Friday, July 30, 2010

Lost Planet 2 stage design lessons

One of the games that my brother and I have been playing a lot of over the last few months is Lost Planet 2, the third person shooter made by Capcom. The game's strong emphasis on co-op gameplay makes it a great game for our purposes. However, the game definitely has its mix of great stages and poorly designed stages. After a while, I think I have noticed a few patterns, which could be taken as lessons to be learned for future games.

Ironically enough, mission 1-1 is one of the hardest stages in the entire game, particularly at higher difficulty levels. This is entirely due to the final segment of the stage, where the players have to seize and maintain control over a mine. There are two main causes for this difficulty. The first is because of the open layout of the mine; it consists of a large open central area surrounded by multi-level structures with lots of open doors and windows. Essentially, every enemy in the mine area can easily get opportunities to shoot at the players, meaning that the players have to deal with all of the enemies at once, making it easy to get swarmed or surrounded. Second, the game asks the players to maintain control over four control posts at once for a certain amount of time. Because of the spacing of the control points and constant reinforcement of enemies from multiple entry points, this is very difficult to do with even two people. It feels like this part of the mission more or less requires four players in order to be easily feasible.

This second point is seen again in mission 5-1, where the players are asked to once again maintain control over certain control posts for a certain amount of time; this time it is two control posts located in different rooms separated by several corridors. In our case, my brother and I were each able to guard a room, but it was a very difficult fight for us, since powerful enemies constantly stream in from three entrances into each room. Based on these missions, it feels like missions that require maintaining control over multiple places at once vary in difficulty a lot more significantly based on the number of players than missions that require simply advancing forward. This is probably because these missions force players to split up, which is more punishing for two players than for four. I have no idea if it is even possible for a solo player to tackle these (disregarding AI helpers).

Perhaps a more positive and interesting way the game takes advantage of its co-op focus is in its stages with multiple routes. Most stages in the game are very good at having at least two routes to get from one room to the next. For example, 4-1's first area consists of a multistory building. Many floors in this building have two or even three stairways leading to the next floor. This opens up a lot of room for strategy. The players can all go down the same path and try to combine their firepower, or they can choose to split up, take different routes, and flank the enemy. This multi-route design even lets players split up to clear out enemies and independently secure objectives if they so wish.

On the flip side of the coin is the final section of mission 4-1: a narrow choke-point leading into a large area where there are multiple heavily armed mechs ready to shoot anyone passing through the choke-point. To make matters worse, the only weapons capable of talking out the enemy mechs are on the other side of the choke-point. This kind of design comes up in about three or four places in Lost Planet 2, and it really is inexcusable. In all of these occasions, the only choice is to make a beeline for the usable mechs or good weapons and pray you can find some cover before the enemy fire tears you to shreds. This kind of area layout gives an overwhelming advantage to the enemies, and can quickly become very frustrating to the players.

On a final note, I really liked the co-op cannon segments of the game. In these, all of the players need to to work together to man powerful weapons against giant enemies. While one player is aiming the cannon itself, other players are manning anti-aircraft guns to hold off threats, fighting off enemies who have boarded the player's vehicle, or working to power-up the main gunner's next big shot. These battle can be frantic, complex, exciting, and very epic. Unfortunately, I gather that they aren't very fun when one player is playing solo. That is a real shame. Unless the game designers can write AI that can actually behave the way a player wants them to in such a complex situation, this is going to remain a trade-off when choosing between making a primarily co-op vs single-player experience.

I think there is a lot about Lost Planet 2 that shows that a game built around four player co-op does have significantly different level design considerations than a game built just for single-player. As a big fan of co-op games, I do kind of wish that there were more games like Lost Planet 2 out there.

Wednesday, July 21, 2010

Communication and Interaction in Fable 2

It has been far too long since I have last written a blog entry here. I have finally sat down to revive this blog with a post about a game that has been out for a couple of years: Lionhead Studio's Fable 2. I finally got around to playing it for the first time over the last week, and it was actually a fair bit better than I was expecting based on some reviews I have read about it. Unfortunately, one of the key selling points of the game, the ability to interact with the common characters throughout the world to earn their love or their fear, just fell flat. In practice, the lack of any real communication and the ability to only interact with faceless NPCs devoid of characterization makes it feel like the whole game world, including the player's own character, feel empty and lifeless. Fable 2 doesn't create the feeling that there are actual relationships between characters.

The fundamental problem with Fable 2's interaction is that it is impossible to actually talk to people; all you can do is use emotes called Expressions to get vague ideas across. Every NPC in the game-world seems to have their own base impression of the hero based on a combination of his renown (a value built up by doing quests), his alignment, and his appearance. From there, it is possible to further manipulate the impressions of the NPCs by taking various action, primarily Expressions, around them. For example you can pose heroically to impress people. NPCs will then occasionally make voice-acted comments as the hero walks by based on their personality traits and current impressions of the hero. However, at no point in this process do the NPCs and the hero actually interact in any significant way. It just feels like the player and the game-world are just talking at each other rather than engaging in anything substantial. It isn't even possible to tell which specific man or woman is actually doing the talking when they are in a crowd.

The biggest reason this interaction feels so hollow is because the people filling the game world really are hollow and lifeless. They are nothing more than names pasted onto three to six character personality traits, a handful of likes and dislikes, a generic character model, and a generic voice. I wouldn't be surprised if most of them where cranked out by a random NPC generator program. Furthermore, all of those statistics are devoted to determining how the NPC reacts to the hero performing any given Expression. All these NPCs are capable of is wandering around and reacting to Expressions. At the same time, none of them really stand out at all. They have no interesting personalities, they all look the same, and, worst of all, they are all equally frivolous in their emotions. Just by having my hero put on some nice clothes, half a city fell in love with my hero. Why should the player even care what the NPCs think of the hero when the NPCs are nothing more than generic background characters?

There are several honestly interesting characters in the game who are part of the story and the games various quests, such as Barnum, Hannah, and Garth. Unfortunately, these few interesting characters are completely segregated from any interaction. They generally have no reactions to the hero performing Expressions nearby, and they can't even be killed (one quest giver still offered me a job after I shot him in the head a couple dozen times). You can't even lock onto them to check their stats like you can with every other character in the game. It is really frustration that the only characters in the game that I actually care about don't really care much about my character.

The whole problem is compounded by the limitations of the Expressions. The Expressions are grouped based on how they influence people, and these groups include Flirty, Scary, Rude, and Fun. Pretty much all of the Expressions represent conscious performances put on by the hero to influence people's emotions and impressions. What is missing are Expressions that actually express the hero's own emotions. For example, it is possible to point and laugh at people to piss them off or humiliate them, but there is no crying emote to express sadness. This actually comes up in the handful of cutscenes where the player is asked to use Expressions to interact with story characters, such as when the player is attending the funeral of a slain monk. Without the ability to actually express the hero's own emotions, it is nearly impossible to engage in anything resembling a conversation. While the game does offer up Expressions for the player to use in such situations, trying to figure out the meaning of any given Expression in the scene's context is often very difficult. As such, the player's own character feels just as hollow and lifeless as the NPCs filling the world.

In the end, I never really bothered interacting with the NPCs of Fable 2 much during my go through. The game didn't give me much real motivation to do so. The NPCs and Expressions just felt too much like a mechanical puzzle to be cracked than anything with actual emotional payoff. In order to succeed with this kind of thing, you need to breathe life into the NPCs and treat them like actual characters, not just set-pieces.

Saturday, August 8, 2009

Avalon Code: Some Random Musings

I really have been meaning to write a follow-up to the post in which I assaulted the Book of Prophecy system in Avalon Code, but I feel rather torn regarding how I should follow that one up. On one hand, I feel like I really should elaborate a bit on how I think the Book of Prophecy idea could have been better implemented, but on another hand I also feel like I should just ignoring the positive side and instead describe how badly the game tortures you through the "Book Value" system. I guess all I can do is write a bit about both.

First, some more negativity.

One of the biggest issues with the Book of Prophecy is that every last page of the book has an associated "value" given by a number. Pretty much everything you do in the game other than pursuing the main storyline is related to this value. This number supposedly relates the "amount of information" contained within that page, but the method used to determine that number depends on the type of page in the book. For example, any page that allows Code alteration gives a higher value for putting more Codes on an object and giving the object more potent "Titles" created from Code combinations, with special bonuses for changing an item's form or giving a character the Title that they want the most. However, you also increase a page's book value by various things like talking to people and giving them gifts, killing monsters and juggling them with the irritatingly necessary Judgement Link attack, or just talking to your Spirits and using the powerful Spirit Magic attacks. At this point, the system doesn't sound so bad, but there are a few issues...

First, the connection between what builds up book value and the rewards you get for doing so simply are not very clear. I have absolutely no idea why making stronger weapons will cause the town's mayor to start holding a quiz minigame. I have no idea why doing well in that quiz minigame may cause a rare monster to appear. I have no idea why defeating that monster may cause some random part of the map to start producing Magic Jewels (the game's mostly unneeded currency). In effect, you just get randomly rewarded for going about and doing nothing particularly special. It is rather unsatisfying, in my opinion.

More problematically, the game isn't really rewarding you for any particular kind of activity. It is not like the game is rewarding you for doing the things you would normally do to win the game, since certain things that raise book value actually make the game more difficult, such as making monsters stronger by giving them powerful Codes and Titles. On the other hand, the game certainly isn't rewarding you for taking on greater challenges, since creating absurdly strong weapons that make battles a cakewalk will give you a lot of book value. The real problem, though, is that the game generally tends to give you a lot of book value (and other rewards) if you go around and do a lot of tedious, meaningless stuff that is mind-numbingly boring and frustrating, like checking every last square inch of most of the map squares in the game looking for things to examine (and watching the hero/heroine jump around and yell like an idiot whenever you don't find anything because the examine button is the same as the Judgement Link attack button). Basically, the game gives you a lot of rewards for doing things you will hate doing, which is a terrible game design.

Actually, there are a number more things I could complain about regarding how the game is so miserly in handing out good rewards and seems to enjoy in tormenting the player, but I really should focus a bit more on more constructive and positive endeavors. If I don't, I may go crazy thinking about how stupid I was to fall prey to some of that inanity. At least the experience taught me the rules behind solving 24-puzzles...

So, on to how things could have been better.

There are two things you see Avalon Code's Book of prophecy system that probably could have been used to better effect: unique special Codes and Codes that can't be freely removed from a page of the Book. In the actual game, these two things always coincide, and are always seen as problems that need to be removed, with the special Codes always vanishing when the problems they represent are cleared up. I think the system would be a bit better if unalterable Codes were far more common (with even normal, non-unique Codes being often unalterable), and unique special Codes could be removed and used on other pages of the Book. This arrangement would enable three things. First, making it harder to alter the Codes of something like an NPC or a monster would help the Code system have a stronger connection to the actual gameplay and characterization, since it would be easier to determine what codes like "Justice" or "Snake" are actually supposed to represent. Second, having two different tiers of Codes, with both common Codes (which you can create as many of as you like) and unique Codes, would allow a system where it is valuable to collect Codes, but also easy to alter the arrangement of a single page of the Book without having to flip through pages of the book searching for Codes or pull apart other code arrangements. Finally, it would allow a differentiation between Codes that add Titles and Codes that change the nature of an object in a way that could potentially remove the need for Metalize recipes, thus making Codes treasures in of themselves and encouraging greater player creativity.

Of course, a few other things would also need to be done. A greater connection between the Book and the reality of the game world could be emphasized by letting actions of the game characters more directly affect the contents of the Book. For example, instead of having to unlock a problem Code by adding a Title to a character, you could unlock it through the direct actions of solving that character's problems. As another example, instead of directly removing an enemy's Stone code using the Book whenever you want, you would instead weaken the monster's Code by adding another Code that contradicts the Stone, then hit the enemy with bombs and hammers in order to deactivate the Code, which would give you an opportunity to hurt the enemy with sword attacks as if the Stone code didn't exist. Any kind of greater interaction between Codes and real game states would have made Avalon Code a lot more interesting.

I think that is about enough commentary on that game for now. I've got too many other games to write about.

Romancing SaGa: BP, DP, and LP

Among the games I have been playing during my break from blogging is Romancing SaGa, a somewhat older game for the PS2 I recently stumbled across. While Romancing SaGa is a remake of an old SNES RPG that never made it state-side, it is pretty clear that it is a loose remake, since there are some pretty interesting innovations derived from more recent SaGa games. One of the best of these innovations is the BP/DP system, which is involved in the costs for using the various spells and special weapon techniques in the game. This system is light-years ahead of traditional MP costs and results in much more dynamic and strategic combat.

BP is short for "Battle Points", and it is the primarily limiting factor on what special moves that you can use in battle. Unlike in a traditional MP system, where characters have MP totals that are carried over between battles, BP resets at the beginning of every encounter. Each character has three pertinent BP values: their starting BP amount, their maximum BP total, and the amount of BP that they gain at the start of each turn of combat. As such, a character's BP total is in constant flux, going up and down as the battle progresses and the character uses his spells and special moves, which in turn means that what moves the player has access to are also in constant flux. For example, a character who starts a fight may have to wait until turn 2 before he can cast a certain spell. However, if that character keeps using weaker spells in the mean-time, he might never get to use the stronger spell. So, BP makes the player have to constantly weigh the opportunity cost of actions.

I find that this approach is much more interesting than the traditional per-adventure MP model, which doesn't usually force the player to make interesting resource-management decisions on a round-to-round basis. In most RPGs that use MP, the dominant strategy consists of two parts: conserve MP as much as possible during regular battles, then use your MP with complete abandon during boss battles. Even if MP gets low, there are typically plenty of items available to restore it, so low MP is rarely a factor in determining which move a character has access to in a major battle. In the BP system though, the best special moves often consume from a quarter to half of a character's max total. So even though it is constantly regenerating, using your best attacks constantly isn't necessarily a good idea. On the other hand, since BP is encounter based, it means that characters can use their various special moves without worrying about saving them for boss battles, which makes regular battles generally more exciting.

However, the biggest innovation of Romancing SaGa is that it does incorporate aspects of more traditional per-adventure resource management as well using DP (Durability Points). DP isn't a character statistic; rather, it is a quality of equipped weapons. A weapon has a maximum DP ranging from 20 to 60 or so, with 50 being pretty common. Attacks can deplete anywhere from 0 to 10 DP with each use, and when a weapon reaches 0 DP, it becomes useless until repaired. However, normal weapons can only have their DP restored by resting in an inn, while special weapons can only be repaired by forking over a lot of money to a blacksmith. As such, it is usually worth conserving DP until you need it. However, an interesting effect of the DP system is that it encourages characters to carry multiple weapons (such as a regular weapon to use DP draining attacks with and a more expensive weapon to use low DP attacks with), which adds some interesting tactical layers to combat.

Where the BP and DP systems really shine is in how they interact. Since every attack has two costs, the system allows for a wide range of special moves that are useful in an equally wide range of uses. For example, weak, low BP, zero DP cost attacks are great for saving up BP for a bigger attack when dealing with normal enemies. High damage, high BP, low DP attacks are good at quickly eliminating regular enemies, but are generally unsustainable during protracted combat. High damage, low BP, high DP attacks are great for dealing solid damage to a boss round after round, but they chew through weapons and are thus useless for fighting regular enemies. Since exact details such as individual BP regeneration rates, current BP totals, what weapons are available, what the costs of various moves are, and so forth are so variable, there are rarely obvious choices to make. The sum result is that each character has a wide range of interesting moves available and the player is forced to make interesting resource management decisions every turn.

However, the system does have a few weakpoints. Most notably, spells use BP, but not DP; instead, powerful spells consume a character's LP (Life Points). However, each character only has six to eight or so LP, and when they run out, they die. Since this is hardly the only way to lose LP in a fight either, using spells with an LP cost can be suicidal. Another problem is that the number of actually usable moves is a lot lower than it could of been due to poor balancing of moves and limited availability of various special moves. The system could definitely use some tweaking and refinement.

In the end though, I think that an RPG combat system that incorporates multiple resources instead of just one ultimately produces a more interesting combat experience than one that uses just one if handled well. Mixing per-encounter and per-adventure resources together works particularly well. I would definitely like to see more games use systems like this.

Wednesday, July 29, 2009

Avalon Code: Limitations of the Code

It certainly has been a while since I last posted an entry here...

In the interim since my last post, I have played through quite a few different videogames. I will try to make at least one post for every game I played in that time. Also, I have been taking a break from Persona 4 for about the same amount of time that I have been taking a break from blogging, so I will probably pick up my commentary about that game when I get back to playing it (which shouldn't be all that long now, though Shin Megami Tensei: Devil Survivor is calling to me...). For today, though, the game I want to write about is a DS action-RPG called Avalon Code, which I completed a few months ago.

Avalon Code is something of an experimental game, basing much of its system on the idea of the "Book of Prophecy" that records the "Codes" of every object, monster, and character in the game world. The game allows you to alter these codes using the DS's touch controls, so that you can change the properties of almost everything you encounter. It is a system that promises an incredible amount of player freedom and seems like a great playground for player creativity, but it does little to fulfill those promises. In fact, the system seems to do little more than add a few unnecessary levels of frustration onto fairly typical RPG mechanics, and does little to live up to the concept's incredible potential.

Honestly, I really want to start this off by listing a few of the things in the Book of Prophecy that are implemented in an interesting way, but I seem to come up with a caveat for each one. The Book of Prophecy is an incredible resource of information, giving you a complete record of everything in the game and everything you have done that is available at any time, but it is so bloated with information that it can be really hard to navigate the Book and find the information you are interested in. Every section of the Book has a detailed index that serves as a shortcut for navigation, but the index is the same size for every section, whether it is a short chapter of just 10 entries or a bloated monster with hundreds, making the indexing either a bit excessive or totally insufficient, with little in-between. The four spirit bookmarks you are given help a lot to quickly find the information you need, but they come and go with the plot far too much and the spirits that accompany them are mostly useless and are annoying enough to somehow manage to overcome my fairly generous tolerances for irritating companion creatures (I mostly used the spirit that can't talk, and she still somehow managed to grate on my nerves). Still, all these complaints are insignificant compared to some of the real flaws in Avalon Code's system.

I may as well just tackle the big problems in order...

First, it is a pain and a half to actually do any serious modification to Codes using the Book of Prophecy. Codes take the form of Tetris-piece like building blocks that have attactched generic properties like "Fire" or "Justice". Each object in the book has sixteen spaces to place Codes in, and individual Codes may occupy from one to four spaces. The problem is that you can only put four Codes into a holding area that can be carried between pages of the book, and every other individual Code piece has to be attached to some character, item, or creature. This means that if you want to complete rebuild the Codes for a character or item (which is necessary quite often, for a reason I will get to later), you need to take apart all of that object's Codes a few pieces at a time, finding spots on random creatures and characters for the unneeded Codes as you do so, and then you need to hunt down all the Codes you need and place them a few Codes at a time on the object you are creating. To make this all the more bothersome, there is no way to have the game find the Codes you need for you, so if you need something specific, like a two-space Fire Code (this kind of need comes up a lot), you might need to manually search through hundreds of character, item, and monster entries looking for the Code you need, and you may do so only discover that the copy you have of the Code you need is attached to some item that you can't do without, forcing you into a hard choice and often making the entire search a big waste of time. A basic search function, or even a list of how many you have of each kind of Code, would have done a lot to make the system a lot more usable. Better yet, the system could have been designed without the "every code needs to have a place" and the "you only have a finite number of any given type of code" assumptions. Because of these choices, the Book of Prophecy system is extremely user-unfriendly.

The next big problem with the book is that you are only allowed to have a single version of any category of item at one time. There are dozens of different kinds of sword, but they all just variations off the same sword archetype that are created with Codes. As such, even if you have the knowledge and Codes required to make both the Kaleila Sword and the Rune Blade, it is simply impossible within to use both at the same time because both are created by modifying the basic Sword with Codes. This is somewhat acceptable, given the nature of the game system, but the problem is that changing between the different weapons requires going through all the hassle I just described in the paragraph above. You can't save a record of how the Codes were arranged previously, or create a second sword, so changing from one sword to another requires completely dismantling the sword you are currently using and creating a new sword. This makes experimenting and trying new strategies rather bothersome. And annoyingly enough, the game asks you to modify your existing weapons for unusual situations or rebuild your weapon into something an NPC wants as a present far too regularly, so you may need to disassemble even your favorite ultimate weapon every so often.

Still, the biggest problems stem from the fact that, for all the headaches you have to endure in modifying items and altering the Codes of monsters and characters, there really isn't any point in doing so. No matter what you do to the Codes of a character, it doesn't change the way that character looks, acts, or talks. No matter how you play with the Codes of an item, the amount of HP it restores and the amount of MP it costs to use still remain the same. The only properties you can change on a weapon are its attack power, its knockback power, and its element. The only properties you can control for a monster are its HP, its defense, its weight, and its element. Other than that, you really can't control anything with Codes. The only point in changing the Codes of a weapon is to maximize attack power, since even controlling a weapon's element is nearly impossible given certain aspects of the Code system, and one of the best weapons in the game, the main character's unarmed attack, doesn't even use the Code system! The only real point in changing the codes of a monster is to make it weaker by removing the Codes that increase its HP. Most of the items that you can create with codes are just variations of keys needed to unlock doors and healing items with different combinations of HP gain and MP cost. All told, it is simply boring.

One of the most tragic flaws with the system, though, is the simple fact that it doesn't reward creativity and experimentation. In order to get any real results from the Code system, you need to find recipes called "Metalizes" and follow them to the letter. If you want a powerful sword, you need to find a Metalize for a powerful sword and follow the recipe. If you want to create a new kind of healing potion, you need to find a potion Metalize and follow the recipe. If you don't have the Metalize, then you can't create the weapon or item you are trying to make, even if you use the right Codes. Nothing you can create without Metalizes compares to the power of items that follow recipes, and trying to improve upon a recipe by adding some Codes to a completed Metalize item tends to cause that item to revert more generic and useless form. Basically, Metalizes just take the place of the treasures you would acquire in a normal RPG, and the only point of the Book of Prophecy is to make you jump through some hoops in order to get less functionality from those items than you would get in said normal RPG.

The final tragedy of this whole system is that the Book of Prophecy doesn't have a vital role in puzzles, dungeons, or boss fights outside of its role in providing you with the keys, items, and equipment you need and the rare case of dismantling an enemy's invincibility. Almost all of the game's puzzles are solved by clever use of the "style" attacks you learn by progressing through the game and using the different weapon categories. So, even the potential for the Book to be used in puzzles and story events is squandered. In many ways, the "Scepter" from the old Playstation game Granstream Saga did all of the interesting things the Book of Prophecy did in a less user-unfriendly manner, so I really am not impressed by the few unique things Avalon Code does with the idea, and the all the complexity just feels meaningless.

Honestly, I can also name any number of other things I don't like about the system, like how the whole game is a slave to the idea of "Book Value", how only the four metal Codes are worth anything, how incredibly annoying it is to be forced to take a break from modifying weapons and go juggling because modifying Codes costs MP, how annoying it is that one guy can only be Code-Scanned in a single scene in the game (and I missed that chance), and countless other not-insignificant problems. I would probably need to double the length of this post to point them all out in detail.

What really bothers me is that I still really love the concept behind the Book of Prophecy. It has a ton of potential, and a game that realized that potential would be incredible. However, in the case of Avalon Code, that potential is completely squandered by a series of flawed design choices. Any game trying to fulfill the potential of the concept would need to re-imagine the implementation at its most basic level.

Shadow of the Colossus: Argo

I finally managed to track down a copy of Shadow of the Colossus not too long ago, and I have finally beat it. Despite the games short length, I think that it is a real masterpiece of a game. Shadow of the Colossus is such a success because it manages to create powerful experiences and evoke strong emotions in those who play it. Since, I have already discussed fighting the Colossi before, this time I will talk about perhaps the most compelling character in the game: the hero's horse, Argo.

Even though Argo is merely a horse, and thus has no dialogue, he is nonetheless a character that I became attached to over the course of the game. Argo is the hero's, and thus the player's, constant companion throughout the course of the game. In particular, almost the entire entire exploration and journeying aspect of the game takes place from Argo's back. While the player often has to leave Argo behind in order to actually confront the Colossi, Argo does play a pivotal role in several of the game's epic battles. Furthermore, Argo is really the only friendly character the player has to interact with in the game, since the only other characters are Dormin, who is really only an ominous disembodied voice, and the hero's deceased love. As such, Argo's unwavering loyalty to the hero is the only thing that breaks up the quiet solitude of the game.

What really distinguishes Argo is that he is not just a vehicle for the hero to ride around on; Argo was created to be a character for the player to interact with. Argo doesn't just sit around waiting for the player, he often wanders off on his own, generally following the player around if you walk around on foot. Furthermore, once you saddle up, Argo proves to be realistically stubborn. I found that I had to be persistent and keep prodding Argo to get him to turn and run like I wanted him to. While Argo's tendancy to be difficult to control was annoying at first, I quickly got used to it and in the long run it made him a very realistic character. Pretty soon, I even discovered that I could use his intelligence to my advantage, since Argo was perfectly capable of navigating narrow canyon roads and such without any guidance from me other than an occasional reminder to keep running fast.

This characteristic of Argo's ended up being critical during the fight against the tenth colossus. In that battle, I needed to lure the fast-moving Colossus to pursue me as I fled away from it on horseback, at which point I could turn around in the saddle and fire an arrow at it's weakpoint. At first though, I kept aborting my attacks in order to guide Argo and make turns to avoid running into walls. As such, I had trouble making well-aimed shots, and I couldn't hit the enemy's weakpoint. I was only able to persevere in that battle when I came to the realization that I had to stop treating Argo like a vehicle under my control and I started to actually put my faith in him. So, I readied the bow and stopped worrying about trying to guide Argo around obstacles. Argo rewarded my faith several times over. Not only could Argo avoid obstacles and running into walls without my guidance, but he ended up being much better at dodging the attacks of the enemy colossus without my input. With Argo keeping us away from the enemy on his own, I was able to focus all of my attention on making my arrows hit their mark. Shortly thereafter, I started to make real headway in that battle. That was the point in the game where I fully stopped looking at Argo as tool in my arsenal for fighting the Colossi, and I started to see Argo as my partner.

By the end of the game, I had really grown attached to Argo. That is why the scene where he sacrifices himself and falls off a crumbling bridge t his apparent so that the hero could face the final colossus felt like a cheap gut-punch. During the game's ending, probably the most emotional part for me was seeing Argo walking back up the outer steps of the shrine like he always did. I was both very glad to see that he had survived the fall, and felt bad seeing him limping with a wounded leg. In the end, the fate of a horse overshadowed all of the other crazy things that happened during that ending. I suppose that is prook that the creators of Shadow of the Colossus made an amazing character.

I definitely wouldn't mind seeing more horses in future videogames act more like Argo. Maybe Epona could learn a thing or two.

Thursday, February 19, 2009

Devil May Cry 4: Gating and Mission Design

I was reading an interesting feature article on the website Gamasutra.com earlier today where a game designer was analyzing various elements of encounter and AI design in 3D brawler videogames. I was rather disappointed that he didn't analyze any entries in the Devil May Cry series; in fact, I am considering trying out his method of analysis on the game later. In the mean time though, one aspect of his article caught my eye: the concept of gating. While the concept behind gating is something that I was already familiar with, it is nice to finally have an accepted word for it.

In essence, gating is where the game developers put in obstacles to prevent the player from running away from or circumventing a battle. It is a valuable tool of the game developer, since it is pretty pointless to make a brawler-style game where the player can just walk right past every fight. There are a couple different ways of implementing gating: one way is to explicitly lock a player into a room until every enemy is dead, and the other is to design the enemies' attacks and AI to punish the player for trying to run away. While Devil May Cry 4 uses both methods of gating, it leans heavily towards the former in every mission, to the point where I think it begins to be detrimental to the game.

As is traditional in the Devil May Cry series, DMC 4's most common form of gating is where the game blocks off all of the doors and escape routes out of a room with red or white walls of energy. The games even use a specific color of wall to differentiate them from walls that can only be removed by solving puzzles or advancing the plot. Now then, there is nothing wrong with this strategy in of itself; plenty of videogames, including most incarnations of the Legend of Zelda series, lock the player in a room until the player defeats every enemy present. I think doing this is a great way to mark major battles. However, I think the technique is severely over-used in Devil May Cry 4. The typical mission structure of DMC 4 can be safely summed up as a string of three to five rooms where the player has to defeat every enemy in the group in order to proceed. While optional enemies sometimes appear, they are in the minority. After the first few missions set up like this, I actually began to forget that I could avoid some battles.

The problem is compounded by the fact that the majority of the battles that don't have explicit gating often are designed to take advantage of "soft" gating. For example, one of the few areas of the game where the player can run away from battles is in the long corridor sections of Fortuna castle. However, these long corridors are typically populated by either Frosts or Basilisks, both of which possess a mix of both high pursuit abilities and dangerous long-range attacks. So while the player can try to run from these enemies, he will probably take a few bad hits from enemy attacks. Furthermore, these groups of enemies are usually comparable in numbers and fighting strength to enemy groups that the player is locked in with. So even they still count as part of the linear chain of major fights.

Honestly, I think the developers went overboard with the gating in Devil May Cry 4. Gating is particularly useful for differentiating big, important fights from small, inconsequential fights. However, the developers decided to make just about every fight in Devil May Cry 4 a big fight, and cut out all of the little fights, and then they used strict gating to lock the player into that structure. However, that uniformity gets tiresome after a while. While Devil May Cry 4 has plenty of variety in enemies, it doesn't have enough variety in the kinds of encounters that the player faces. The consistent use of gating also denies the chance for the player to use running away, even of the temporary tactical withdrawal variety, as a viable strategy.