Monday, April 23, 2012

Opening Analysis: Fallout

Older RPGs are generally not considered to be the easiest games to get into, especially on the PC.  Complex systems of rules and mechanics to learn, stories steeped in arcane lore and fully of dense sub-plots, and a focus on managing multiple characters, quests, and other goals all at once often leads to a sense of things being impenetrable.  Even experienced RPG fans can sometimes find themselves intimidated by older games, or those based on rulesets they may not be familiar with.

The original Fallout is a game with such a reputation.  Generally considered uninviting and not very good about telling the player how to play or where to go, it's fair to say that many people stop playing before they even really get into the game.  However, I feel this reputation isn't really appropriate to the game itself.  Indeed, Fallout, despite its age, actually has one of the strongest opening segments in an RPG that I've seen in quite some time.  In this article, I'll be examining the early goings of Fallout and will demonstrate how it manages to ensure that players are given a thorough introduction to its gameplay and themes, without relying on heavy-handed tutorials or narration.

What's in a Introduction?

The goals of a game's introduction sequence are usually twofold: introduce the player to the rules of the game in such a way that prepares him or her for the challenges ahead, while also providing a solid narrative basis for the player's actions.  The latter is what we generally call backstory, although the extent of this can vary wildly - Fallout might require quite a lot to establish its world and characters, but Angry Birds doesn't have much to explain so it can get its narrative elements out of the way almost immediately and without having to spend a single word or line of dialogue on it.

The former is usually the bigger challenge, and it almost always takes the form of some sort of learning sequence.  Learning in the early stages of a game is the most critical part of the entire game - all game functions necessary for playing the game must be demonstrated.  This is best done through interactivity - a set of simple challenges designed to introduce game elements one by one and by providing a series of gates such that the player cannot progress without proving that he/she has a grasp of the mechanics, controls and/or interface.  This can be everything from "calibrating your HUD" in a shooter to teach basic camera controls, to requiring the player bash down a barrier to proceed.

Role-playing games, especially ones with complex rule systems, have things doubly hard.  Not only does a designer have to ensure the player is familiar with the mechanics of the game before starting, but the user interface has to be navigable by the player in a way that expresses all of those mechanics clearly and allows for easy interaction with them.  Typically, this is why an isometric, old-school RPG such as Icewind Dale is a bit more imposing than a modern action-RPG like Mass Effect - with more party members to manage, more ways to interact with the world, more abilities to use, etc., the interface is going to necessarily become more complicated.  The trade-off made is that once the player is familiar with the interface, there's typically a lot more the player can actually do in a game (such as Fallout allowing the player to use specific skills with specific objects or characters in the game world).

So, that out of the way, here's exactly what an opening to a game like Fallout needs to teach the player:
  • The key backstory (United States vs. China, world apocalypse, Vaults, etc.)
  • The most important characters (the player character, the Overseer - others are only relevant later in the game)
  • The player's goals and objectives (search the surface world for a Water Chip and bring it back to Vault 13 within 150 days, in order to save the Vault)
  • Movement (point and click)
  • Interacting with the game world (right-click context menu, hotkeys, skill use)
  • Inventory use (managing and equipping items, looting containers)
  • Combat (turn-based system, melee attacks, ranged attacks, aimed attacks, action points, hit points, reloading)
  • Different weapons (guns, unarmed, melee, etc.)
  • Using the world map
  • Random encounters
  • Following directions towards goals
  • Interacting with other NPCs (speaking, bartering, "ask me about" feature)
  • Importance of skill use in the game world
  • Choice and consequence within the game world
  • Distinction between combat and non-combat zones
  • The player's relative position of strength in the game world
Seems like a lot of things to teach, more than most other games, and you'd be right... but at the same time, Fallout is able to pull just about all of these off, and does so without ever using a single tutorial message or pop-up.

Interface and Controls

After its introductory sequence, which gets the first several points out of the way (the backstory, the player's goals, etc.), the player will find him or herself inside a cave, the huge Vault 13 door shut behind.  Interestingly, despite being a far more complicated and story-heavy game than most modern titles, its cutscenes still total just about five minutes, and can be easily skipped.  Many other games with much less to them have far, far more elaborate openings, which demonstrates how effective Fallout's relatively short and to the point beginning really is.

So, what have we got?  There's a corpse at the door, a computer console, and the path leading forward.  Already, things are tense: the music is moody and atmospheric, the corpse implies hostile circumstances, and the darkness of the cave ahead is certainly not very inviting.  Instantly, the stage for the bleak tone of the game is set.

The Vault-suited corpse and the unresponsive radio near the door is the first of many subtle, dark jokes throughout the game that say all they need to without a word.
Fallout teaches the fundamentals of its interface in surprisingly intuitive and covert ways.  This is accomplished through the corpse near the player - after all, just about everyone's going to want to inspect it.  To do so, Fallout requires the player right-click to change the cursor to "interact mode" and then left-click on the corpse once more.  This opens up the looting panel, which clearly displays both the player's inventory and the container's own items (in this case, a machete).  When the player's done, however, another problem appears - the player can't move anymore!  Well, right-clicking worked once, so what about again?  Sure as can be, a second right-click returns the cursor to movement mode and lets the player continue exploring.

But what about the machete on the corpse?  Provided the player can read, he/she will no doubt notice the "INV" icon prominently displayed on the user interface.  Clicking it opens up the inventory, where the player will notice a few things: first, that he/she hasn't been sent out without supplies; second, that the machete appeared at the top of the player's list; third, that any items can be equipped in the "Item 1" and "Item 2" slots, not just weapons.  Thus, the player has learned not just how to use the interface for moving, interacting, looting, and equipping items, but also that items can be used in the weapon slots - suggesting not only that, say, Stimpaks can be used to heal one's self, but that they can be used to heal others as well.

Like most things, Fallout appreciates that players are smart enough to figure the basics of inventory management for themselves, building scenarios to introduce it without telling the player to do anything.
Although this seems counter-intuitive and difficult to figure out, this is actually one of the most effective "invisible" tutorials I've ever seen.  With a single sequence, the game has taught the player the fundamental differences between movement mode and interaction mode, how to change between the two modes, how to interact with the environment, how to manage inventory, and a whole other mode of interacting with the world, through items.  It's not mandatory, but unless the designers wanted to break immersion with tutorial messages, this is about as quick and effective a way as you can teach the player these things.

Many modern games would attempt to map all the same functions to a single mouse button, and while that might seem more intuitive and easy to get a hold of, in drawing clear lines between different types of interactions, Fallout also avoids many of the interface muck-ups that other games have (such as attacking an NPC instead of speaking with him/her).  With a clear interface, there can be no ambiguity or mistakes in the player's own actions - to do something, you really do have to do it.


With those basics out of the way, the player will feel confident to set out into the cave ahead.  Doing so without running into one of the local denizens, however, is nearly impossible - they blend in with the game background, they're small, and moving nearby initiates combat automatically.  Could it be anything other than rats?

Rats.  RPGs, RPGs never change.
Due to the rats' small size and placement right on the path ahead, he player is almost guaranteed to end up in combat mode, and will more or less be forced to learn the basics of the interface and mechanics to proceed - taking minor damage demonstrates the hit point mechanics, seeing scrolling text in the bottom-left corner of the screen teaches the player about the value of the info log (which is used for both combat and non-combat functions), how to pass turns, and, of course, the player now knows how to shoot, stab and punch others (by clicking the big weapon button on the bottom of the screen).

Once again, the player has to figure all of this out.  Unlike most games, combat is something the player has to figure out on his or her own, without tutorial messages.  How do I attack?   How do I move?  Why can't I attack anymore?  How do I end a turn?  How do I change weapons?  Normally, this would be a lot to take in, and it is.  However, that's also the beauty of a turn-based game - the player has plenty of time to find out what does what, and the low-level rats the player will fight are almost impossible to lose to even with a very physically weak character.

In a game where hand-holding is almost nonexistent, tutorial pop-ups everywhere would have killed the flow of the game and would have coddled the player, leaving him/her unprepared for later on... by making the player use his/her head, not only is there a genuine feeling of learning and progress, but the player now knows to expect little help when the game gets more difficult.  By modern standards, this is sacrilege, but even today, getting into Fallout is fairly easy because of the way it's set up to encourage learning.  Even if the player misses out on something right now, the player is almost guaranteed to learn the fundamentals.

Once the player has dealt with the first enemies in the game just south of the Vault door, chances are he or she will be much more alert and ready for the next encounter - now that the player knows what rats are, it'll be much harder to run into them unintentionally.  The player may also have noticed that going into combat mode highlights available targets, making it an effective way of spotting them in the darkness.  This means that the player likely won't blunder into combat again, unless it really is his or her own fault.

The additional caverns to explore don't just provide extra XP and loot - they also teach the player about risk/reward, and about the optional nature of most combat in the game.
More importantly, though, this combat sequence and the level design of the area actually teach the player that much of the game's combat is avoidable.  The opening cave area has two caverns off to the east and west, both of them completely optional to explore, and with them, the combat within.  Although the player does receive some rewards for exploring and defeating the rats, the large groups of them off the beaten path may be off-putting.  If the player invested in Sneak, now's the time to try it out.  This lesson about avoiding fights, especially when outmatched, will be invaluable as the game goes on and the player learns that often, running is the best solution when faced with tough odds.  What's more, on subsequent plays, there's nothing to stop the player from simply rushing past the minor enemies and heading out into daylight.

World Map

When the player emerges into daylight for the first time, he/she will be presented with a brand-new interface element: the world map.  Aside from looking neat, it forms the basis for overland travel and is the lifeline between the different settlements across the game world.  Fallout's world map deviates from the norm in a key way - whereas most games have world maps that display the entire world on a single screen, in Fallout, the player only ever sees a 9x9 grid, with each settlement taking up one space on the grid.  To find more locations, the player will have to penetrate the fog-of-war and explore the outlying area.  Even in the heat of the sun, the feeling of a hostile, mysterious world is still preserved.

Starting out can be intimidating, but handily, we already have a list of two locations - Vault 13 and Vault 15.  The player knows (or should know) that Vault 13 is where the game began, so the most logical step to take, as per the Overseer's instructions in the intro video, is to head to Vault 15.  To do this, the player need only click the Vault 15 button on the interface, and he/she will begin traveling to the east.

Interestingly, there's some additional videogame logic at work here to guide the player.   Upon first appearing on the world map screen, Vault 13 is shown on the top-leftof the screen.  The black fog-of-war to the right and bottom of the location suggests the player go in those directions, rather than left or up.  Going up will eventually take the player to the top of the world, with little of interest otherwise, while going left will soon cause the player to stumble into far more difficult enemies and, more than likely, instant death.  Meanwhile, going right or down will take the player in the general directions of civilization.  It's a subtle touch, but very effective in steering the player.

The layout of the world map screen and the starting locations relative to each other is just one more way that Fallout teaches the player about the game and its world without breaking the fourth wall.
Pressing the button to initiate travel towards Vault 15 reveals a few interesting things.  First, moving around removes the fog-of-war.  Second, different types of terrain affect movement speed, which in turn affect time spent traveling.  Last, and most importantly, there's another settlement on the way to Vault 15!  The player will no doubt recognize it as such due to the big green circle (which looks exactly like the Vault 13 marker), but since the player hasn't arrived at Vault 15 yet, it must not be something entirely different.  But what?

This opens up an interesting dilemma.  Does the player continue on to Vault 15 or stop at this first settlement (by left-clicking the map to interrupt the current course)?  If the player continues on to Vault 15, he or she will find a few things - more difficult enemies to deal with (bigger rats), and soon after, it will be impossible to proceed, owing to a non-functional elevator that needs to be bypassed.  Even if the player skips by the first town initially, it will become necessary to visit it shortly after, either because the player runs from a tough fight, or because the player simply can't move on.  The decision to stop or continue is a compelling one to make and is the first step in cementing Fallout's open world nature, even if there's nothing really lost either way.

The world map also teaches another few important narrative facts about the game world.  The first: there is almost no civilization left.  The world as we know it has been destroyed, if not erased.   The land is barren and inhospitable.  This is something communicated in the introduction, but seeing it in-game has its own strong emotional resonance.  Most games would see the player exploring a wilderness full of life, or a sprawling metropolis - in Fallout, it's desolate, bleak deserts.

Should the player stray too far, there's an interesting gating mechanism at work.  Although the player is free to move in any direction, there are tough enemies off the beaten path - to the far west, Super Mutants (which are effectively impossible to fight until much later in the game), and to the south, Radscorpions and raiders (both of which are also more difficult and imposing than the rats the player is now used to).  Thus, it's pretty much impossible not to end up at the first town, Shady Sands.
Shady Sands

Arriving at Shady Sands, the player's learning phase is beginning to come to an end.  However, there's still a couple very important things left to see before the player can move on to the goal of Vault 15 - namely, more complex interaction with the world, speaking with other characters, bartering, and accepting quests and earning reputation.

The gate at Shady Sands and the warning to unequip weapons marks the difference in gameplay between combat and non-combat areas, as well as the narrative distinction between wasteland and civiliation that accompany them.
Right off the bat, the player will learn that Shady Sands is a peaceful place, as the guards at the gate tell the player to put away his/her weapons.  Speaking to one of the NPCs, Katrina, will give the player a valuable and welcomed exposition dump - basics on the game world, as well as mechanics like healing, buying new items, and so on.  This is the most direct tutorial the player will ever get, and speaking to Katrina will yield an XP bonus.  In another subtle bit of teaching, this sequence reinforces the value of learning through interacting and accomplishing goals rather than simply killing enemies (as 500 XP is likely more than what the player earned fighting the rats in the cave).

Inside Shady Sands, the player will spend most of his/her time exploring and learning about some of the game's finer points.  There is an early game quest to pick up, which sees the player ridding the town of its Radscorpion problem, which will likely provide the player with his/her first level-up and some more advanced combat experience (though there's an alternate non-combat solution, involving collapsing the Radscorpion cave with explosives, as well).  There's a lot of loot to find in the various containers around town.  There's a new companion to pick up, Ian, and doing so will most likely require that the player barter for some extra caps to pay him with.  There's the "tell me about" feature which can provide the player with extra information in dialogue (though sadly it's not very useful, and was removed from Fallout 2).  There are even a few Speech and Science checks to make that allow the player to appreciate the value of intellect over brawn.

Teaching farmers about crop rotation rewards the player's character-building choices, while also fleshing out the game world in intelligent ways.
Importantly, many of these early moments also reinforce certain narrative qualities as well.  The Radscoprion quest demonstrates that in the wasteland, humans are constantly struggling to survive against the wildlife.  The farmers, who lack understanding of crop rotation (which a smarter player can teach them), are only barely managing to get by.  The player even learns, through Tandi, the village leader's daughter, that human nature - desire for excitement, adventure and recklessness in youth - has remained largely the same even in the face of apocalypse.

Long story short, when the player is done with Shady Sands, he/she will feel truly ready to go ahead and tackle Vault 15.  If the player already visited it, then this learning phase in Shady Sands will instill a lot of extra confidence and will allow the player to gear up for its challenges.  If the player stopped at Shady Sands first, then the player will probably have to backtrack at least once, but won't find it truly intimidating and difficult either.  Most importantly, though, the player now has a clear distinction between and understanding of all the different modes of play the game has to offer - combat, exploration and conversation, and world map travel.
Vault 15

Vault 15 is the game's first major goal and marks the transition point from introduction and tutorial into the meat of the game.  Not only does it put the player's skills to the test, it also throws a huge monkey wrench in what likely seemed like a fairly straightforward experience up to this point.

The first, and most obvious sign that Vault 15 is a bigger threat is that it contains some substantially larger enemies.  If the player fought off the Radscorpions in Shady Sands, they won't seem nearly as threatening, but compared to the rats encountered earlier, the larger pig rats and the monstrous mole rats are certainly deadly.  Although the choice of so many rats might seem a bit odd, it actually provides a nice sense of mastery over the game world that another enemy type wouldn't provide, and draws attention to the effects of radiation on the local wildlife in increasingly more dangerous ways.  Once the player's dealt with the big critters, he/she will have symbolically moved beyond the intro phase.

Battling through the monsters, the player will come across the first major roadblock - the elevator is broken, and there's no way down!  In what is quite possibly one of the most infamous moments in RPG history, the player is unable to progress until he/she collects, of all things, a rope.  I cannot count the number of stories I've heard of players throwing their hands up in disbelief at this realization.  For many players, this will mean traveling back to Shady Sands to find one - there are a few available for barter, and one to find/steal.  Although this moment is often critiqued, I actually think it's a pivotal moment in Fallout - it forces the player to understand the importance of interacting with the environment, of using items to solve puzzles, and importantly, that not all problems can be solved with bullets.

Vault 15 is the player's first true test, and has both combat and environmental challenges that must be solved to make sure the player is adequately equipped to deal with the rest of the game.
After fighting more enemies, and gathering some more loot (the player will no doubt appreciate the ammo and the leather jacket, the first new piece of armor in the game, and another symbolic indicator of progress and mastery), the player will finally come to the Vault's control center.  If the player decided to revisit Vault 13 before coming here, then this will no doubt look familiar, but even so, reaching the end is fairly simple and straightforward, with a strong critical path and a few optional rooms to explore off to the sides.

Unfortunately, it also presents a huge twist in Fallout's gameplay and completely changes the player's relationship to the game world.  Simply put, there is no Water Chip in Vault 15.  Just about every player who ever gets this far in Fallout will have the same reaction: "so... now what?"
This really marks the moment when Fallout moves from being just another straightforward RPG into a fully open-ended title with non-linear narrative and goals.  While the player will always find the replacement Water Chip, eventually, at the same place every time, the ways to get there can be remarkably different.  Even returning to Vault 13 and speaking to the Overseer won't really provide much help - the player is truly on his or her own to move forward in the game, and Fallout completely embraces this by providing only vague directions and forcing the player to explore and interact, rather than follow instructions.

After all the basics of interface, mechanics, and gameplay modes are out of the way, the player is truly ready to take on the world.  This is where the tutorial ends, and where the real game begins.

Despite all these great things Fallout does to ease players into its inhospitable world, there are some definite issues that many players and designers over the years have correctly and fairly pointed out.  The first of these, the relatively complicated character system and the relative uselessness of certain skills compared to others, is by far the most damaging.  Unfortunately, it is possible to make an outright bad character in Fallout, and while it's not quite as easy to mess up as some players occasionally suggest, there's no way of knowing what's on the outside world, and thus what will be most useful, until going out yourself.  Woe to thee who tagged the Gambling skill.

There's also, unfortunately, the issue of the timer.  Specifically, 150 days doesn't seem like a lot, and while it's more time than the ticking click might suggest, putting the player on a time limit from the very beginning can feel stifling.  Worst case scenario, the player really does run out of time, earning a non-standard game over.  Starting over a game was actually fairly common back in 1997 when Fallout came out, especially in RPGs, but by modern standards, being forced to replay the beginning of a game due to a few bad early choices seems patently ridiculous.

Last, the tutorial isn't explicit about a few things.  Using skills in the environment may not be entirely clear, although with experimentation most players will eventually figure it out.  The same goes for more advanced combat functions like aimed shots, and even some of the basic rules in combat like action points.  These are all important, and unfortunately they're very hard to communicate in the passive way Fallout attempts.  Fortunately, back then, games still came with instruction manuals!

Fifteen years later, the first Fallout still demonstrates effective teaching of both its narrative and gameplay through subtle gating, smart scenarios of vary scales which lead from one to another, and by giving the player increasingly challenging tests with a good feeling of progression and mastery.  There might be a few rough edges, but for a game with so many moving parts, its opening accomplishes a lot without resorting to a single pop-up.  Compared to contemporaries like Baldur's Gate, Fallout's opening still stands the test of time as a smart and effective piece of game design.

Monday, April 9, 2012

Assassin's Creed: Controls Make the Game

Assassin's Creed is one of my favorite game series of the last generation.  It features a lot of really great things to like: interesting characters, a conspiracy nut storyline, an expansive world to explore, clever (although not at all accurate) integration with real-world history, and more.  Now that more information on the long-awaited Assassin's Creed III is starting to come out of the woodwork, we're also beginning to see some of Ubisoft's renewed vision for the franchise.

Among this is a statement by Alex Hutchinson, Creative Director on the title, about the revised controls for Assassin's Creed III.  It's an interesting read, and it got me thinking about the control scheme of Assassin's Creed, as well as how it ties into the overarching design of the series.  Unfortunately, that train of thought also led me to realize that the existing control scheme of the Assassin's Creed series is, in many ways, deeply flawed.  In this article, I'll be going into detail about that control scheme, and examine how and why it doesn't work, as well as how it can be improved upon.

The Puppeteer System

The original Assassin's Creed control scheme was one of the first things discussed in the run-up to the game's release.  Whereas most games control by having actions tied to specific buttons, in Assassin's Creed, the complexity of the environmental interactions necessitated by the game's parkour-style climbing in turn requires a different approach to controls.  When you can climb, jump, grab, swing, and everything in between, managing all those movements across a small handful of buttons can get very difficult.

Enter the "puppeteer" control system.  The principle behind it, which has since been used for all subsequent games in the Assassin's Creed series, is that the player's controller can be conceptually thought of as puppet strings connected to the different appendages of the avatar.  The A button controls the legs, for instance, while X controls the weapon hand, B controls the off hand, and Y controls the head.  The second component of this system is the high and low profile configurations.  By default, the player is in "stealth mode", or low profile, but holding RT will cause the player to enter high profile mode.  In high profile mode, attacking enemies is messy and loud, the player can run faster, jump farther, and generally perform actions that raise suspicion.

For a stealth game, this, in theory, is a fantastic control system.  Not only does it get rid of the "am I sneaking or not sneaking?" problem that some stealth games exhibit, it also consistently maps certain types of conceptual actions to certain buttons.  Pressing Y will allow the player to examine objects or scenes in the game world, just as holding Y will allow the player to enter Eagle Vision mode, highlighting different characters and interactive objects, for example.  When something involving the head, eyes, or looking occurs, the player can be confident that pressing Y will trigger the correct action.

Tyranny of the Open World

Unfortunately, the control scheme itself does not actually work very well in Assassin's Creed as a final game.  The original title was built around key assassinations, as well as performing mini-challenges to collect more information on the target - collecting part of the information would reveal the target's location, while collecting all of it would often provide some sort of bonus in the confrontation.  All of this took place in an expansive open world, relegated into very large areas on a city-by-city basis (with one large transitional area between them).

The big problem with the controls as realized in Assassin's Creed isn't really the controls themselves.  It's the rest of the game.  See, Assassin's Creed, especially the later titles in the franchise, isn't really a stealth game.  The marketing and the image might tell you it's about stealth, about assassinating, about being a silent predator... but it really isn't.  No, Assassin's Creed an open world game that is primarily about getting from A to B.  Whether you're running from objective to objective, or trying to solve a challenge, or escaping the guards, there's a good chance 90% of your game experience will be running and jumping.  Bearing this in mind, the controls, as they are implemented, do not support this mode of gameplay well.

Take, for example, the necessary input to run at top speed, something the player is bound to want to do just about all the time.  Assassin's Creed requires the player press and hold two buttons, the A button (for sprinting and jumping) and RT (for high profile mode), in order to move quickly.  Two buttons, plus the analogue stick?  Really?  When running for several seconds or even minutes, this becomes fatiguing.  On the PlayStation 3 controller, the extra distance to reach and force required to press the trigger makes this action even more awkward over extended periods.  Surely, it would be much easier and more comfortable if the player could run at full speed solely by moving the analogue stick, without pressing any buttons at all?

Despite featuring a beautiful open world, the Assassin's Creed series struggles to provide controls that work for navigating its complex geometry.
This was made worse in subsequent Assassin's Creed games.  Whereas the first in the series had a lot of additional gameplay to perform while the player was running from A to B, in the form of guards to avoid, cover-blowing beggars to fend off, and so on, this was mostly removed from Assassin's Creed II onward, presumably because players were annoyed with having to stop every few minutes.  While the first game could be said to resemble one big stealth and movement puzzle, becoming quite difficult in the later levels, Assassin's Creed II was effectively Grand Theft Auto: Italia, minus the cars and guns.

There is one argument that could be made, and it's a very poor one, in my opinion.  Grand Theft Auto IV (and some of the prior games) featured a very annoying control mechanism, whereby the player had to tap the run button as quickly as possible in order to sprint.  The reasoning behind this mechanism is that the player feels more involved when running around if he or she can press a button - it triggers the "something's happening!" sensation that interface designers will know makes animated loading screens a necessity.  But, open world games aren't about mashing buttons - the open world exists to give the player choice in experiencing and solving gameplay challenges.  Moving around is a function of choice-making, not the game in itself.  If you need to resort to the player holding down or mashing buttons to make things "interesting", you have a very serious problem with your game.

The end result of Assassin's Creed overall design is that, given the open world nature of the game and its heavy focus on traversal, the existing control scheme is simply not conducive to smooth and fun gameplay.  Players just want to get from A to B and stab someone, after all, and doing that is much more complicated and awkward than it needs to be.  I, for one, don't find it surprising that a control system designed primarily for a game about stealthily sneaking around, evading guards, blending in crowds, and so forth is not suited to the new model of gameplay realized especially in Assassin's Creed II.

Contextual Controls & Obfuscation

Spending more time with Assassin's Creed reveals another very interesting side of the gameplay - despite being more complex than they need to be, the controls are actually very deep in terms of the move set offered to the player.  There are all sorts of really cool things you can do: short jump vs. long jump, run across walls, wall jump, etc.  While Assassin's Creed is often derided for its imprecise controls, it actually has extremely precise ones - in fact, just about all the game world is assembled in a way which makes them useful.  If you take the time to master the puppeteer system, you will, in theory, be able to play the entire game without ever making a single mistake.

Despite this high level of precision, though, almost no player I've seen has actually been able to control Assassin's Creed in a way I'd call precise.  The context-sensitive nature of many of the controls, combined with the expectations players bring into the game, often ensures that our old habits simply do not work.  Additionally, by making everything context-sensitive, the game has to effectively guess at what the player wants to do when more than one action is available.

In a game like Uncharted, the controls are so smooth and precise because there is always only one path to follow: pressing up can only either mean "forward" or "up", not both.  But Assassin's Creed, set in an open world, has to contend with all sorts of problems.  Does the player want to climb?  Jump up?  Spin around and hang from the nearby ledge?  Leap off a roof and stab someone on the way down?  Wall jump?  The controls are precise as they can be contextually, but they are not intuitive because the context is very often different than what the player anticipates.
Uncharted lets the player do all sorts of crazy context-sensitive stuff, but it only works because of how limited the player's world is.  In an open world, those context-sensitive controls become a lot harder to fine-tune.
Part of this can be mitigated by proper teaching, which, unfortunately Assassin's Creed does not do very well.  Even though the first two games in the series have extended introductory sequences and take close to an hour just to give the player a weapon, I can guarantee almost no player will have a keen grasp of the controls or game functions by that point.  Even though we're master assassins, we're still running into walls and accidentally leaping to our deaths.

The reason for this is that the Assassin's Creed games do a poor job of actually teaching the player the more advanced maneuvers available.  While it will explain high and low profile, or how to run, or stab someone, it won't explain the precise control combination needed to wall jump in a certain direction, or how to hang from a ledge rather than jump over it, or how to fall straight down from a rooftop rather than leaping forward at full speed.

All of this more advanced information, fundamental to smoothly navigating the complex 3D geography of the world, is instead buried in optional tutorial videos and challenge modes.  The only reason I even know about all this is because I took the time in Assassin's Creed: Brotherhood to get gold medals on all of the optional challenges.  Most players will not do this, and yet the game almost requires the knowledge in order to control properly.

The Solution?

I don't have a perfect way to solve this problem.  It's a difficult and complex one that Ubisoft have clearly struggled with over the last several games in the series.  Still, there are some very basic and fundamental changes which I would make to the existing scheme that would go a long way towards mitigating the issues introduced by using a stealth-driven control interface in an open-world action-adventure.
  1. Ditch the puppeteer system.  I realize that this is one of the Assassin's Creed hallmarks and probably one of the biggest innovations the game has to offer.  But to be frank, it just does not work.  It is needlessly complicated and does not mesh well with the core gameplay, as I've explained above.
  2. No more high and low profile modes.  Like the puppeteering system, it's needlessly complicated and only adds additional button presses to rote game functions.  Walking slowly can be done by moving the control stick gently, or pressing the stick once to enter stealth mode.  Stealth kills and messy kills can be assigned to different buttons, or tap vs. hold.  That's all players really need the high/low profile modes for anyway.
  3. Play to player's expectations, not with them.  Familiar control schemes may not always be perfect, but if they work because players know them, that's better evidence than any to adopt them.  That's why most modern shooters use the left stick for moving and the right for looking - it may not be perfect depending on who you ask, but defying convention here is just going to increase a game's learning curve to no real benefit.
  4. Provide adequate tutorials that cover when and how players can use certain moves and perform certain actions.  Assassin's Creed does a really good job of this with combat, with environment interaction, and so on... but it falls on its face with the most important thing, movement.  If the player is going to need to learn how to do something to simply navigate the environment, then ensure the player actually learns it, and learns it well.
  5. Avoid feature bloat.  This is a more general design comment, but I would argue that the control situation with Assassin's Creed has only worsened over time.  More features mean more items, weapons, moves, and so on, and that means more buttons are going to be occupied on the controller, and there are more context-sensitive scenarios.  Keep the controls simple from the beginning and only ever add a new function to a button if you're a) sure it won't interfere with the core control concept, and b) certain it's necessary to the gameplay you're trying to achieve.
If I had to devise a control scheme for the game, I'd come up with something resembling the following instead:
  • A is your all-purpose jump and climb button.  Pressing it once near a wall initiates climbing, and subsequently it allows the player to jump in a direction based on the analogue stick's direction.
  • B is a universal "cancel" button, as well as a "drop down" button.
  • X interacts with an object.  For civilians, this means picking pockets.  In combat, this means drawing and using your sword.  For objects, it means opening doors or picking up crates.
  • Y activates the currently selected sub-item, like the player's gun, crossbow, throwing knife, etc.
  • LB provides loud and lethal assassinations.
  • RB provides quiet and stealthy assassinations.
  • LT initiates lock-on mode, which ideally would take the general shape of Zelda-style targeting.
  • RT blocks when in combat.
You may notice that some of these ideas are, in fact, borrowed from other Ubisoft games - specifically, LB and RB mapped to different types of attacks was first seen in Splinter Cell: Chaos Theory, which mapped lethal and non-lethal to the triggers.  Meanwhile, I do think some of the existing control ideas in Assassin's Creed are pretty good - I like the universal climb/jump button, I like the targeting mode, I like the radial menu for weapons and items, and I like the general interface navigation.  Those things don't need to change because they work just fine.  Unfortunately, they're strapped onto a problematic core control scheme.

The radial menu introduced in Assassin's Creed II is a smart solution to feature bloat, and is well laid out to allow for reflex-action selection of key items and weapons.
 There are still some issues with the above layout, as well.  Having different buttons mixed up a little between combat and exploration game modes is probably more inconsistent than it needs to be.  At the same time, if the goal is precision when performing different assassinations, then the bumpers work well; meanwhile, if you want button-mashy, timing-based combat, then using the X button is more appropriate in my opinion.  And, of course, you can't market the game based on your inventive control scheme, but that's a small loss when all's said and done.

Closing Thoughts

Assassin's Creed, despite all its issues with controls in my opinion, is still a fantastic game series and one of the few truly original and entertaining properties to come out of this console generation, especially one that deviates from the standard shooter model.  There is a lot of room for improvement, and I hope I've outlined that clearly in this article, but I do want to stress that the franchise still has plenty of successes as well, and they far outweigh the issues with controls.

For Assassin's Creed III, I hope Ubisoft will take a long and hard look at how to overhaul the control scheme, not just in terms of modifying the existing setup - they need to radically rethink it from the ground up, studying other titles in the same open world genre, and examining what works in other games, and why.  Here's hoping the next game is the first one where I don't end up accidentally killing myself as often as my targets.

Sunday, April 1, 2012

Bulletstorm: When Presentation Isn't Enough

Bulletstorm, released in February 2011, is the first game by People Can Fly after they were purchased by Epic Games.  As the developers of Painkiller, one of my favorite first-person shooters of the decade, I went into Bulletstorm expecting intense action, entertaining and unique weapons, varied enemies, and great, inventive level design.  Oh, and fantastic visuals as well.

While the graphics behind the game are as strong as ever, though, Bulletstorm completely missed its mark for me.  Although it still manages to have some enjoyable core action, I found myself constantly being pulled out of the experience.  In this article, I'll be examining some of the many, many questionable design decisions in Bulletstorm, both ideas that could have been improved through better execution, and ones that should have simply been left on the cutting room floor.

Introductions Aren't My Strong Suit

The first, and most completely perplexing part of Bulletstorm for me, is the surprisingly heavy focus on storytelling.  The opening hour or so of Bulletstorm has a ratio of cutscenes and quick-time-events to gameplay that I almost stopped playing outright.  I think it's fair of me to, when I start up something called Bulletstorm, expect a game that gets right to the action, but Bulletstorm more than wears out its welcome

Out of the initial time spent with the game, approximately 30-40 minutes are spent watching cutscenes or "playing" heavily scripted sequences, and the bulk of gameplay involves performing quick-time events and going through standard tutorial motions.  What purpose do all these cutscenes and flashbacks serve?  This is a game about impaling mutants on gigantic cacti - why is it so necessary to spend so much time establishing a setting and characters?  Most of those ideas are all eradicated by the plot about ten minutes into the game, anyway, so not only is the beginning boring (despite the pretty visuals), it's also completely irrelevant to anything going on in the rest of the game.  Half-Life 2, a game with ten times the narrative weight and substance than Bulletstorm, accomplished more in its opening 5 minutes than this game can in an hour.

You know a game's got issues when it starts with a quick-time-event.  A tasteless one, too.
Moreover, of all that time spent, much of it is completely irrelevant to the actual game.  About five minutes of the opening is dedicated to a shooting gallery section which has just about no relation to the rest of the game and occurs before the player has even been introduced to the core gameplay experience.  In fact, it takes close to an hour to even be introduced to the game's core defining mechanic, skillshots.  Why am I being made to sit through this lengthy, uninteresting sequence for over a half-hour before the game even starts?  This game is called Bulletstorm, and its tagline is "kill with skill" - so let me do that!  Most games begin to bore players if they can't get into the meat of a game in 10 minutes, and Bulletstorm has the gall to think it's worthy of asking even more time before it lets the player have fun.

Once the game actually gets into motion about an hour in, things generally improve, and the seriousness of the introduction completely evaporates as the bandit slaughter and explosions begin, but there is no excuse for how overblown, bloated, poorly paced, lengthy, and completely pointless Bulletstorm's opening act is.  It's hard to imagine just how this segment came to be, what the rationale was.  This game has the intellectual capability and grace of a drunk rhino - nobody is here to understand the twisted alcoholism-stained motives of grizzled protagonist Grey, or why he cares so deeply about his crew of cardboard cut-outs.  I have the sneaking suspicion this was all intended to be ironic, but that's no excuse to force the player to endure such a poor intro sequence.

Skillshots, or Not

As I mentioned, the core gameplay of Bulletstorm revolves around "skillshots", which could adequately be described as gun combos - depending on your means of killing enemies, whether that's using the environment, melee-type actions, guns, or all in concert, you're rewarded with points.  These points, in addition to tying into the game's online leaderboards (the only real "multiplayer" the game has), also feed into the game's upgrade system, which basically boils down to purchasing capacity increases and secondary fire options for the game's firearms.

On paper, all of this sounds great.  The problem with the skillshot system is that, as executed, it isn't particularly fun or skillful, and is actually undermined by other mechanics in the game.  It's a real shame, too, because it's one of the few original ideas the game has, and it isn't even able to use it to its full potential.

The "kill with skill" tagline featured heavily in Bulletstorm's marketing, but the in-game manifestation isn't too much more exciting than a slot machine.
The first major issue is that skillshots are just not hard to perform, and many occur near-randomly in standard gameplay.  The vast majority involve the following:
  • Killing an enemy by kicking him/her into an environmental hazard
  • Killing an enemy by shooting an explosive object
  • Killing an enemy by shooting him/her in a specific body part (usually the head or groin)
  • Killing an enemy by using a gun's secondary fire
  • Killing an enemy by combining melee attacks and gunfire
Noticed anything suspicious?  Three of these - secondary attacks, explosive objects, and headshots - are all things that most players do in virtually every other shooter.  These might require some modicum of skill, but they aren't really creative acts, and certainly not original.  Already, a majority of the game's rewards come from doing things that are standard fare in shooters.

The two remaining categories, environmental hazards and melee combos, are a bit more interesting, but even these become stale once it becomes clear that these almost exclusively revolve around kicking an enemy at something or using the Leash, which is able to stun and throw enemies around.  Moreover, the vast majority of skillshots are only compatible in very specific situations, and the things that differentiate them are basically aesthetic.  Kicking an enemy into the water vs. off a cliff is functionally identical, which becomes apparent after about 10 seconds of gameplay

Skillshots are fun to pull off, but many are so routine and standard that they rarely feel special or require creativity.
Additionally, and damningly in my opinion, skillshots don't really test your combat skill - rather, they test pattern recognition.  Most skillshots are only applicable to specific guns, enemies and environmental situations, and almost every single skillshot is instantly lethal to that enemy.  This means, that skillshots aren't really about executing - they're about recognizing enemy X, situation Y, and then putting two and two together.  While there is risk/reward in play, the reward for performing skillshots isn't particularly compelling, especially as so many kills that provide high point values are so easy to do.  Even near the end of the game, things do not improve - the player doesn't need to get creative, enemies do not require multiple skillshots to take down, and so forth.  The mechanic literally is a one-trick pony, only given an illusion of depth by having so many aesthetic variations on the same themes.

Last, the final nail in the coffin - Bulletstorm's skillshots are all about risk and reward, about going in to do reckless things and coming out on top, covered in blood... yet the game sports a regenerating health system and a very heavy emphasis on using cover... yeah, uh-oh.  Although the temptation to bunny-hop around Quake-style is always there, it's rarely possible because the game moves at a very slow pace.  Shoot for 15-20 seconds, and you need to hide behind a chest-high wall for another 10 seconds as your health comes back.  Playing a Vanguard in Mass Effect 2 felt more reckless and risky, and that was in a game with a formalized cover system!  Fixing this would have been pretty easy, as well - giving the player a health bonus for a skillful kill would have made skillshots both rewarding and necessary for survival.  As it stands, it's just a way to pretty up what is otherwise a very straightforward shooter.

Level Design?  What's That?
One of the things that made Painkiller so much fun was its smart and semi-open level design.  While the game was definitely linear, it had a ton of variety in its environments which directly impacted gameplay... situations with cover, without cover, where the environment was both a friend and enemy, multi-level areas with an emphasis on verticality, and so on.  The unhingedness of it all was probably the key to its success, but even so the game was so much fun because it made each combat encounter feel different.

Bulletstorm feels like it was designed by a computer.  That's not an understatement - I think it may have some of the most boring and by-the-numbers level design I have ever seen.  It's beautiful, yes, and I think the game's artists deserve a huge amount of credit for what they did there - but graphics aside, there just isn't anything substantial or interesting to see or do (with very rare exceptions).

Bulletstorm's levels sure are pretty.  Too bad they're almost entirely non-interactive and rarely feature more than straight, narrow corridors and copy-paste arena fights.
I mentioned quick-time-events earlier, but truth be told, they make up a surprisingly large portion of Bulletstorm's gameplay.  Whether it's an actual "press X not to die" situation, or a climbing mini-game that requires mashing the mouse buttons/triggers one after the other (which is repeated about five times more often than it should be), or a standard procession of obstacles which must be kicked down or pulled away with the Leash, the core of Bulletstorm, when not shooting anything, boils down to: walk down a narrow corridor, and every 20 seconds, press the indicated button when prompted.

I think it goes without saying: this is not fun.  But why is this?  I think it comes down mostly to the way the game prompts and guides every single action that isn't giving someone .44 caliber brain surgery.  Whereas games like Half-Life 2 are able to break up their action by giving the player some time to explore an environment at his or her own pace, and then find a solution on how to proceed, Bulletstorm flat-out tells you "KICK THIS THING" or "PULL THAT THING" in big, bright blue letters.  Any illusion of freedom is completely sucked away, and what should be breaks in the combat to absorb, explore and appreciate the hyper-detailed world are instead turned into an opportunity to usher the player to the next combat sequence.  Look, Bulletstorm - just because you are a stupid game does not mean you need to assume I am stupid.

While I can understand highlighting things like explosive barrels and other interactive objects the first time around, Bulletstorm does this throughout its entire length, and ruins all sense of exploration and problem-solving as a result.
The worst part is how completely unnecessary all of this is.  Bulletstorm is a game that, again, is about shooting people in ridiculous and satisfyingly brutal fashion.  It does not need these unfun quick-time-events every few minutes, nor does it need non-interactive dialogue sequences and cutscenes bookending everything.  It does not need to treat me like a moron and baby me through scenarios which never, ever go beyond "gee, maybe I should blow up the bright red barrel next to this breakable wall."  Most shooters, even dumb ones, are content to give the player a degree of authorship in solving puzzles - maybe you can shoot something, or kick it, or throw something, etc.  In Bulletstorm, a door can either be opened with a kick, or the Leash, or gunfire - not "all of the above."

There are a few times when Bulletstorm ever shows any real ingenuity or inventiveness in level design.  One of the most memorable comes in a sequence about a third of the way through the game, where the player remote-controls a gigantic robot dinosaur with rocket launchers attached to it and tears through a villa (now we're talking!).  Unlike most parts of the game, this sequence actually introduces a new play mechanic and challenges into both the narrative and the level design (there's some foreshadowing for it early on), and requires the player to change strategy for the duration.  While the segment is extremely easy, this kind of change-up shows a degree of creativity that is almost entirely missing from the rest of the game.  It's fun, and actually plays differently - whereas every level in the game should have been like this, refreshing, over the top, and ludicrous, in Bulletstorm it's a rare treat in between the rest of the monotony.

Other Issues

I know I'm nitpicking here, but here's another list of design sins that Bulletstorm committed that I can't fit anywhere else.
  • Many unskippable cutscenes, even after you've finished the game.  At least I can avoid the intro.
  • Poor context-sensitive controls.  Reload is mapped to the same key as interact.  There were multiple times where I performed an action when I was trying to reload my gun, once or twice resulting in death.
  • Poor checkpoint placement.  If I die, don't make me waste a minute running down a corridor, kicking down doors, and re-collecting doodads.
  • If you discover a tough skillshot and die, you don't retain knowledge of it in your database upon reloading... despite Bulletstorm being all about replaying for a higher score.
  • Most ammo must be purchased with points, but guns must also be upgraded with points.  Suffice to say, I almost never bought any ammo.
  • Very hard mode is quite easy, except for quick-time-events, which made up 90% of my deaths.
  • The game has regenerating health and a heavy emphasis on using cover, yet a lot of the cover in the game isn't tall enough to shield you fully, meaning your head gets blown off with surprising regularity even when you're ducked behind something.
  • No hard saves, which means starting a new game (which I did to grab screenshots for this article) kills your existing checkpoint.
  • Very limited ammo for most guns - like regenerating health, the game encourages you to avoid using the more exotic weapons and skillshots.
Closing Thoughts

The regrettable thing about this is that Bulletstorm is, for the most part, a competent game.  It's not extremely short unlike most other shooters, it's technically very competent, the guns are all fun to use and fairly distinct, it's got a reasonably unique look (even with Epic's artistry all over it),  and it goes back to a time where shooters could have bright colors, cartoonish violence, aliens and monsters, and didn't take place in a nebulously defined portion of the Middle East.  Heck, it's even got a couple of original mechanics - and that's saying something in this day and age.

And yet, it also succumbs to huge design flaws and oversights which probably could have been solved with more effort and care.  Skillshots needed to be about skill and talent instead of a way of giving out disproportionately large XP rewards and feedback for mundane tasks.  Enemies and levels needed to be built around the idea of using special skillshots with unique effects, instead of just creating aesthetic variations on the same repeating themes over and over.  The quick-time-events should have been either turned into a legitimate and interesting game mechanic, or eliminated.  And that introduction, along with all those story sequences and cutscenes, really, really needed to be cut.

 Bulletstorm finds itself in the peculiar situation of nailing all the fundamentals, and then completely botching the landing.  Here's hoping People Can Fly will be able to stretch their creative and design muscle a little bit more for their next game, whatever it might be.