the most efficient way, even if it’s boring, and they’ll blame YOU.” Of course, if the monsters in Atlus games acted rationally and efficiently, they would focus fire on the main character and win much more easily. So really the beloved heroes of Persona 3 & 4 only succeeded because Shadows are dumb as hell.

Valkyrie Profile 2: Purify Weird Game

tri-Ace’s Valkyrie Profile is weird even by JRPG standards. It throws out nearly every genre convention in order to tell the Norse story of Ragnarok in RPG form, much though it may deviate from the source material. The valkyrie Lenneth must prepare Valhalla for cataclysmic war by recruiting the worthiest dying mortals to serve as divine soldiers (einherjar): therefore there is a limit on the player’s activities before Ragnarok – the endgame – begins, and those activities consist of visiting einherjar at the moment of their tragic demise, then delving into dungeons to level them up before sending the most heroic to Valhalla, never to return.

Everything that you do in Valkyrie Profile follows from this premise. (All Valkyrie Profile images via yannick.fleurit.free.fr)

Although some of the einherjar’s stories are interconnected, they have little to no story role as soon as they become party members. The bulk of the game’s storytelling occurs in disjointed vignettes that establish the tone of the setting but don’t give you much reason to get attached to specific characters – which is just as well since you have to let a good number of them go in order to fulfill Odin’s demand for recruits. You’re free to view these scenes in any order, though only a few are revealed during each of the game’s eight chapters. At the end of each chapter you’re shown how the quality of the einherjar you transferred affects progress in the Valhalla war effort, as well as the bonus you receive from Odin. All of this combines to create a game with a very specific mood and a very circumspect story. Lenneth is more or less defined by her job, and the particulars of how she does it have a gameplay impact but are mostly irrelevant to how Ragnarok unfolds.

…at least, that’s the case if you get the “B” ending. If you notice an odd feature of the status menu, if you piece together vague references in certain characters’ dialogue, if you do the exact right thing in the exact right place at the exact right time – in other words, if you use a walkthrough – you can instead trigger the “A” ending, which reveals the true nature of Lenneth and gives her a real character arc. Almost the whole of the game’s actual plot, the stuff that would be doled out in cutscenes every 30 minutes in Final Fantasy, is locked behind an incredibly arcane puzzle that demands counterintuitive actions at extremely specific moments spread out over the course of a 30-hour game – which boils down to tri-Ace doing for storytelling what it normally does for game mechanics.

Valkyrie Profile combat looks turn based until you press a button, and then things get craaaaaaazy.

VP’s battle system bears more resemblance to Marvel vs. Capcom than to Final Fantasy. Your party chains attacks together into combos, which need to be carefully timed just as in a fighting game – e.g. a low attack will miss if the target has just been launched into the air. Juggling enemies yields experience bonuses; pounding an enemy that’s been knocked down reduces the cooldown on special attacks. Those special attacks can only be used after a combo long enough to fully charge the special attack meter, but multiple specials can be chained together if used in the correct order. You’re rewarded for experimenting with the composition of your party, the order in which their three attacks are used, and the order you send them into a combo. Unfortunately, once you’ve found a pattern that lets you chain as many attacks as possible, you are rarely forced to switch it up; aside from a few enemies with small hitboxes, your technique will be universally applicable. Strategy boils down to choosing the order in which you’ll beat down on enemies – especially important in some Hard mode battles where enemies can revive each other or absorb the strength of defeated comrades – and when to give up a character’s turn to use healing items or spells.

Despite the dynamic feel of combat, it quickly becomes repetitive, so it’s good that most non-boss enemy encounters can be avoided while exploring dungeons. Exploring in VP means 2D platforming, rather than the usual maze running. Combat is initiated by touching an enemy, and it’s usually not very difficult to play keep-away if you don’t feel like fighting. Enemies don’t respawn until you leave a dungeon entirely – and since re-entering a dungeon advances the Ragnarok clock, this ensures that the experience you need to prepare yourself and your einherjar for the endgame is a limited resource. The other major mechanic in the exploration phase is Lenneth’s ability to shoot crystals which can stun enemies or create platforms. The dungeon design gets a lot of mileage of this twist, but overall the platforming in VP is at best a novelty and at worst a frustrating mess – most of the dungeons exclusive to Hard mode have some kind of gimmick that is far too demanding for the sloppy jumping physics you have to work with, and failing often means being sent back to the beginning, or even being kicked back to the world map, pushing the clock forward and most likely causing you to reload your save.

Most of these don’t do what they sound like they should do. Some don’t do anything.

tri-Ace loves giving you a huge list of skills to choose form when developing characters. In VP, many of those skills don’t have any effect in battle, but are instead used to assess the capability of einherjar sent to Valhalla. The distinction isn’t always clear, either – Find Trap will not show you traps in dungeons, of which there are plenty, but an einherjar sent to Valhalla with Find Trap will occasionally get an award for detecting a trap. Hooray! But that’s nothing compared to the opacity of the Transmutation system that converts one item into another – or maybe a third or fourth type, if you have the right item equipped… one of which you create by transmuting something else… again, the sheer possibility space means you’ll be using a walkthrough. And if you use transmutation to its fullest potential, you end up with equipment that frankly breaks the game, even on Hard difficulty.

So Valkyrie Profile sets itself apart from other JRPGs in nearly every aspect of its design and execution. Not all of those differences are for the better, but when it fails, it fails with character. Six years later, tri-Ace released Valkyrie Profile 2, which took the essential mechanics of the original and transplanted them into a much more conventional, and much more polished, JRPG. The narrative conceit of Ragnarok’s impending arrival, and the game structures built around that premise, are entirely absent: VP2 follows a party of eccentrics on a linear quest to find a MacGuffin, traveling from town to dungeon to town, resting at inns, buying items from shops with money, and advancing the story through a steady trickle of cutscenes, like every other JRPG. Transmutation is gone, and there is only one end to the story. Einherjar can be collected in dungeons (with only a brief line of dialogue to establish them as characters) and released after leveling up, but the reward is merely a stat-boosting item. Many character skills are available, but all have a clearly defined purpose, though many are highly situational.

The sequel is a prettier game, and it’s not always just about polygon count – like the visually pleasing rune mechanic that unlocks new skills.

None of this is to say VP2 is simplistic or unoriginal. Skill acquisition by equipping items is a common trope, but here a combination of items in adjacent slots on a grid is required to unlock each skill. While this could have been an overwhelming combinatorics problem, giving each item one of four colors and limiting each skill to a color-matched set gives the player a friendly visual handle on the possibility space, and the equipment grid is a very pleasant piece of user interface design. Monster parts can be sold to stores to craft powerful items and equipment, but how those parts are obtained is married beautifully to the unique Valkyrie Profile style of combat (more on that later).

One of VP2’s most unique additions is the ability to collect and manipulate Sealstones, which create dungeon-wide effects on the party and enemies. This single mechanic enables not only some very clever puzzle designs in dungeons but also another layer of player customization that gently ramps up  in complexity over the course of the game as more sealstones come into your possession and more of them can be equipped at once. (Among the few bits of unfortunately opaque design in VP2 is that those slots are gained by completing optional dungeons which only appear if you talk to certain random NPCs, and sometimes talk to them more than once; NPC interactions are otherwise totally disposable, so it’s easy to miss these very important rewards.)

The Sealstone system will throw a new set of variables at you in each new dungeon, and then let you customize those effects to your taste.

The Sealstone system will throw a new set of variables at you in each new dungeon, and then let you customize those effects to your taste.

The pieces of Valkyrie Profile that survive in VP2 are the platforming dungeon sequences and the basic fighting-game-y combat mechanics, but each of those elements is refined and elevated here. The feel of VP2’s dungeon exploration is terrific, and even though the platforming is much improved, the finicky twitch-based jumping puzzles are gone, just to be safe. Replacing Lenneth’s crystals are new protagonist Alicia’s photons; these let Alicia jump on floating enemies by freezing them, in the finest tradition of Samus Aran. Even better, a second photon fired at a frozen enemy swaps its position with Alicia’s, opening up a whole new set of possibilities for platforming puzzles, not to mention lightning-fast movement across areas you don’t feel like fighting in. Getting all the treasures in each dungeon usually requires some very creative use of photons, though the item you get is usually less satisfying than the feeling of accomplishment. Yggdrasil, the third-to-last dungeon of the game, is a brilliant challenge that melds platforming, photons, and sealstones together in a completely new and intuitive way.

Alicia can not only create platforms out of frozen enemies but also swap positions with a second hit.

Alicia can not only freeze enemies to create platforms but also switch positions with a second hit.

The improvement to dungeon exploration would have been enough, but VP2 also pushes the original game’s combat forward by literally adding a new dimension. Battles take place in an arena which the party must traverse to reach enemy targets. Enemies can attack if the party steps into their target range, a visible cone or circle on the map. Enemies only move when the party moves, but the party can also dash to pass through a danger zone unscathed, at the cost of Action Points, which are regenerated by moving normally. This, in itself, is tremendous fun, and you’re incentivized to do it a lot since there is a big XP reward for heading straight for the lone enemy leader, whose defeat means immediate victory. I love the leader idea and the snappy pace of combat it allows. The faster the leader goes down, the more bonus magic crystals you get – magic crystals still being the reward for juggling enemies with a mid-air combo – and in VP2 these serve not only as a multiplier to the XP reward in battle but also, in a nice touch, as the currency used to permanently add sealstones to your collection. Although I didn’t often make use of it, there’s also the ability to split your party into two groups so the beefier characters can draw fire away from their fragile partners. So tactical movement is a crucial aspect of the new battle system, and this goes a long way toward relieving the grind of combat.

There are flaws with this system, for sure; characters will sometimes unexpectedly get stuck on geometry during a dash, leaving them right in the line of fire. The fact that enemies have multiple attacks with different ranges, and their chosen attack often changes following a player action, means a safe spot regularly becomes an unsafe spot without an opportunity to react. This is forgivable since otherwise you could probably finish 99% of battles without the enemy taking a turn, and also because you can manipulate this phenomenon to your advantage by using a menu action to force the enemy to switch to a more easily avoidable attack.

VP2 battles take place on a tactics-y sort of map; stay out of the red zones and enemies can't attack.

VP2 battles incorporate a mild level of tactics; stay out of the red zones and enemies can’t attack.

The battle menu is an interesting idea but definitely the one that I’m most ambivalent about. Casting non-attack spells, changing equipment, and using items all make use of the battle menu, and once any character does any of those, the menu is unusable until a cooldown period elapses. Moving, attacking, or being attacked will advance time until the menu is available again. Initially this requires judicious decisions about healing, especially in boss fights, since you’ll probably only get to use one item or spell between the boss’ big party-wide beatdowns, but that soon becomes a moot point as you accumulate enough money to have 99 heal-everybody-to-almost-full items on you at all times. Ultimately the menu limit ends up only being relevant in desperate scenarios when your party is down to one survivor and you need to slowly reverse momentum with a series of revivals and healing. In that case, having to avoid contact with the enemy for such a long time can be very challenging, but not exactly fun.

Once your party is in position and you press the first attack button, the camera switches to a side-on view and the familiar mode of combo-based combat from VP takes over. The major difference is that each attack now consumes AP, which complicates fights against enemies that can survive a single turn of all-out assault. Whereas in VP you could attack the same number of times each turn and there was rarely a reason to do less, VP2 frequently requires you to choose between attacking now with a limited AP pool or trying to move around the battlefield, putting the party at risk, to put more AP into a single combo. Or, since being attacked restores a large chunk of AP, you could just stand in front of the target and trade blows – but unless you’re doing a lot of grinding, enemies in VP2 hit rather hard.

Enemies in VP2 are made up of several body parts - this attack made contact with the gryphon's legs.

Enemies in VP2 are made up of several body parts – this attack made contact with the monster’s legs.

The other new mechanic when attacking is enemy body parts. Where you position yourself on the field in relation to the enemy, and where your attacks land, determines whether you hit their sensitive tail or their armored plating. Enough sustained damage to a single body part will break that body part off, which may yield a crafting item (told you we’d get there!) and may also activate Break Mode, a short period during which you can just button mash the enemy into paste without worrying about AP usage. Visual flourishes and some nice sound design make it viscerally satisfying when skeletons explode with goodies like an undead piñata.

Special attacks are back, and the rules for using them are the same except there is no per-character cooldown between rounds; if you’ve got the AP to get the meter to 100, you can use your specials. Blessed be tri-Ace for adding the ability to skip past the special attack cutscene that you see every single time you use it; by the end of my recent VP playthrough I was building my strategy around how to minimize the number of times per minute I had to watch them. One step backward in VP2 is that mages always have the same special regardless of the spell they have equipped, whereas in VP there was a corresponding special for each regular spell; this means if you run into a type of enemy with an elemental resistance to your mage’s special, you can’t just change spells, you have to change characters, which is something the game makes you do frustratingly often as it is.

Party management in the original VP may be odd, but in many ways it’s much friendlier to the player than a more conventional JRPG. Completing dungeons and solving puzzles rewards you with Event XP, which can then be freely applied to whichever characters you wish at your leisure. You can level up your mainstay characters faster, or – my preference – get the characters you haven’t been using up to snuff before sending them to Valhalla. It’s very nice to have that buffer rather than constantly juggling your party formation to keep everyone equally leveled. By contrast, VP2 is a downright bastard about surprising you with characters leaving your party at the whim of the story – sometimes right before tough boss fights – and instead of a bank of Event XP you get a handful of items that award a certain amount of XP to one character, that amount being insignificant by the endgame.

Sure, the localization is better in the sequel, but the original has so much more personality. Polish isn't everything!

Sure, the localization is better in the sequel, but the original has so much more personality. Polish isn’t everything! …no you’re being nostalgic. Shut up.

Still, this is the only aspect of VP’s design which I would call mechanically superior to its sequel. VP2 throws as much stuff at the player but that stuff is more coherent, more polished, more balanced. There’s no doubt in my mind that in terms of mechanics the sequel is the better game.

But though I might sometimes wish otherwise, mechanics aren’t everything, and VP2’s failure to connect on a narrative or emotional level taints the whole experience. It got rid of a lot of problems the first game had, but it may have thrown the baby out with the bathwater. The countdown to Ragnarok is a compelling idea to build your game around; there is no compelling idea at the center of VP2. The dialogue in VP is stilted and the voice acting is notoriously uneven, rather behind the curve as far as turn-of-the-millenium Japanese localizations go, but VP2’s storytelling, with the benefit of improved translation and okay performances, is just incredibly boring. In terms of story, the relationship between VP and VP2 is roughly analogous to that between Chrono Trigger and Chrono Cross – the sequel takes a fairly minor unresolved plot thread and builds an entire new game out of it, more or less abandoning the theme and tone of the original, and also wreaking such havoc with convoluted time travel metaphysics shenanigans that you’re left not knowing whether anything that happened in either game ends up mattering. But Chrono Cross has its own identity independent of Chrono Trigger, and a charming one at that, while VP2 never aspires to be more than an appendix to its predecessor.

Even though this avalanche of confusing data is mechanically irrelevant, it's all in service of the game's core concept that your einherjar are needed for Ragnarok. Valkyrie Profile thrives on messy details like this while its sequel is very tidy and totally bland.

Even though this avalanche of confusing data is mechanically irrelevant, it’s all in service of the game’s core concept – that your reason for existing is to recruit einherjar to prepare for Ragnarok. Valkyrie Profile thrives on messy details like this while its sequel is very tidy and totally bland.

Valkyrie Profile’s story didn’t need a sequel (or prequel), but unfortunately tri-Ace seems unable to hold onto nice things; its flagship RPG franchise Star Ocean ended up being sunk by sequels that needlessly undermined the story its fans were invested in with retcons that made the whole setting absolutely pointless – sound familiar? Given that Valkyrie Profile 2 isn’t really about being a Valkyrie and doing Valkyrie things anymore, conceivably it could have been given a completely new setting, story and title and been just as fun without the baggage of being a sequel to a beloved game… but new IP is risky. tri-Ace’s most recent console release, excluding their contract work on the sequels to Final Fantasy 13, is Resonance of Fate, which mostly flew under the radar – despite having a terrific battle system, according to Bob Mackey on this episode of Retronauts on RPG battle systems (which you should really listen to even though it lionizes Active Time Battle more than I would). Maybe I wouldn’t have played Valkyrie Profile 2 if it didn’t have the name, and that thought makes me sad. Because it’s a great game… it’s just a lousy Valkyrie Profile.


From the beginning I knew I wanted this blog to look at the mechanics of Chrono Cross. It was my first JRPG, and changed my gaming taste forever. (It was not, I later realized, my first RPG – that turned out to be Steve Meretzky’s 1994 adventure game/RPG mashup Superhero League of Hoboken, though I couldn’t have put a name on either genre back then.) Thirteen years later, the clunky party management interface and Masato Kato’s storytelling excesses are more obvious, but back then I went in with essentially no expectations, and certainly no knowledge of Chrono Trigger to spoil the fun. Despite its flaws I find it holds up much better than Square’s PS1 Final Fantasies. And let me be the 1,000th person to tell you that Yasunori Mitsuda’s soundtrack is in a class above anything else in games.

Capturing everything I love about the battle system of Chrono Cross seemed like a daunting task, which is why I kept putting off writing about it, but I think the majority of what makes it great, and the design lessons I would take from it, can be distilled down to two fairly simple ideas:

Karsh is very very very wrong. (Image via The GIA)

Karsh is very very very wrong. (Image via The GIA)

1) Everything you do has multiple effects, which means no decision is as simple as it looks.

In a typical RPG, successfully attacking an enemy deals damage. In Chrono Cross, successfully attacking an enemy:

  • increases the probability that subsequent melee attacks will succeed,
  • charges the Element meter that powers magic and other special skills,
  • restores other party members’ stamina so they can take more actions (this happens even on a miss), and
  • deals damage

Using an Element not only does whatever it says in the description but also adds the Element’s color to the Field Effect, calculated from the color of the last three Elements used, which modifies the effectiveness of all ally and enemy actions based on their innate color. If two of the last three Elements were red, blue characters are weakened, and vice versa. And in the rare case that the Field Effect is pure red, the red summon Element can be used, but since the colors of enemy Elements are also included, getting a pure Field Effect can be tricky.

Multiple effects of a single action is the shortest path to making something interesting happen in your system. Knowing that every action you take changes the game state in a non-trivial way (i.e. not just dealing damage and moving in a straight line toward victory) makes every choice an interesting one. But when every choice is interesting, you can just as easily induce paralysis through overwhelming complexity of the decision tree. So it’s just as important that…

There beeba lot of information to process here, but it's never overwhelming. (Image via lparchive.org)

There beeba lot of information to process here, but it’s never overwhelming. (Image via lparchive.org)

2) Despite the complexity of the systems, the game teaches you a viable “default” strategy to fall back on.

The maximum amount of stamina a character can have is 7. This is enough to perform a light attack (costs 1 stamina), a medium attack (2), and a heavy attack (3) before casting an Element, which can be performed so long as the meter is at least 1 (even though it costs 7). Because heavier melee attacks have a lower probability of hitting, but each successful hit increases that probability, the light-medium-heavy combo is an obvious progression that strikes a good risk/reward balance. And since Elements necessarily end your turn because of their high stamina cost, and require melee hits first, they make a natural combo finisher. By default, the interface will explicitly recommend this strategy, automatically advancing the cursor from 1 to 2 to 3 as long as you connect with each attack and bring the next attack level above an 80% chance of success.

For most normal encounters, you don’t really need to deviate from this pattern, reducing the complexity of the decision tree to which party member attacks what enemy, and which Element to cast at the end. This strategy won’t suffice for tougher enemies and bosses – for instance, you might want to think about defending once in a while – and it can be rewarding to experiment with a fancier approach, jumping between party members that still have stamina and then unleashing multiple powerful Elements back-to-back. In fact, solving the “final boss puzzle” to get the good ending requires exactly this kind of stamina/Element coordination across your party.  But by establishing the 1-2-3-element template early in the game, Chrono Cross gives the player a comfortable place from which to start exploring the depth of its systems.

Pulling this off means employing your mastery the game's mechanics toward a completely different goal than you're used to - my favorite kind of final boss. (Image via lparchive.org)

Pulling this off means employing your mastery of the game’s mechanics toward a completely different goal than you’re used to – my favorite kind of final boss. (Image via lparchive.org)

In terms of battle mechanics I only have one real quibble with Chrono Cross’ design, which is the lack of any indication of the enemy’s turn order. Until recently I wasn’t even aware there was a predictable pattern to enemy attacks (each enemy has a counter and after that many player actions they get a turn), but since it is predictable if you count manually, why not let the player see the number? If your answer is that it adds to the difficulty, please see my post on Active Time Battle for my opinion on adding difficulty through a gimped interface. Still, it’s not nearly enough to taint the joy of playing the game, and if I were going to “fix” anything in Chrono Cross, it would be the miserable UI experience of moving a full loadout of Elements from one of your 4,000 party members to another.

Death and Taxes in Rogue Legacy and Dark Souls

I know you’re just exhausted by the multiple-post-per-year pace of updates around here, so let’s take a break from turn-based combat systems and talk about the progression systems of two very, very good action RPGs.

Welp. Image via PCGamer (http://www.pcgamer.com/review/dark-souls-review/)

You did the thing they told you to prepare to do. (Image via PCGamer)

When you die in Dark Souls – and if you know nothing else about the game, you have certainly heard how often you die in it – all your unspent currency (souls) drop in a bloodstain at the spot you were standing a few moments before death. If you can return to that spot before dying again, you can reclaim all that currency, plus anything you collected on the return trip. In order to spend the souls, you have to reach either a bonfire or a merchant. The risk of Dark Souls is overextending – going so far without banking your earnings that when you die, your bloodstain is too difficult to reach. If you have enough souls to convert into something permanent, the conservative play is to head back to safety immediately. Using a bonfire to heal and level up will cause any enemies you defeated to respawn, but as those stats increase, along with your skill and confidence as a player, you begin to view enemies as fodder for growth rather than obstacles. After all, as long as you can get as far as your bloodstain, nothing is lost forever – and with each cycle you accumulate more souls. But seasoned Dark Souls players will also tell you: ignore the souls, unless you’re specifically grinding. Souls are a renewable resource, and only a means to an end in any case. Permanent progress through the world – opening shortcuts, collecting keys, defeating bosses – is paramount.

You CAN take it with you, but you can't come back with it. (Image via Steam user Interloper)

You CAN take it with you, but you can’t go back with it. (Image via unfortunate Steam user Interloper)

When you die in Rogue Legacy, you keep all the currency (gold) you collected, but you must spend it before re-entering the castle; Charon claims anything you have left before allowing you to enter. At great expense you can invest gold in a skill that reduces the amount he takes from 100% down to a minimum of 70%, and I regret having gone down that upgrade path in my first playthrough – I didn’t recoup anything like enough gold to justify the thousands I spent on the discount. If your run was particularly short, you may not be able to afford any of the available upgrades or equipment, in which case you get nothing from it. The risk of Rogue Legacy is not reaching a milestone – the gold required for a new upgrade – before death. It is always best to go as far as possible in one run, to have as much gold in hand as possible when the inevitable happens. Until you discover a boss door, that is – then, if you are prepared for the fight, you can pay the architect, who takes a cut of any gold you earn in return for locking the map on your next run, allowing you to teleport straight to the boss at full strength. The game steps right up and tells you this is a great idea. As in Dark Souls, it’s best to have an objective in mind on any given run. Currency, or progress? The customizable equipment and runes let you decide how to balance combat mastery and gold farming.

SKILLZ (Image via Steam user Cannibalpea)

The money pit. (Image via Steam user Cannibalpea)

In both games, the cost of an extra point in each stat or skill increases with every purchase. How you invest points doesn’t affect the cost in Dark Souls – your 21st upgrade will cost the same every time, regardless of the distribution of the previous 20. As a result, veterans tend to focus on two or three primary stats to level and invest as little as possible in the others; a level added to a mediocre stat drives up the cost of an excellent one. In Rogue Legacy, on the other hand, certain upgrades always cost more than others. Incremental increases to base stats are the cheapest, while high-return skills like new character classes and boosts to gold collection are much more expensive. But again, buying a level of any skill increases the price for all future skills.

Since you can’t save up gold, a twisted thought wormed its way into my head as I watched the price of the gold boost skill become more and more exorbitant: should I start throwing my gold away instead of dumping it into cheap upgrades, so that the expensive ones – which provide a huge return on investment – don’t become even harder to attain in a single run? I didn’t follow through on this idea, and in hindsight I’m pretty sure I was right not to – although the little upgrades don’t make a huge impact immediately, longer survival times correlate pretty directly with bigger gold earnings. But in the moment, standing between Charon and the castle, it wasn’t clear why I was desperately buying anything I could to avoid “wasting” a currency that existed in infinite supply – to a more skilled player.

The combat system of Dark Souls – attacks, blocks, dodges, distance from the enemy, and how the stamina bar ties them all together – is such a confident, elegant design that I get wistful thinking about it. It is possible to beat Dark Souls on skill alone – without spending a single soul. I am not capable of doing it. I need the safety net of more health, more damage, and especially more stamina, so I grind and grind. But where in most RPGs that currency is a marker of my progress – I made it to level 40! – in Dark Souls, it’s a monument to my shortcomings: I had to dick around earning 40 levels before surviving the Capra Demon and his effing dogs.

He is about to PUNCH the Capra Demon to death. With no shirt on. (Image via NeoGAF)

He is about to PUNCH the Capra Demon to death. With no shirt on. (Image via NeoGAF)

It may also be theoretically possible to beat Rogue Legacy without spending gold, though I can barely imagine surviving the final boss’s attacks without unlocking at least one double jump or dash. But looking back, I’d argue that denying yourself permanent upgrades isn’t how you demonstrate mastery of Rogue Legacy, as it would be in Dark Souls. Dark Souls is thoroughly deterministic; like Mega Man, its levels and the behavior of enemies can be (must be) learned. By contrast, the test of Rogue Legacy is managing the unexpected – random levels, random pickups, and most of all, random characters. After each run, you choose from three new random characters to play as, a combination of a class and a few helpful or harmful traits. The skills, equipment and runes you unlock with gold let you play to the strengths of your current character, and survive while exploring each freshly generated map.

So yeah, I was around level 100 when I beat Rogue Legacy. But I beat it with a character afflicted with Vertigo – i.e., the screen was upside down, with left and right inputs switched. Therefore, in conclusion, I’m awesome.

What's up? Image via Rogue Legacy on Wikia

What’s up? (Image via Rogue Legacy on Wikia)

…In actual conclusion: Dark Souls, just two years old, has already become a touchstone for indie developers who want to differentiate their games from the press-this-button-to-win stereotype of current mainstream offerings (I’ve lost count of the games on Kickstarter and Greenlight that are “like Dark Souls but”). Rogue Legacy is this year’s breakout indie hit, likely to inspire further iterations on the “roguelike plus X” genre. I hope that these future games learn the right lessons from the success of Dark Souls and Rogue Legacy – and aren’t just punishingly difficult for difficulty’s sake, but deftly incorporate death and penalization into their progression loops.

Plans Within Plans: Optional Goals Spice Up JRPG Combat

JRPGs tend by nature to be repetitive. You slog through the same random encounters over and over, and the same strategy works every time; occasionally a preemptive strike will shake things up, or leveling up will unlock a new ability that changes the boilerplate formula you apply for the next hour or two. Giving the player optional goals to pursue during combat can make that slog a lot more engaging, especially if those goals come into conflict with the primary goal of “win the battle efficiently.” The simplest example is the ability to steal items, common in Final Fantasy and elsewhere – you can get by without doing it, by and large, but it usually pays to trade some time and damage output for a chance at extra resources. Here are three games that provide great reasons to think about the long-term effects of your choices in battle instead of sleepwalking through.

Grandia Xtreme: XP Bonus for taking no damage

A screenshot would not do justice to the symphony of chaos that is a Grandia battle.

Grandia’s exceptional combat system frankly doesn’t need any tricks to keep the player interested – if anything needs help in Grandia it’s the bottom-of-the-anime-barrel writing and voice work – but the dungeon crawler spinoff Xtreme is more of a grind than the mainline entries. So it offers an optional challenge in every battle: if you can win without taking any damage, which probably means before the enemy gets a turn, you get a bonus multiplier on the XP earned in that battle.

For fights against enemies that are much lower level than your party, this is pretty easy to achieve – which is nice, because as you’d expect, the XP you get from fighting scrubs is unimpressive. For fights where the monsters aren’t such a pushover, deciding whether to go for the XP bonus requires some risk/reward calculation: it’ll probably take some expensive special moves to wipe the enemy out that quickly. If you fall short, you’ll have blown resources for no gain, but if you pull it off, the XP boost is substantial.

Skies of Arcadia: Equip the right color weapon on the last turn

Skies of Arcadia characters use Fruit Gushers to cast magic. Don't ask how Silver tastes.

Skies of Arcadia characters use Fruit Gushers to cast magic. Don’t ask how Silver tastes.
Image from lparchive.org

Enemies in Skies of Arcadia are each aligned with one of six colors (elemental affinities, in function if not flavor), and will take more damage when you use attacks of a superior color, determined by a rock-paper-scissors graph: red beats green beats purple, and so on. On each character’s turn you can rotate between any of the colors you’ve unlocked so far.

At the end of each battle, your party earns both normal XP and magic XP. Each color of magic gains XP independently, and the color of weapon each character has equipped when the battle ends determines how the XP is invested. The colors are far from interchangeable; all the healing spells are green, and silver has both the revival spells and the instant death attacks.

So long-term, you want to end battles with the color of weapon that will unlock the spells you need most, even if in the short-term it hurts your damage output to use anything but the color the enemy is weakest to. The rate of random encounters in SoA is super high, and the combat itself does little to provide variety between battles (though it has one great mechanic in the shared SP bar, which I will talk about more in a later post and which I totally ripped off in Demon Thesis). Switching your characters’ weapon colors each turn to pursue either short-term combat efficiency or long-term character development adds an interesting wrinkle that the game needs badly.

Xenosaga Episodes I & II: End the battle on a bonus rewards turn

The SKL icon in the lower right means this is a good turn to kill a dude. Image from IGN

The SKL icon in the lower right means this is a good turn to kill a dude.
Image from IGN

Xenosaga takes (and merits) a lot of abuse, but there are some really great ideas in each installment’s combat system. My favorite is the Event Slot from Episodes I & II – a sort of turn clock which rotates through a set of effects that drastically alter the decision making process. One slot increases damage from attacks on that turn, one slot increases the efficiency of charging the “boost meter” that allows for interrupting the normal turn queue, and one slot grants bonus XP* for killing enemies while it’s active.

Because I am mildly obsessive-compulsive, I went to extreme lengths to ensure that as many enemies as possible died while that bonus was in effect. This might only mean holding back a deathblow until the right moment, but sometimes I had to spend a boost to get out of a cycle where the bonus always came on the enemy’s turn. If I accidentally killed an enemy a turn too quickly because I misjudged how much damage I could deal in a turn, I felt like I messed up.

Did I mention the bonus randomly varied between a factor of two, five, and ten? Even a less (ahem) dedicated player than myself would have to stop and think real hard about how to time their actions just right to take advantage of a tenfold increase in reward.

*It’s not actually “XP”, it’s Tech Points, Ether Points, and Skill Points, but c’mon.


Active Time Battle Drives Me Crazy

There is a gradient of abstraction in RPG combat systems; at one end you have the pure turn-based systems, in which you choose from a defined set of possible actions and the result is simulated, e.g. the original Final Fantasy; at the other end you have pure action systems, like Dark Souls, in which every facet of the action is determined by your fine-grained input on movement and timing. There is also a gradient in the combinatorial complexity of decision making in RPGs. Depending on the design a character may have a handful of spells and items to choose from, or hundreds.

The interface by which the player navigates the game’s decision tree and executes their plan of attack should reflect how abstract and how combinatorial the combat system’s design is. This old Penny Arcade strip about Final Fantasy 12’s Gambit system, while in my opinion not a very smart take on FF12 per se, makes the point that game interfaces are not one-size-fits-all.

Until taking screenshots for this article I had forgotten about the existence of "Drink."

Until taking screenshots for this post I had forgotten about the existence of “Drink.”

Japanese RPGs began as (and in some studios’ catalogs, continue to be) highly abstract games with moderately deep complexity; in the original Final Fantasy for NES you can face upwards of eight enemies in a battle, spellcasters might have a dozen or two spells, and there are three types of usable item. Given that the NES controller has just a directional pad and four buttons, a nested menu interface was probably the only sensible choice for FF1. For each layer of the decision tree a menu lists the available choices; you scroll through until you find the choice you want and hit A to confirm. Since then, menus have been a mainstay of JRPGs – even when the underlying gameplay shifted.

Square’s first Super Nintendo entry in the series, Final Fantasy 4, introduced the Active Time Battle system. Unlike previous installments, FF4 considers a “turn” to be a character-specific event, with a frequency independent of other characters’ turns. Inspired by watching Formula One racers lap each other, the designers gave each character a speed which determined how often they could take action. Each player character (and enemy) has a meter that fills up over time according to their speed; when the meter is full, they are ready to receive input. Status effects can speed up, slow down, or even stop individual characters’ meters. Depending on which action the player selects, the character spends a certain amount of time preparing the action and then executes it – powerful magic takes several seconds, while a basic attack or item is nearly (but not quite) instant. A caster with a sliver of health remaining might not survive the prep time required to perform a battle-ending spell. The decision tree looks essentially just like the original Final Fantasy’s, with more spells, more items, and time simply being another factor to weigh against others.

The ATB system made battles far more engaging and fast-paced than the original Final Fantasy, never going more than a few seconds without requiring the player’s input, and it certainly succeeded in increasing the series’ appeal. Though the long spellcasting times were abandoned after FF4, Square kept the core conceit of ATB around for each of its SNES and Playstation Final Fantasy games, plus perennial “best RPG ever” contender Chrono Trigger. When an American gamer thinks “JRPG”, their mental archetype is probably a game with Active Time Battle.

The new mechanic introduced by FF4 is not  visible in the interface.

The new mechanic introduced by FF4 is not visible in the interface.

I hate Active Time Battle. I also think it’s objectively bad, which is not the same thing, but the reasons for both are the same.

Here’s an extreme hypothetical scenario to demonstrate what drives me crazy about ATB. At the end of a grueling battle, you have one surviving character with 1 HP, facing one remaining enemy also with 1 HP. Your character’s ATB meter fills up slightly before the enemy’s. If you can hit the “confirm” button twice – once to select Attack, once to select the target – in the gap between the menu popping up and the enemy starting its attack action, you win. If you can’t, game over. The “time management” skill that ATB demands from the player extends to managing the time your fingers take to enter inputs through the menu system – introducing a twitchy, execution-centric challenge into a game with an interface designed for complex, abstract decision making. Best of all, in the original version of FF4 you can’t even see the meter to anticipate the appearance of the menu! I have lost many a Final Fantasy battle because of split-second delays in my input, because Square chose not to stop the timer while the menu is open.

What’s most peculiar about this problem, to me, is the existence of Wait mode. In the settings menu of every game with ATB is a toggle between Active mode, in which time continues to pass no matter how deep you go into the menus during battle, and Wait mode, in which time pauses when you are browsing the lists of spells or items, as a concession to people who might not enjoy the gaming equivalent of looking up names in a phone book on a time limit. The international release of FF4 – notoriously rebalanced, simplified, and called “Final Fantasy 2” so as not to overwhelm Western players who missed the Japan-only sequels that expanded on the systems of the original – actually doesn’t offer a choice: the game is in Wait mode, period. U.S. gamers first got a taste of Active mode in FF6 (sigh, “3”).

In FF7, Wait mode expanded to include target selection for basic attacks, but time continues to pass on the primary Attack-Magic-Item menu and so requires at least one quick-trigger button press to avoid wasting time; meanwhile what had been called Wait mode in previous FFs was rebranded “Recommended” and became the default option. So Squaresoft wanted to give players some control over how much pressure there was to navigate menus quickly, and even considered that to be the default experience players should have, but always stopped short of including an option to remove the pressure entirely. Why?

FF7 took one more step towards eliminating twitch from ATB, but stopped short.

FF7 took one more step towards eliminating twitch from ATB, but stopped short.

Many fans of the series say that they don’t play Active Time Battle games in Wait mode because it makes combat “too easy” – and they could well be right. The roughest Final Fantasy is a tea party compared to the brutality of the friendliest Atlus RPG. Difficulty is subjective, of course, so games have different settings to let players choose their level of challenge; it’s not much of a stretch to call Active mode Final Fantasy’s “veteran” setting for players who would breeze through the game otherwise. Square’s decision to remove Active mode from the “Easy Type” international version of FF4 is certainly suggestive.

But the history of game design is littered with cheap, lazy, artificial solutions to making a game more challenging. I’m reminded of how the survival horror genre clung to unintuitive tank controls long after consoles had overcome the technical limitations that necessitated them; the die-hards who defend tank controls argue that making it harder to run away from enemies adds to the tension. Yes, having to input a Killer Instinct combo to cast Fire-3 before my mage gets his face ripped off increases the difficulty. Taping my fingers together or putting on a blindfold would also increase the difficulty.

Wouldn’t it be more interesting if the game became harder because I couldn’t bring 99 of every healing item into the final boss fight? Call me crazy, but how about tuning the numbers that comprise your game – you know, balancing it – until it’s as hard as it’s supposed to be? Etrian Odyssey can afford to give me all the time I want to make a decision, because the margin for error is tiny even in a random battle against enemies at your level. ATB is a crutch, substituting twitch gameplay for strategic depth.

Square's latest take on ATB is just as fast-paced, but de-emphasizes convoluted menu input.

FF13’s revision of ATB is just as fast-paced, but de-emphasizes the speed of your menu input.

Although my taste in RPGs definitely runs toward pure turn-based games where I can consider options carefully, I don’t mean to suggest that decision making on the clock has no place in RPGs. Final Fantasy 13 represents a substantial improvement over ATB in the compromise between menus and a fast pace of action. Crucially, the player can input commands while the character’s action meter fills, so even though time passes in the menu, that time is not necessarily wasted – it may even be to your advantage to delay the execution of your queued commands to synergize with your AI-controlled teammates. By contrast, only in rare boss battles is it ever strategically sound to spend more than the minimum possible time in menus in FF4.

To be fair to Final Fantasy, the combat system fundamentally works, despite this one thing that makes me hate it so much. Now could someone explain to me why KINGDOM HEARTS, a fully real-time action RPG, makes you pick attacks from a goddamn menu?

Welcome Back, Blogger


I have needed a new website for a while. Here be it! Though I’m proud that my homemade blog/RSS solution on Gigaville actually held together for 10 years, it should have been taken out back and shot long before now.

The new site serves a few purposes:

  • Hosting this blog, a real blog with WordPress and everything, where I intend to infrequently post some long-winded analyses of JRPG combat systems
  • Moving my old projects to a sensible archival location (see sidebar), with prominent commentary to let new viewers know I am adequately embarrassed by how they’ve aged
  • Motivating me to produce new stuff, e.g. the aforementioned long-winded JRPG combat system analyses
  • “Consolidating my personal brand” because that’s what “Millenials” do, “apparently”

Gigaville.com will redirect here for another year or so. I suggest switching over to the new RSS feed, though.

Also, if you are responsible for one of the bazillion links to individual Last Days of FOXHOUND comics on TVTropes, you may want to update those or they will eventually break – and in 5000 years, when the TVTropes database is the only source of information on 21st century culture, no one will know about my contributions to society.

My first real post, in which I attempt to convert my seething hatred for Active Time Battle into a halfway professional critique, is forthcoming.