How Monaco Fails Its Players

I picked up (against the better judgment of my bank account) several games during the last Steam Sale. Three of those were stealth titles of recent publication - Monaco, Gunpoint, and Mark of the Ninja (the last a gift from my friend Pat). And I bought them despite the fact that I have very real problems with the stealth genre.

Monaco has gotten the most play so far. The Girlfriend and I are moving through it, and the emotions it evokes have been a little roller-coaster-y. When it's working, it's fantastic - the art design is incredible, the character abilities well-balanced, and the challenge often invigorating. Until we hit the wrong mission, anyway. Then the alarms go off, the guards start shooting, and soon we're dead.

Now, Monaco, as far as stealth games go, is quite forgiving. You're given plenty of information about what the guards are thinking and seeing, and they have pleasantly elastic memories - if your character can stay out of sight for long enough, you'll be forgotten, and the level will go back to an un-alerted status. And in co-op, a dead player can be revived by his or her co-conspirator, making survival much easier. And yet...

There is no sensation in gaming more infuriating than loss of progress. One of the reasons we play games is that it gives us a pleasant, if artificial, sensation of achievement. We are doing things, growing stronger, passing through levels. And when those achievements are stripped from us by our own mistakes or failures, the feeling is crushing. Feeling crushed isn't why I play games.

To give a Monaco-specific example: it is a significant task to grab every single piece of loot on a floor of Monaco. It involves stealth, hacking, combat, all of your characters' skills. And if, after achieving that task, you die on another floor of the same level, it's all gone. That progress is deleted as though you'd done nothing for the last half hour of your life.

Worse, you're forced to repeat the areas you've already completed. Novelty is another important aspect of games, and by punishing players by stripping them of progress when they die, you're also stealing novelty from them. I want to experience new challenges, not be forced to re-complete ones I've already done because I died on an unrelated part of the level (this is a problem I had with Hotline Miami, as well). If a challenge kills me, I want to try that challenge again. Not spend twenty minutes doing the things that lead up to the challenge.

Am I just whining and asking to have my hand held? I don't think so. I still want hard, nasty challenges in my games, but I want the challenges to be their own reward. I don't need the stakes of knowing that failure means losing half an hour of progress in order to take the challenge seriously.

For an example of someone doing it right, look to Gunpoint. One of Tom Francis's stated design goals with the game was to never make players feel like their time had been wasted. As such, Gunpoint saves every few seconds, and after a failure, you can simply rewind a few clicks to try again. You still have to solve puzzles and execute tricky maneuvers, but there's no sense that you're being punished by the game. If you're seen, you die pretty much instantly, and then the game immediately reloads an old save; it's exactly what players have been doing when being spotted in stealth games for years, it's just automatic and far less painful.

I love Monaco. But it has made me feel legitimately upset because of how it treats its failure state, and that's not a great way to feel while playing a game. There are better ways.

Driving Me Psycho(nauts): One of the Worst Aspects of One of the Best Games

Been re-playing Psychonauts, on the theory that its hilarious dialogue and gorgeous art would make it a good game for The Girlfriend to watch me play (she enjoys it, I swear). And I was right! The only problem (and it's the thing that always trips me up when re-playing Double Fine's best game) is the tedium of some of the collection aspects of the game.

Now, to be fair, Psychonauts gets half of its collection sidequest stuff right. The game is divided between the 'real' world and the mental world, and in the real world, collection works like this: Collectible items (cards, challenge markers, scavenger hunt items, and brains) are brightly colored and mobile, making them pop against the game's backgrounds. You're given a counter in each area, telling you how many many of each collectible are left. And, most importantly, each collectible has a reasonable amount of worth. Every collectible is worth either 1, 1/2, or 1/9th of one of 101 Psi-Ranks (the game's leveling-up aspect, which is completely based on collection), and the challenges associated with each collectible are proportional to the value. A Psi-card (1/9th of a rank) is easy to grab, while a Psi Challenge Marker (1 whole rank) will involve some serious searching and acrobatics.

 

In the mental world, however, things are different. Here, the primary collectible is the 'figment' - rogue expressions of the subconscious of the mind you're currently in. Here's the problem with figments: Each level has roughly a hundred. They move. They're 2D in a 3D world. And they're transparent. Admittedly, the transparent neon look of the figments is cool, but it also makes quickly scanning a level for them an utter chore. And their sheer volume means that, even though you're given the number of figments in an area, the act of finding that last rogue semi-invisible, moving, random shape can be a chore of hours.

 

I'm not against collection sidequests at all. In a platforming game, they make much more sense than combat (never a 3D platformer's strong suit) as a metric of success. Collectibles can encourage exploration and tricky jumping, testing the core mechanics of the game. It's only when the collection becomes an exercise in frustration that I protest.

Compare Psychonauts to Mario 64. Figments are worth a variable amount of a percentage of a Psi-Rank (usually proportional to how well they're hidden), and discovering them all unlocks a hidden ending cutscene. Which is another way of saying 'There's part of the game's story that you don't get to see if you don't get every figment' (barring Youtube, of course). In Mario, the basic collectible unit is the coin. Most levels have a total of between 110 and 140 coins, and the player is rewarded with a star for collecting 100 in a single run. After that 100, the only benefit to collecting coins is health restoration. That makes the 100-coin star challenging, but not maddeningly frustrating. It's a challenge to your collection abilities, but doesn't force unpleasant, completionist behavior on the player.

The irony is, if Psychonauts DIDN'T penalize/reward you with additional story content for collecting every figment, it would make the act of completing the task (which the game never explicitly asks you to do, but which is implicitly demanded by the ____ out of _____ figment count for every mental level) even more hollow. How many games have you played where, upon reaching 100% completion, you get, at best, a text box, and then nothing else? Once you've set a goal (by 'telling' players to get EVERY figment), there must be a reward, or the player feels cheated. (All of this presumes that gameplay like this can't be its own reward, which is easy to say, given how stressful, nerve-wracking, and boring scouring a level for a single missing figment can be. At this point, you're essentially being bribed by the game into playing it in a particular way. Bad design, I think, especially when contrasted to the elegant way Mario 64 handles it). The game has asked you to do something that most players would consider unpleasant, so it has to use the carrot of unlocked content to guide you forward.

How to fix it? Glad you asked! First, there's the Mario solution: only a subset of collectibles are needed. This is probably the most pleasant way to go around this, because you can tune and playtest to find the right percentage to avoid needless player frustration. You could also go the "collectible finder" route - Saints Row 3 does this, and it works fairly well. This can go too far, of course - Far Cry 3 is overly insistent about notifying you of where collectibles are, turning the game from one of free will into one of watching the dot that represents your character as it moves across the mini-map, creeping toward a collectible icon. It's better than leaving players to waste hours in the unpleasant act of searching, but playing as a dot making his rounds on a mini-map just isn't as fun as playing as a character exploring their environment. It takes exploration, which is ostensibly what all this collecting nonsense is supposed to be in service of, and makes it safe, boring, and rote.

Linkpost: Timmy, Johnny, and Spike

No time for a full post today (moving/painting), so, since conversation on Facebook about yesterday's post got pretty detailed, I thought I'd just leave a link to this, one of the most interesting articles I've ever read about game design. It's from Wizards of the Coast, designers of Magic: The Gathering, and it discusses the three player motivations they design cards to appease. Yesterday, I revealed myself as more of a Spike than I really thought I was (although I imagine my friends weren't surprised). Although, I'd like to think there's a healthy streak of Johnny in me, as well. Which psychograph do you fall into? Leave a comment to that effect, if you like.

Timmy, Johnny, and Spike Revisited.

Why Winning Matters

My friend Matt (who blogs over at Killer Tofu) pointed me to a Kotaku piece by Quentin Smith titled "Video Games' Obsession With Winning is Killing Them". It's an interesting read, with the central thesis that video games like Call of Duty, with their focus on every interaction having a clear winner and a clear loser, create more wounded feelings and sensations of failure than necessary. He points to modern boardgames as a model that video games could learn from on this score, with their focus on non-direct competition and creating experiences more than winners.

Which implies that he and I have been playing board games very differently.

Smith cites Twilight Imperium as a game that prioritizes the creating of experiences over winning. In TI, there is a vast array of available actions to take, and the brutality of combat ensures that diplomacy is generally a first option. This encourages talking around the table and gives players breathing room to try different parts of the game world. You can focus on researching interesting things, build Death Stars, role play and sway votes in the Galactic Council. And while you're doing all that, I'll be busy accruing Victory Points so that, once the game is over, I'll be the winner. And one of TI's flaws is that the accumulation of VPs often stands in contrast to doing fun things, which means you can't have as much fun playing to win as you could simply screwing around.

 

Maybe this need to win speaks to something flawed in my character. Maybe I've been indoctrinated by video games to believe that winning matters, and I'm robbing myself of joy. But here's my question: If winning is so unimportant to modern board games, why is it there at all? Why does Twilight Imperium even have a Victory Point system if winning is such a secondary concern?

Because 'winning' is the goad that ensures good play. When we all sit down around a table, we enter the so-called 'magic circle', in which we all agree that the outcome of a fundamentally meaningless activity like playing a game matters. Each player strives, to the best of their ability, to follow the rules of the game to ensure the best outcome for themselves. I bet you've played board games before with someone who doesn't care about winning, or who doesn't fully understand the rules that the game operates under. It's horrible, right? The table basically has a hole, sucking all of the fun down it as the magic circle breaks whenever it's that player's turn. They don't care about winning, so they don't make interesting moves; they act randomly, or to suit some personal whim. They're probably having fun, but it's fun at the expense of the table, and it's toxic to good gaming.

This is often the consequence of improperly defined or easily ignored victory systems. A player who feels like they can't win, either because the game mechanics make it too easy for a savvy player to pull ahead, or because it's too easy to lose track of how winning is achieved, will begin to strike out randomly, essentially attempting to pull the game down around them in the interest of their own enjoyment. This problem plagues Twilight Imperium, with its Victory Point system often lost in the haze of all the available options. A player will see another player pull ahead, and, feeling the game's stated goal pull out of their reach, will start acting only to amuse themselves.

Games are more fun when people play to the best of their abilities. People play their best when there are stakes. The agreement that winning 'matters' and is achievable by all players is the best way to create those stakes. Ill-defined or unimportant win conditions in board games aren't a blessing or a boon that video games need to adapt; they're a curse that needs to be avoided.

(I will say: Smith's last point, about the competitive party game Bang, is dead-on. A party game that relies largely on randomness is a great palate cleanser after a long, winning-focused strategy game like Twilight Imperium, and a quick game where winning is more a matter of luck than skill can be a great game to heal wounded feelings. That being said, for games where skill IS paramount, winning must and should be the player's goal).

Failure to (Stop) Launching - How Launcher Games Hook You

I play a lot of Flash games (especially on Kongregate), for two basic reasons. One, they're free. And two, their low budgets and quick development time mean you often encounter innovative ideas that wouldn't function in a big studio release, or even as a dedicated indie game.

At least, that's what I tell myself, as I load up another damn 'Launcher' game and start sending my (pirate/buffalo/monkey/rocket/penguin/nerd/whatever pick one as applicable) hurtling through landscapes full of stuff designed to either impede or increase its progress upward or outward. So I find myself wondering: What, exactly, is so compulsive about this genre of games?

The first Launcher game I ever played was the inexplicably Japanese NANACA(Cross)CRASH!! a weird, Flash-based spin-off of a Japanese game I've only barely ever heard of (I said Japanese twice in that last sentence, because this game is very, very Japanese). The premise is simple: You're a dude. A girl hits you with her bike. AWAY YOU GO! There are people on the ground, some boost you, some slow you down. If you hit them in certain orders, weird special effects (none well explained) occur, usually to your benefit.

Back in high school, this game was VERY big in my social circle on AOL Instant Messenger, with people constantly updating their profiles (does anyone else remember how amazingly important it was to have a properly curated AIM profile, with only the best hideously colored backgrounds, the deepest, most meaningful terrible song lyrics? Just me? Right, I'm old) with the highest scores to show they were the best NANACA(Cross)CRASH!!ers (or, as we usually just called it, Weird Bike Game). So let's unpack what about it is so compelling, and see what we can extrapolate to the 'Launcher' genre as a whole.

First, the game's playtimes are very short - each session takes about a minute, unless you do very well, or very poorly. In each of these sessions, you get a wide array of emotional moments - the giddy speed of the initial launch, the pleasures or disappointments of near-misses or barely-grabbed boosts, the thrill as you catch a last-minute reprieve, the slow decline (or immediate abrupt stop) of lost momentum and the end of the session. It's very much a roller coaster (except, in this case, you can immediately get back on).

Second, the game sits in the sweet spot between luck and player skill. The initial launch is entirely dependent on the player's reflexes, but after that, control is extremely limited. You get 3 "Boosts", where you can knock your airborne character up and give him a little speed, and a slowly recharging down-kick that imparts speed and lets you push him towards the characters on the ground. That's it. The rest of the game is watching your character float along, waiting for the ideal moments to deploy your limited controls. The order of characters on the ground is random (although the player is given a small indicator so that they know which boosts or detriments are coming up), so it's very easy for a lucky player to hit several boosts in a row, massively increasing speed and distance traveled, and for an unlucky player to hit an instant stop. However, the player control means that, when you use one of your limited interventions and it works, the feeling is intoxicating. Essentially, we have a situation where happy outcomes feel like a result of player skill, while negative results feel like bad luck. That's a recipe for players deriving pride from their successes without being too discouraged by their losses, and feeding that positive feeling is a great way to keep people playing.

One thing NANACA(Cross)CRASH!! doesn't have that its descendents almost always do, though, is any kind of progression structure. Every launch starts with the same probabilities of success - you can't upgrade the bike or pay an in-game currency to make the beneficial characters more prevalent. This makes NCC!! ideal as a score game, one where you can post a crowing high score to your friends with no caveats about whichever upgrades you've purchased, or how long you had to play the game to unlock the SUPER BIKE that got you your score - there's only 'skill' (which, see point 2 above). However, it also means that, once you've gotten a score you like, there's no sense of investment in the game, no feeling of sunk costs to pull you back in for another play.

In contrast, let's look at a game I've talked about on here before, Burrito Bison Revenge. BBR (produced by Juicy Beast) shares many design elements with NNC!!, - timing based launching, special units that boost or reduce speed, a limited control scheme prioritizing carefully timed interventions), but adds a few things that really let the game put its hooks in. (An easy way to know when a game 'has' you, and one I'll refer to repeatedly here, is when you have the thought "I'll just play until _____," where _____ is some measure of in-game progression. Once you've started bargaining with yourself for more playtime, you know you're in trouble). The first is an upgrade system, where in-game currency can be spent to enhance various features of your flight (stronger launches, less speed lost from mistakes, more control interventions, and several others). This does two important things: One, it gives the player a small-scale goal to play toward - "I'll stop once I've got enough money for that top speed upgrade" (but of course, you're going to want to TRY OUT the new upgrade, aren't you? And now you're so close to that bounciness upgrade...)- and two, it ensures that the player, regardless of skill level, will do better and better as play continues. Even if the player never gets better at using the rocket slams (the game's limited control scheme), by nature of upgrading their character, they'll go, generally, further and further with every launch. Thus, you avoid the discouraging scenario often found in NANACA, where a great launch can be followed by an abysmally cut-short one; instead, you have a constant, pleasing feeling of (fake) improvement.

 

Juicy Beast also introduced a multi-tiered progression scheme into the game to keep players hooked. The game world is made up of four regions, each separated by a large door which requires a sufficient amount of speed and strength (a stat only affected by upgrading) to pass through. Once the player can break through the fourth door, the game is 'over', and the push to get through each, more sturdy door, makes up the highest level of the game's progression. The doors are a milestone to aim for, a large-scale 'I'll play until I get through the next one' measurement. However, the game also has a smaller scale sense of progression, through in-game missions. These (as I described back during my Achievements posts), are small, variably difficult goals that are presented, three at a time. Once one (say, do two perfect launches in a row, or earn x amount of money in a single launch, or any permutation of the game's mechanics) of these is completed, another one takes its place. Thus, even when the gameplay distance between breaking through one door and the next is too big for the player to comfortably commit themselves to, they have these mini goals to focus on as they build up upgrades (which, as pointed out above, make it easier and easier to progress to and break through the next door). The missions act as connective tissue for the other goals in the game, a way to keep players committed even during lulls.

 

So, there you have it: the modern 'launcher' game. The initially compelling aspects (short session times with a wide emotional range combined with a healthy amount of randomness and limited, but meaningful control) have been fine-tuned and married to progression mechanics designed to keep the player going for 'one more launch' (inevitably, more like 20). In a way, they're quite cynical in how effectively they suck down player time, but it's hard to stay mad at them, because it's hard to do anything right now except PLAY LAUNCHER GAMES (Will quickly alt-tabs to the three different browsers he has with launcher games waiting).

A short list of good launchers:

1) NANACA(Cross)CRASH - Weird, and addictive. Feel free to post your best score in the comments here; mine's 10,145.

2) Burrito Bison Revenge - The king of launchers - gorgeous graphics, varied controls, and an expertly machined sense of progression.

3) Learn to Fly 2 - This one offers a great deal more control than the first two, and also has an EXTREMELY CHARMING PENGUIN

4) Into Space 2 - the best of the Vertical Launcher genre, where your goal is to go up, not out. A really, really detailed upgrade system.

There, that should be enough to waste a few of your days.

Why I Love Jetpack Joyride But Will Never Pay For It

 

So I'm playing Jetpack Joyride on The Girlfriend's shiny new Windows 8 computer (Windows 8 - Like a cell phone, but it won't fit in your pocket!), and I get to thinking about how much the whole thing cost to make. Rocking soundtrack, gorgeous art, programming... My Google-Fu is too weak to find it the exact development cost, but it clearly wasn't cheap. And I've been playing it, now, ad-free, for about 2 hours. And the guilt starts my eye crawling down to the 'Purchase coins" button... And I balk.

The advent of free-to-play has done a lot of weird things to the gaming industry. Back in the day, it was simple - you gave some money, you got some game. Now, here are the models I can pull off the top of my head: the game dev gets paid in attention and future interest in products, the game dev gets paid in ad revenue per play, the game dev gets paid for the second half or last two-third of their game, the game dev gets paid before they even start development through Kickstarter, the game dev gets paid through in-app purchases.

That last one (the one that Jetpack Joyride uses) has a lot of latitude in how it can be applied. Cosmetics, extra in-game currency, bonus levels... They all boil down to one question: Can you 'finish' the game without ever making a purchase?

Of course, this means we have to define how a game is 'finished,' which is increasingly nebulous as mobile gaming's focus on pick-up-and-play has created a resurgence in arcade-esque endless games. Still, most games have some sense of progression, usually through the purchase of upgrades or the completion of achievements. Jetpack Joyride (where everything can be purchased with the in-game currency, which collects slowly, but not so slowly as to make buying things impossible) can be completed without making a single purchase. Which is good, I guess, because it avoids the upsetting bait-and-switch thing that happens when a 'free' game starts piling on inconvenience after inconvenience - which can, of course, be alleviated with a cash purchase, just as you're getting into the meat of the game.

Here's the problem, though: The major gameplay of Jetpack Joyride (which you should totally play if you haven't, it's an amazingly well crafted little thing) is collecting coins, and the major part of the game's meta-structure (that is, the system of persistent upgrades that gives you a feeling of progress) is deciding how to spend those coins. And the only way you can give the game-makers money? Buying coins.

If you think about it for a second, you'll see the paradox. If I want to compensate the game devs (and I do, because they've given me a lot of fun), I have to cheat myself out of some gameplay. I have to make some of my upgrade choices meaningless by filling my coffers with purchased lucre. It feels like cheating, and that's why I balked from, essentially, giving Halfbrick Studios a tip for the enjoyment their game gave me.

Basically, what I'm arguing here is that the free-to-play elements of a game should be disentangled from its progression structure (this is something that goes back to the Dead Space 3 "Pay for Crafting Resources" thing from last year - if earning resources is fun, then don't give me the option to pay to take away the fun, and if earning resources ISN'T fun - why is it in your game?). Include them, certainly, but in their own separate areas - cosmetic items are ideal for this. Hell, I'm even fine with selling levels, as long as you don't go crazy with it. But when games like Jetpack Joyride merge their for-pay elements with the in-game activity of gaining money, they're going to end up disincentivizing either gameplay, or paying. And a game this fun doesn't deserve either.

Short Fiction: Rogue-Ish

(I wrote this on a whim to show that I could write something fantasy-ish. It's short, but... I don't know, I kind of like it).

The thing about rogues is that they are, at the end of the day, profoundly roguish. And sure, that sounds great in the tavern. Everyone’s heard stories about the party that would have been chewed apart by scorpions at the bottom of a pitfall, if not for the tireless efforts of an auric-hearted rogue. Or of steel-eyed queens charmed into leniency (and out of lingerie) by a talented bard.

Sure, they were, technically, thieves. But, the stories always emphasized, FUN thieves. And when the chips were down, at the end of every story, wasn’t it always the seemingly traitorous rogue who came back to save you? A glint in his eye, aiming a crossbow bolt at your chest, but no! He was only shooting the orc behind you. He might steal your gold pouch, but he’d never stab you in the back to do it.

Which presents the question, Thak of Grimmeld, Warrior Lord of the Far Steppes, mused to himself as he lay bleeding to death on a filthy stone floor with a dagger wound in the small of his back: Who’s making up these stories, anyway?

 ------------------------------------------------------------------------------------------------

Glenn, as he had called himself, was the most roguish of rogues, if his own stories were to be believed. Despite his simple name and unassuming appearance, he claimed to have stolen more gold, platinum, and jewels than the combined efforts of years of work from the Steppe Horde raiders who had trained Thak in his youth. He claimed to be the assassin of The Unknown Emperor, which, Thak thought now, he probably should have been more suspicious about verifying. And Glenn had a reputation, one that, Thak’s wife, the ‘Virgin’ Sorceress Aurora, assured him, made him the ideal, trustworthy companion on their next venture: He had betrayed every single person he had ever worked for.

It had taken Thak some time to work through the logic of this benefit. The flask of mead he had consumed (paid for, of course, by Glenn [with money, of course, taken from Thak’s own purse]) had made the efforts doubly difficult, but eventually he had grasped Aurora’s points (while failing to notice that Glenn was, when Thak wasn’t looking, doing some grasping of his own):
 
    1) Everyone knew that in stories, the least trustworthy person could always be trusted. After all, the story wouldn’t be very interesting if it was just ‘The bloodthirsty psychopath turned out to be the murderer’, right?

    2) Who could be less trustworthy (and thus MORE trustworthy, by this new logic), than someone who had betrayed everyone he had ever worked for?

    3) This is damn good mead, isn’t it?

Persuaded by this iron-clad argument, Thak had found himself, the next morning, bleary-eyed and barely able to hold his fabled double axes, Krew and Krag, following behind Glenn and Aurora in pursuit of the fabled treasure of Mak Goughin, which Glenn had conveniently known the location of. Through his raging headache, Thak couldn’t help notice that his ‘virginal’ wife and the youngish rogue were walking closer than comfort would suggest was feasible. But he held his thick, fuzzy-feeling tongue, not wanting to give Aurora ‘paranoid jealousy’ as a weapon in their increasingly constant arguments. And so, as they ventured into the filthily-floored dungeon, he simply watched.

Not closely enough, it turned out. And so, he lay dying on the incredibly poorly kept stone floor of the dungeon, bleeding swiftly from the wounds in his back, one placed by Glenn with a merry laugh in his throat, and the other by Aurora, cold as a winter’s sunrise. Was the flickering in his vision the last of his life ebbing away? Or the guttering of the black, oily torches that illuminated the chamber? Or, possibly, the Lichstone he had stolen, years ago, from the Wizard Jandar, finally fulfilling its dark purpose and reanimating him as a murderous revenant bound to avenge his own demise?

It was that last one, happily. And as the last of his mortal existence faded away, Thak smiled at the thought that Glenn really, really should have done more research before seducing his wife and murdering him. But then, that’s the thing about rogues.

They’re roguish. That’s not the same as smart.

How Sadism and the Wii U Saved Gaming (Potentially)

Confession: My favorite Mario game is New Super Mario Bros. for the Wii. It's not the most polished or the most inventive, sure. It doesn't have the most beloved power-ups or the best level design. But it has one thing that has given me more joy than any other feature in a Mario game: multiplayer.

Not just "trade off when someone dies and every once in a while battle in a coin-collecting minigame" multiplayer. Not "Player 2 collects star bits" multiplayer. Not even "there's a separate battle mode and you can run around fighting in it" multiplayer. Real, honest-to-goodness, four-people-struggling-to-make-it-through-every-level multiplayer. And it was glorious.

 

And the best thing about it was that you could play it more than one way. Sure, you could be nice, with the more experienced players carrying the weaker ones through difficult sections, politely sharing power-ups, using teamwork to get the hard-to-reach star coins. And that's fine! Do that, you'll have a good time.

I'll be busy playing MY way - a brutal race to the finish of every level. Dirty tricks, hoarding of power-ups, intentionally throwing the others into bottomless pits. Nothing's off-limits, as long as one person's alive at the end to grab the flag. And the game's mechanics are perfectly built for this (power-up stealing, picking people up and throwing them, the way the screen scrolls), but without any explicit instructions to do so - the perfect recipe for building your own game mode.

 

(An aside: in 2006, my friend Nathan had the perfect ingredients for fun - an original XBox, a copy of Crimson Skies, and a projector. Dog-fighting on a big screen was, even with the muddy resolutions, suitably epic, but we eventually tired of the basic multiplayer modes, with their focus on combat over flying. And so we invented a new mode - a variation on the game's King of the Hill (where the player who holds a single flag on the map for the longest wins) with one key distinction: the player with the flag wasn't allowed to shoot. Instantly, what had been a fun-but-generic bout of dogfighting became a tense, exciting game of cat-and-mouse. Especially on the game's 'Chicago' map, with its just-wide-enough streets between towering skyscrapers, it was the perfect, player-created game of chase and be-chased.)

 

As previously mentioned, I spent some time with the Wii U last week. Along with NintendoLand, the other game I played (along with The Girlfriend, my friend John, and his wife Desiree) was New Super Mario Bros. U. Because I was (and remain) fascinated by the GamePad, I asked if I could play the game in what's called 'Boost Mode' - where 1-4 players play as normal with Wiimotes while the player with the GamePad uses the stylus to place blocks, uncover secrets, and stun enemies, ostensibly to help the players on their way.

Ostensibly.

Instead, it quickly became clear to my fellow players that I was using Boost for a different purpose - to turn every level into a maze of suddenly-appearing platforms, erratically moving enemies, and sudden death traps. I would do my best to block jumps, move power-ups out of the way, and basically take on the role of a cruel dungeon master. (I cleared this with my friends, by the way. Mostly. They appreciated the challenge! I choose to believe). The mechanics for Boost Mode are less clearly designed for abuse than the ones in the Wii game, but opportunities are still there. The platforms you can create with a touch of the stylus are the most obvious method of interference, but there are other, more subtle ways to play vengeful God. 3-Up blocks can be revealed by the GamePad player, acting as perfect bait to force players into traps. And a few levels have things like giant gearwheels that respond to the stylus's touch, allowing a savvy bastard to trap and crush his... opponents? Is opponents the right word? in between their gears.

 

As gamers, we've always dealt with the fact that no challenge from a video game can be as organic (and nasty) as one presented by a human. Part of the pleasure of tabletop gaming is the sense that you're playing 'against' an opponent who can tailor your challenges to your capabilities. The Wii U, by giving one out of the five players (what amounts to) a mouse and monitor set-up, has created an ideal environment for that player to mastermind gameworlds to provide more interesting challenges. (The free Rayman Legends Challenges App takes this even further than NSMBU, with one player having control of most of the features of the platforming world, moving them to help the platforming player traverse). This is fun in a platformer, but what if we pushed it further, applied it to genres like RPGs or puzzle games? Mazes drawn on-the-fly with the stylus. Enemy line-ups drafted from a pool of foes. All those tasks we've relegated to AIs and pattern-generation, back in the hands of a human because we finally have an intuitive way for them to control it.

I know, I know. People have been crowing about the 'potential' of the Wii U since it was first announced. But I feel like we're on the periphery of something great, a chance for gamers to interact like never before. New Super Mario Bros. U and Rayman Legends Challenges App are at the cliff's edge of embracing what the Wii U can really do when you trust your players to help you make amazing experiences, and I still have my fingers crossed that, sometime soon, someone is going to take the plunge.

Would I Lie to (Wii) U?

Finally had a chance to play with the Wii U when The Girlfriend and I stopped at my friend John's house on the drive across the country.

The thing that most appealed to me about the console (we mostly played NintendoLand, although John and I did fiddle for a bit with the Rayman Challenge App, which I might write more about later) was the way the asymmetric gameplay provided by the GamePad allowed for one of my favorite flavors of gaming: Deception.

I like lying. Telling a lie well, and convincing the people around you to act on its information, is one of the great social challenges that exist. Unfortunately, in most situations indulging in that challenge makes you an asshole or a sociopath, and the negative consequences almost always outweigh the benefits (insert witty comment about politics, business, and all other profitable lines of human behavior here.) Which is why some of the games I love most are ones in which lying is a codified part of the experience, where such normally vilified behaviors are encouraged and rewarded.

This can either be lying through game mechanics (the first example that comes to mind being bluffing in poker, although my personal favorite is Letters from Whitechapel, where one player is invisible and must cross a large game map while being hunted by trail-sniffing detectives), or through outright social manipulation (Werewolf and its cousin Epic Mafia are probably the best examples of this, where successful manipulation of the other players is a key strategy for both sides). It's fun to lie, because lying has stakes, and you're tricking the human mind, one of the greatest bullshit-detecting computers in existence. When you pull it off, the rush of adrenaline is incredible (I may have just outed myself as a psychopath).

Lying in video games has historically been much more difficult. Early games were played with two players on the same screen, meaning that secrets were impossible to keep, even if you 'totally for sure pinkie-swear' not to look. The only example I can think of with successful same-screen secret-keeping are the play-choosing modes in sports games, where the cursor can be made invisible so that the other player can't see your choices. Playing via network made this much easier, of course, leading to all sorts of games where controlling access to the other player's information was a key part of strategy (Starcraft and other RTS games that use fog-of-war come to mind). But there's something empty about beating someone over a network, about lying to them from across a hundred miles instead of being in the same room. It lacks... intimacy.

That's what made the Wii U so fascinating for me. Playing Luigi's Mansion in NintendoLand as the ghost (who is invisible on the main screen but visible on the GamePad), I had that thrill of sneaking up on someone, of misdirecting them and then suddenly striking. Trash talk (and communication between the other players) becomes a key strategy of manipulation and coordination. When playing as one of the non-ghost players, I had the thrill of anticipating the movements of an invisible foe and taking him down just as he was prepared to strike.

I've played a lot of games where people sitting in the same room never broke eye contact with the screen, never talked to each other. That's not possible with the games we played in NintendoLand; communication is key. The WiiU has the potential to change the way we game together, adding elements of social manipulation and teamwork to same-room gaming. If the software is there to support it (eh....), this could be a new dawn of the fine art of lying through games.

Walking (Dead) With My Girlfriend Finale: Keep that hair short, girl

The Girlfriend and I have now finished the last three episodes of Telltale's The Walking Dead (and then moved across the country to Washington, hence, the delay of this article), and I thought I'd get a few more thoughts down on the game. This article will contain a few spoilers, although I'll try to keep it as clean as possible. Still, if you don't want at least some elements of the game's story spoiled, read no further.

The thing that became clear, once The Walking Dead was over, was that it's not Lee's story at all. Lee is the protagonist, sure, and the games stick very closely to his perspective. But the actual story, the character who the player influences in the most interesting ways on their journey, is Clementine's.

There was a moment, after the games were over, when Shanna chastised me for being too honest with Clem. Indeed, anytime the option to sugarcoat or euphemise or lie to spare her young feelings was presented, I would argue strongly against it. "She's just a little girl," Shanna argued. "She deserves the truth," I would reply. We were having a sincere discussion about the best way to parent a child who was not a child, just a deft creation of 3D modelling and writing and Melissa Hutchison's excellent voice work. But Clementine felt real to us, and her opinions mattered.

When we chose to do the right thing, it was as often as not because we were worried what Clementine would think (and the game is brutal about inserting her into moments when the urge to do wrong is strongest). Her disdain is the worst punishment the game can deliver, because, thanks to her childish naivete, she never hesitates in opting for the "good," "righteous" choice. Sure, you can rationalize your actions (and the game's structure, which requires that Clem remain devoted to Lee, ensures that she'll at least partially accept your excuse), but the punishment for doing so is the sense that you've made a permanent influence on Clem's impressionable mind.

Telltale aren't the first people to realize that a child's judgment is an excellent way to make a player give considerable weight to their actions. It's easy, especially for jaded gamers, to treat fictional worlds like consequence-less playgrounds where the id can run free. This is all well and good if that's what you designed your gameworld to be, but it can absolutely wreck an attempt at a serious tone. By placing the watchful eyes of an impressionable being on the player, learning from their actions, it's possible to give normally sociopathic players pause.

The difference here, and the reason Clementine is so compelling, I think, lays in the fact that most games that employ this mechanic (Bioshock 2, The Witcher, and Dishonored come to mind) tie your influence on the child to some set outcome for the game. You are told, explicitly, that your choices had a concrete impact, that you've pushed this child to some specific life route. In short, you're given closure on the choices you made, assured that what you did 'mattered,' because in a video game you expect to be given a clear metric for the choices you made. But The Walking Dead's conclusion derives its poignancy and meaning from the fact that we are utterly denied our closure, our ability to see how we've shaped the future.

When the game is over, your ability to influence Clementine is gone. There's no jump cut to her as a heroic messiah or a blood-thirsty warlord, guided by your parting words. There's only a scared little girl, still trapped in a bad situation, and the hope, a hope which exists only in the hearts of the player (or players), that the influence we had on her will be enough to keep her safe. That we taught our little girl enough to make her strong and smart and healthy. There's no guarantee that it will, that all those "Clementine will remember you said that"s will amount to anything. We just have to hope that it was enough.

I think that's called 'parenting'.

Papers, Please, and the Joys of Being Mindlessly Amoral

In my gaming experience (now stretching, ugh, 24 years), I've portrayed: floating talking skulls, item shop merchants, sentient meat, a fake hacker, God, Satan, a Japanese death pinball, and a shrunken planet with a ray gun. It's STILL weird to be playing (and enjoying) a game about being an immigration checkpoint attendant.

Papers, Please, by Lucas Pope, is still in beta, but it's already compellingly playable. Set in the same post-Soviet dystopia as Pope's previous game, The Republia Times (in which your role was editor-in-chief of the state-controlled newspaper, ordered, at gun point, to keep the people happy and docile through story selection), Papers, Please is, mechanically, very simple. Potential immigrants step up to your window, give you their documents, and wait as you peruse them for forgeries or mistakes. If you don't find any, you take your giant stamper machine and CLUNK "Approved" on their passports. If they do, you reject them (or interrogate them to figure out the meaning behind the discrepancies). The heart of the gameplay is a series of very basic rules that you apply to every set of documents - are the dates right? The issuing country? Do they have the right visa? It's essentially a game of pattern recognition and anal-retentive detail-noticing, and with the wrong implementation, it could be incredibly dry. But somehow, here, it isn't.

A huge part of that is the interface. Almost everything is done in-game, with no need for menus. Your desk is your interface - you shuffle the documents around on it, pull out your stamping machine, fumble through your instruction manual. If you need an added feature (say, to search or fingerprint a subject), a button pops out on the desk. Besides a few opening instructions, almost everything exists in-game, and it gives your job a pleasantly tactile feel.

Adding to that feeling is the solid CLUNK of the stamping machine. It's hard to overstate how good it feels to CLUNK a document with a big, satisfying stamp. It gives a happy little climax to every encounter - CLUNK! Denied! Go away, forger! CLUNK! Welcome to glorious free state of Arstotzka, citizen! I don't know where Pope got that sound file, but it's the game's true star. CLUNK! I am seriously considering pausing writing this to play the game some more, just so I can get my CLUNK! fix.

This lady SEEMS on the level...

That joy in tiny detail, in the pleasures of executing bureaucracy, is what makes the game intellectually fascinating. Pope has done a lot of great work in making each immigrant seem unique through a very effective randomization system (along with several scripted immigrants who build up the 'story', so far as it goes during the beta). They'll praise Arstotzka as you let them in, or complain bitterly about having to travel though your hayseed country, or tell you a sob story when you ask them why they're trying to pass some expired bullshit through YOUR checkpoint. CLUNK! Bye, lady! Because, sorry, lady, but I'm on a timer. I could take the time to resolve your issue, but every time I CLUNK! my CLUNK!er, I get $5, and every night I've got to feed my family, pay my rent, and make sure the heat is on. Every day has JUST enough time for an efficient CLUNK!er to make enough to keep their son from getting sick and leaving their wife hungry, but it's close. Too close to spend a lot of time worrying about the people you're sending away unhappy.
 
That's the genius of Pope's game - it presents situations where empathy is called for, and then makes that empathy harmful, or at least inconvenient, for the player. For every second you spend asking someone for an extra document, or double-checking their fingerprints, you could just CLUNK! them and send them away, with you, at least, none the worse for wear. The game has scripted characters who offer more tangible moral choices, but the game's most effective moral lesson is that it's remarkably easy (and, in fact, kind of quietly pleasant), to turn off your humanity in favor of efficiency. (Or even patriotism! There's something about the game's bombastic, tuba-y soundtrack that gives me a certain pride in protecting my fair Arstotzka from these filthy, dangerous immigrants!)

But I don't like her face. Clunk!

"They're just video game characters, though," I hear you say. And that's the point! Papers, Please, even in its uncompleted beta form, with most of the story segments absent, functions as a beautiful simulation of how it feels when you stop treating people like people, and just as inconvenient, artificial things to be dealt with as quickly as possible. I can't wait to see what the final version can do.

CLUNK!

CLUNK CLUNK CLUNK CLUNK

Papers, Please is currently in beta, and can be found here. Release date for the full game is sometime this summer, and it'll cost REDACTED PLEASE PLAY THIS WEIRD LITTLE MINI-GAME TO LEARN GAME PRICE, ALL GLORY TO WONDERFUL ARSTOTZKA

Walking (Dead) With My Girlfriend Part 2: Less Choices, More Dying

The Girlfriend and I finished the second episode of Telltale's The Walking Dead last night, with the general agreement that, while still quite enjoyable, it was a step down from the first one. Elaborating on why would probably involve spoilers, which I'm striving to avoid with these posts, but I think there's a few points I can make before grousing at a bit more length about a fundamental problem I have with the games.

The biggest problem we had with TWD Episode 2 was that it made the illusion of choice that powers these games more obviously illusory, ironically by offering us more ostensible freedom. The first episode, which was focused on small locations and clear-cut dangers, constrained my possible 'safe' behaviors - the ones that wouldn't obviously get us eaten by zombies - and let us pick a few choices from what was left. The second episode asks much more general questions about which paths could or would be safer, but there's clearly no escape from the general flow of the narrative. Lee (the player character) makes several decisions that neither I or Shanna would, causing several moments where we were like people watching a movie, shouting "Don't go in there!" at the idiot protagonists. The episode clearly had a story it wanted to tell, but in telling it, it robbed us of a sense of agency. There are still plenty of choices to make in Episode Two, but the biggest ones (ie. 'Should we trust these people?') are taken out of our hands, and it makes the experience feel a lot more hollow.

 

Another problem we had (Shanna, especially), was a sense of being let down by the dialogue options in a way we hadn't been with the first episode. "Starved For Help" gets kind of dualistic, as you're pulled between factions in your group, and there (unsurprisingly) came a time when we were forced to choose sides. Afterward, during a conversation reflecting on that event, EVERY conversational choice said essentially the same thing - three different ways of denouncing the character we didn't side with. I guess you could see it as the game locking us in to supporting the decision we had made, but the fact is that none of those things were what we WANTED to say in that moment. It's another one of those times when the game failed to put us in Lee's shoes, and it breaks the sense of "What WOULD I do in these situations?" that makes the games so compelling.

That's not to say we disliked Episode Two. The plot was pleasantly twisty (if a little predictable), and most of the problems fell away as a climax pushed everything back into the tense, exciting "You have three seconds to choose or everyone dies, what do you do, WHAT DO YOU DO" sensation that we loved in Episode One. We'll definitely be back for the third.

 

I do have a complaint, though, about how the games handle something that game designers have struggling with for years - player death. It's probably not surprising that, in a game about a zombie apocalypse, the main character can die. I've died once in each episode (which, I'm not gonna lie, is a little embarrassing in front of your girlfriend), and each time the only thing that really gets killed is my sense of immersion in the game.

Hypothetically, death should be the worst thing that can happen in a game. You know, because it's... death. But in The Walking Dead, the moments where Lee passed away were some of the LEAST tense, because the consequences of it were so quickly reversed. You get a "YOU DIED" screen, and then time is rewound slightly and you get another chance. If I say something wrong in a conversation, it's a mis-step I might never recover from; if my throat is torn out, it's three button presses to get back on the right track.

But what's the alternative? Death erases your save file, Steel Battalion style? Unacceptable - nobody likes to have progress stolen from them. As gamers, we've always had to deal with the fact that, in life, death is the end, but in games, it can only be a setback, because we want to keep playing. Push a player too far, rob them of too much progress, and they'll just abuse savegames (if you let them) to avoid negative consequences completely. Take that away, and most of them will quit (the ones who don't are probably weirdos who play roguelikes, ugh).

 

If there's a solution to this, it's in exploiting the player's attachment to the other characters in the game. Instead of painlessly killing Lee, maybe every time he fails to extricate himself from mortal peril, one of the other characters is hurt saving him. Not killed - that would alter the story too much for the designers to keep the narrative under control - just hurt. Make them look more tired in cutscenes. Hell, maybe even build a sense of obligation between them and the player - "You're siding with HER after I was damn near murdered saving your ass?!" - that opens up new story possibilities or emotional connections.

If you're going to insist on having failable action sequences in your adventure game (and that's a debate, I think, for another time), they need to be invested with consequences that work WITH the adventure portions, not just alongside them. Make them part of the story, not just a quickly reverted sideshow.

Walking (Dead) With My Girlfriend - Some Spoiler Free Thoughts on The Walking Dead Episode One

I picked up The Walking Dead last week during the Telltale Games Humble Sale. I've found their games problematic enough in the past (Sam & Max too clunky, Strong Bad too collectible-obsessed, although the Monkey Island games are quite good) that I'd never have paid full price for it, but I'd heard enough positive things about TWD to happily pay $4 for it. Having recently moved in with my girlfriend, who loves the TV series, I thought the episodic structure would make for great lay-in-bed-together-and-play sessions. Last night, we played through the first of the five episodes, and I thought I'd jot down some thoughts about it.

First of all, the controls have none of the problems of older Telltale games, which is to say, the character moves quickly across the screen (seriously, adventure game designers everywhere: I love that the genre is back! I grew up on Space Quest and Quest for Glory. But, and I mean this as politely as possible, THERE SHOULD BE LESS THAN A FUCKINGSECOND BETWEEN ME CLICKING AND A THING HAPPENING. ALWAYS.). The dialogue wheel is stolen from Alpha Protocol, but we'd all be a lot better off if all games stole from Alpha Protocol, so that's cool.

Actually, the interface is one of my favorite aspects of the game - when you start the episode, you're given an option of a "Normal" or a "Minimal" interface, and I strongly recommend Normal. Normal means, when you make a conversational choice, more often than not a little pop-up will appear on the screen that says something like "Ken will remember you said that." The first time that happened, my girlfriend had a wonderful little fit of paranoia about how a minor conversational lie could come back to bite us in the ass. And for a game that's 60% conversations, it's a wonderful way to give player choices impact. Dialogue can be ambiguous - did that character say he doesn't trust me because that's what he's scripted to say, or because I lied to him? When the interface itself calls you a liar, it makes every decision feel more important.

The timed conversation mechanic is great, too. Once the game establishes that conversational choices have real, meaningful stakes, the added pressure of time-sensitivity amps up the stress in pleasant ways. The best moments are crisis situations, where there's not enough time for me to consult with Shanna about which choice we should make. One of us simply barks out a command, the one with the controller puts it in, and the choice is sealed. And since the most-tightly-timed choices are the ones of most consequence (which is to say, who to save when the walkers start attacking, it creates moments that feel REAL in ways that they couldn't without that urgency. It stands in contrast to the more sedate conversations, where we both try to suggest the 'right' choices, gaming the system or trying to seem morally 'correct'. But there's no time for that when zombies are about to rip a kid to shreds, only blind, instinctual decision-making. I love it, and I love sharing those moments with her.

The game it put me in mind of, unsurprisingly, was Indigo Prophecy (Fahrenheit outside the US and on PC), the first game I can think of that put conversations under the time pressure they would have in real life and assign consequences to how you speak. Of course, the conversational choices in Indigo Prophecy are eventually revealed to be largely meaningless, and are eventually consumed by mindless QTE combat, so it might not be the best game to use as a model.

I'm still not sure how much of The Walking Dead's sense of consequence is real, yet. I've so far resisted the urge to pour over FAQs to figure out which of my decisions actually change things, mostly in deference to the fact that I'm discovering it alongside my girlfriend. The end-of-chapter "On The Next Walking Dead" bit, where almost every decision is reflected, almost made me feel more leery, though. It felt like a checklist, with the game saying, "Seeeeeeee? We remember! Really!" I'm looking forward to seeing how my choices carry over into Episode Two when we play it tonight.

The one real qualm I have with the game so far is that it cloaks the past of player character Lee Everett in ambiguity. Without going into spoilers, Lee begins the game in handcuffs after being accused of a crime, and the game is never clear about whether he really committed it. This would be okay if this was a situation where I could choose his past, like the flashback sequences in Knights of the Old Republic II, where choices made in conversation essentially 'select' which past occurred, but the game makes it fairly clear that there IS a true answer to Lee's guilt, and we just don't know it.

 

In a TV show, this kind of ambiguity is natural and can be used to ratchet up tension and suspense - the first season of Homeland is largely driven by the fact that we don't know what's going on inside one of the main character's heads, and it's riveting - but in a game, it's a flaw. You can make the argument that the gap in our knowledge of Lee represents the fact that, with society having collapsed, people's pasts don't matter. But I'm not just supposed to be watching Lee Everett, I'm supposed to be him, making the choices I think he'd make. By hiding an incredibly important aspect of his past from me, it hamstrings my ability to make those choices. It's a hole in my understanding of the character, and it leaves me feeling like I'm going to be ambushed - and not in a fun, stomp-the-zombie sort of way.

*I'm wondering if this won't turn out to be the sort of situation where Lee's eventual guilt is determined by our behavior throughout the game - a Lee played righteously will be eventually shown to be innocent, a villainous Lee guilty. I'm okay with that sort of adaptive retcon, even if it does damage the possibility of actual redemption by retroactively exonerating a 'good' Lee instead of letting his virtuous actions be a reaction to his past.

Why Etrian Odyssey III is One of the Nintendo DS's Must-Play RPGs

I've been re-playing Etrian Odyssey III: The Drowned City, lately, to the exclusion of pretty much everything else, and I thought I'd try to get down some thoughts about why I find the game so compelling.

Quick basics: Etrian 3 is a first-person dungeon crawler in the style of old school PC games like Wizardry and the early Ultimas - a brutally hard, story-lite descent into a dungeon full of traps and monsters. The genre saw a resurgence in the latter

half of the Nintendo DS's lifespan, mostly thanks to Atlus realizing that the handheld was a perfect platform for graphically light, gameplay-deep games. The Etrian series' primary 'gimmick' is that you use the DS's touchscreen to draw your own map - a throwback to the old days when players were expected to keep books of graph paper next to the computer or risk getting miserably lost.

There are several FPRPGs on the DS (besides the Etrian games, the most prominent is probably the excellent Shin Megami Tensei: Strange Journey). So why does The Drowned City command so much of my attention?

1) Pick up and put down

One reason I've been giving The Drowned City so much of my time is that it's remarkably considerate of it. Previous games in the series had a nasty tendency to stretch dungeon crawls out to unpleasant lengths in the interest of challenge - you had to keep pushing forward or you'd essentially lose your progress. But TDC's map designers sprinkle shortcuts throughout their levels which can only be accessed for the first time from their more difficult side. This leaves the initial pushes into dungeon floors pleasantly challenging, while making progress easy to resume if you're forced back to town, and bypassing a dungeon floor you've already cleared a snap. And since these shortcuts are invisible, they give a nice bonus to the diligent cartographer, since careful notation is the only way to keep track of where they are.

The Drowned City also encourages variation in the gameplay experience through the use of the Ocean Exploration metagame. For a small fee, the player can load their party on a boat and explore the oceans around the titular city of Armoroad. Diligent maritime explorers can make money, discover hidden items and equipment, and unlock a series of challenging bonus bosses. Ocean Exploration is light on combat, acting more as a puzzle game, and it works wonderfully to clear the palate after a tricky boss fight or arduous bout of exploration. The only annoyance is that progress in the metagame is staggered by progress in the main game, which can feel stifling at times. But it also makes sense - if the Ocean's primary purpose is distraction, you can't let players burn through it in a single sitting.

2) Class

The level progression system in The Drowned City is fairly simple - for every level a character gains, they gain a single point that can be invested in their class skills. The innovation comes in the variety of those skills, and the classes that they make up.

The first Etrian game's classes were vanilla - a melee fighter, a tank, a wizard, a healer, etc. Their roles in the party were clearly defined, with the only tactical ambiguity being that, while the front and back party rows each have 3 slots, the games limit you to 5 characters in a battle party, meaning there will always be gaps that need to be covered with savvy strategy.

The Drowned City keeps the five-character limit, but mixes the classes in interesting ways. The tank can be optimized to be a massive damage dealer. The primary mezzer (a role that applies status effects to enemies, vital in Etrian's combat) does so by summoning companion animals into the party's sixth slot who then fight as NPC helpers (the sixth slot is also employed by ninjas, melee fighters who can create clones of themselves. The conflict over who gets to USE the sixth slot is one of the interesting ways the game's classes interact).

The most intriguing of these hybrid classes is the Prince, a kind of passive healer/buffer. The Prince's base abilities are a set of buffs which can increase attack power, protect from status effects, etc. But the rest of his skill tree is devoted to a sort of powerful but slow passive healing. Characters he buffs can regain health. If he removes a debuff placed on a character by an enemy, they regain MP (or, to be technical, TP). And, most usefully, if the Prince is at full health at the end of a round, he emits a party-wide regen effect, perfect for keeping the party in top form but devastating if the Prince takes a hit. The 5-person-party means the Prince will almost certainly be filling the healer role (instead of the Monk, an unarmed combat expert with several direct-healing spells but few buffs), meaning that the entire focus of the party shifts to keeping him (or her, as a Princess) at full health.

The whole system becomes even more fascinating/complicated when, 10 hours in, the option to subclass becomes available. Each character gets access to the almost-complete skill tree of one other class, allowing you to develop synergies between skills and shore up weaknesses in the build. A Prince can be subclassed to a Monk, to be able to heal himself back up to full so that the party can benefit, or to a Ninja to boost his evade so that he's never hit. The complexities of the skill trees and the subclass system lead to a lot of my time away from the game thinking about what builds and what skills would work together to create an optimal party (or, in  some cases, apparently pointless skills make me wonder why they were included at all, leading to excitement when I stumble on a powerful synergy that makes these 'useless' skills incredibly powerful - a late game unlockable class, for instance, relies entirely on inflicting status effects on itself to increases its combat power).

And you need an optimal party. Or, near-optimal at least. Because:

3) Challenge

The Drowned City is, as I said, a brutally hard game, something it inherits from its PC forebears. That difficulty comes in two flavors - exploration, and bosses.

Exploration covers tackling the mazes themselves - solving puzzles, fighting enemies, keeping your characters healthy. The star here is the player's own mapmaking. I know this sounds dumb - like a chore the game designers force upon you - but there is a real thrill in navigating via a map you created yourself. Working out how to best notate traps and dungeon features, leaving notes for yourself - it's just plain fun, and it gives me a real sense of ownership over my exploration.

Exploration also plays heavily in the game's crafting system. Monsters drop items that you sell in town to get better equipment to kill more monsters to get items to etc., etc., etc. But many monsters have 'conditional' drops - only gained when the monster is, say, killed on the first turn of battle, or killed while their head is bound, or with fire damage. The myriad requirements (most of which are hinted at in the game, although some must be puzzled out on one's own) force the player to keep a diverse set of skills on hand lest they lose access to the most powerful equipment.

And you'll need that equipment to take down the bosses on display in this game. These are more-or-less perfectly tuned to the level the player should be at to face them (one of the best things about The Drowned City is that, until you reach the endgame, there is almost no reason for a smart player to grind at all). They use powerful moves (that can often be effectively countered if you have the right characters and a proper observation of patterns), status effects (which can be negated if you have the right build), and huge health bars (many of which can be whittled down more quickly if you take advantage of elemental weaknesses). The best thing I can say of the boss designs in The Drowned City is that, after having my ass handed to me by them, my response is never 'That was unfair. I'd better grind some more levels," but "I need to adjust my strategy and pay more attention next time."

Closing thoughts:

The genesis for this little(?) piece was a brief Twitter conversation Michael Peterson (aka @patchworkearth), author of the reliably fantastic web comic Project: Ballad, about his difficulties in trying to get into the games. The common point of comparison for us was SMT: Strange Journey, The Drowned City's stiffest competition for the title of best first-person dungeon crawl on the DS (a much prized trophy, no doubt). Strange Journey wins, easily, on the basis of being a game 'about' something - it has loads to say about religion, humanity, environmentalism... But from a strictly gameplay point of view, I find The Drowned City's class-system to be significantly more engaging, and less time-consuming, than Strange Journey's demon summoning/fusion system for gaining access to power and skills. A lot of that has to do with transparency - Strange Journey's system involves a lot of futzing around with fusion guides and skill inheritance rules, while Etrian lets you see every skill a class can get from the moment you create a character. That transparency lets you start planning out strategies from the moment you start, and that planning, for me, is the true heart of the game's appeal.