Wednesday, January 30, 2008

Soul Nomad: Combat and Gameplay

I have been playing Soul Nomad obsessively for at least four days now. It has been plenty of time for me to get a pretty good idea of the game system. While Soul Nomad can be a very fun game to play, it also has quite a few glaring faults in its gameplay. These faults can lead to a frustrating experience at times.

The combat system is actually very simple. When a battle starts, the main character's unit is by itself. The first thing the player has to do is summon the other allied units into the battlefield. Summoning units cost the player a certain amount of money, which figures into the game's system of unit upkeep, though it never has been a problem since I have always had enough money to summon all of my units. However, the summoning system does have some difficulties associated with it. The first problem is that summoning units require space. For example, a 2x3 unit requires a 2x3 spot of unobstructed space on the battle grid to summon the unit. Early in the game this is no problem, but later in the game, when the player has seven or so 3x3 units, finding the space to summon every unit becomes a real problem, albeit not an insurmountable one. However, the real problem with the whole summoning system comes as a consequence of other aspects of the combat system.

Combat itself, like in the Ogre Battle series, is very simple. When you attack the enemy, the soldiers in your unit automatically attack the enemy unit using attacks determined by the row they are standing in. Melee attacks are usually executed from the front row, while the best ranged attacks are launched from the middle or back rows. Attacks can hit anything from only a single target on the enemies front row to the entire enemy unit.  After your attack concludes, the enemy will counter attack once, if possible. However, there are quite a few complicating factors of this system:
1) Targeting is completely random (or at least determined by unreliable hidden factors). 
There is no way to control which enemies each individual character attacks. Even if you want your Archers to shoot the row of enemy Pyromages, they might shoot the row of enemy Knights instead, and there is no way of predicting that outcome. This is much more frustrating in the case of allied support characters, who buff or heal at random. A certain degree of dependability is necessary in games where you have no direct control of the characters.
2) Targeting is done in one big wave.
Even if your first soldier's attack kills an enemy soldier, the rest of your unit can still waste their attacks on that dead enemy. This is particularly troublesome for melee units, which can only target the frontmost row of the enemy, making them generally weaker than ranged units. When this fact is combined with the fact that targeting is random, it means that the effectiveness of attacks can vary heavily.
3) Attacks are often very lethal.
It is often possible to wipe out an entire row of enemies, or lose an entire row of allies, in a single exchange. While melee units are pretty durable with support, mages, archers, and support characters are as fragile as glass, and die quickly to even regular attacks.
4) Once a unit's leader is dead, the entire unit disappears once both sides have attacked.
In a unit lead by a fragile character such as a mage, this means that an entire unit can be lost to a single attack by a ranged attacker.
5) Once a unit has lost even a few key units, its combat effectiveness drops quickly.
Many units are built on synergy. A mage unit with 4 mages in it has more destructive power and access to more tactics and special skills than a mage unit with just the leader. And a melee unit is in trouble if it loses its healer.

On top of all of this, there are also special attack skills that can be used by an attacking unit that has lost 20% of its stamina (the main unit resource besides individual soldier's hit points, which is drained after every action). While these skills vary in their effect (some hit only the enemy leader, while others can hit various rows or columns of enemies, or even the entire enemy unit), any skill will generally kill whatever it hits, particularly if the enemy has taken damage. Since it is really easy to target the enemy leader with one of these skills, since each unit probably has several choices by fairly late in the game, a single use of these skills will destroy an enemy unit. There has not been one battle in the game where I have had to deplete close to a third of uses of these skills in order to win, either. On the other hand, every enemy in the game also can use these skills, so if you let an enemy live too long, it will generally wipe out one of your own units.

Because of all of this lethality built into the system, actions are a very valuable resource in the game. Being able to get the drop on a potentially deadly enemy, such as a large mage group, is essential. This means that ACT is the single most important stat in the game. ACT determines both how often a unit's turn comes up, as well as how far it can move. Units with high ACT will be able to act much more quickly than other units, and move further with each individual movement. In other words, high ACT units will be able to rush in first, wipe out the first enemies, and move on before lower ACT units even have a chance of getting into position to help.

This is where the summoning system can become a problem. When a unit is summoned, there is typically a delay before they can take their first turn. This means that the enemy has the chance to move first. This leads to two different possible starting situations in a stage. In the first, the army has to run a certain large distance to reach the enemy. In this case, your units will have plenty of chances to buff themselves up and wear down their stamina to the point where they can use special attacks, and thus will wipe the floor with the enemy when they arrive. However, the slower moving units will probably be dropped and miss most of the fight. In the other situation, the player starts right next to the enemy. In this case, the enemy will get to move first, and might do some serious damage to the player's units before the player can act.

While every unit possesses special skills called Tactics (usally self-buffs), they don't see much use. The only really valuable buffs are ACT boosting ones, which I often use on the first turn so I can close in on the enemy. However, as soon as my units close in to combat range, tactics become useless, thanks to the value of each individual action. Sometimes I wonder if it would have been better to have made them more similar to swift actions in D&D: actions that don't waste a regular action, but can only be done once a round. The only exceptions to this trend are the mage tactics, which include spells that can hit every soldier in three adjacent unit for enough damage to kill every soldier but the leaders. Those spells are so over-powered that they can make mages really annoying opponents.

One last note; there are two kinds of battle you can fight in the game. These are story battles and Inspections. Story battles are the finite in number battles that progress the game's plot, while Inspections are the randomly generated fights you can try at any time. Unfortunately, Inspections tend to be filled with a large number of enemies that all share a predictable power level, and one leader who is usually way higher in level than the party. For example, my level 24-31 party dove into an Inspection with a listed level of 19+. The boss was over level 70. Even characters that were effective against the boss couldn't land a single hit. Thus, Inspections have so far been very frustrating and difficult. Thankfully story battles have provided plenty of rewards to keep me going through the game.

Old Favorites: Way of the Samurai 2

I wrote about Way of the Samurai in my last post, but there are still quite a few things I want to mention about the game and what it tries to achieve, so I decided will write about about its sequel, Way of the Samurai 2.

In many ways, Way of the Samurai 2 plays very differently than its predecessor. Instead of the small, nearly-abandoned train station setting of the last game, this game is set in a thriving, densely populated town. This game spans a longer period of time (more than a week, as opposed to just a few days) which now pass based on actual game time, rather than based on plot events, and the player now needs to rest and maintain a stamina meter. Also, the developers made quite a few changes to the combat system which help speed up combat and make it easier to fight large numbers of opponents (though this introduces an overly simplistic dominant strategy to the game), as well as a few improvements to the sword collection/improvement system.

Despite those other changes, the biggest change comes from the very different way the story is structured in the two games. In the first game, every phase of day has pre-determined plot events open to the player, and all of the important factors that decide what happens in the game come from your choices and actions in these events. However, Way of the Samurai 2 uses a more abstract system of calculating a numerical value for how much each of the different factions likes or hates the hero. With a few exceptions, your actions in story scenes affect that value, rather than affect the story directly. One of the most important ways of affecting that value is to take optional jobs for the different factions and earn their trust through work, completely independent of plot sequences. The final significant change is that none of the plot scenes in Way of the Samurai 2 allow you to control your character normally. Instead, they only present dialog options, so it it is no longer possible to move around freely, ask questions at your own discretion, draw your sword whenever you like, or just walk by a scene and ignore it entirely.

Because of all of the changes, Way of the Samurai 2 does not have its predecessor's perfect fluidity between dialog, free movement, and combat, which was responsible for a lot of the unique charm of the original game. The difference makes the player feel more distant from the character, and less immersed in the game. Also, the importance of faction value, rather than previous decisions, reduces the amount of potential variability of the plot and the potential complexity of the story. In order to get the best or worst endings of the first game, you need to play off of both sides of the central conflict, but doing so is impossible in this game. In fact, some seemingly viable and interesting routes through Way of the Samurai 2, such as helping Sayo and the Magistrates at the same time, are not actually allowed by the game. The game mostly allows you a number of normal endings for each faction path, as well as one or two "bad" endings for each faction path.

A major problem in both games are the "bad" endings for each path. In the first game, after going through a lot of effort in order to unite the different factions to fight against the military, the villain gives you the option to spare his life and join forces with him. Refusing gets you the best ending, and accepting gets you the worst ending. There are no other variables between those two paths. Almost all of the "bad" endings to the sequel are the exact same, including an ending where you take a young girl you have been protecting and teaching for the entire game and hand her over to a the slimiest thug in the city without any qualms. Pretty much all of these endings involve a spectacular reversal of the hero's implied personality and involve complete betrayal of everything he has done up to that point in the game. These kinds of endings feel more like cheap ways to increase the number of endings to the game, rather than proper routes. Important decisions that affect the outcome of a game should not come at the very end, and they should not involve complete departures from the course the hero has taken up to that point.

Anyways, while I am talking about endings, I should say a few good things about Way of the Samurai 2 in which it improves upon the original game. In Way of the Samurai, every story path results in a full-scale battle against the Japanese army, and every final battle is against the general of that army. However, in Way of the Samurai 2, the final sequence can play out very differently depending on which route you take through the game, and there are a few different final opponents (mostly the faction leaders and their lead henchmen). None of the ending sequences are quite as dramatic as the final battles of the original game, but they are more varied and more appropriately reflect the different paths through the game, and the game is more interesting because of the variation,

Before I forget, I should mention that I don't like Way of the Samurai 2's emphasis on earning trust through work, and the large length of time that comprises the game. As a whole, the number of plot events the player will take part in across the length of the game is very similar in both games, but each playthrough of Way of the Samurai 2 is longer, so the gameplay between plot events in the sequel feels more like filler than rewarding gameplay, especially because there are so few jobs available to each faction, and some of them are fairly boring (finding the parcel) or annoying (getting the kid home, pick-pocketing, and extorting money from peddlers). The occasional random attack by a ninja or homicidal maniac helps keep it interesting, but hardly seems to make up for some of the monotony.

Another major complaint I have with the game is the general arbitrariness in the connection between what you say and how it affects events and your standings with certain characters. For example, the very first dialog option you have in the game, after Sayo gives you some food, is a choice between telling her to get lost, and two different ways of thanking her. If you thank her one way, she will wander off and you can go onwards on your own business. If you thank her another way, she will wander off, bump into some thugs, and you will begin a completely different scene. Your choice determines the following scene, but the connection is not logical. As a result, I can never remember which choice does what in that scene, no matter how many time I play through it. Also, your choices in some of these early scenes, seems to affect how Kasumi (another major character with her own faction value) views the hero, even though she is not present to see the scene. This problem gets especially bad in the jobs that involve long conversations (both of which I labeled above as annoying). The consequences of your choices often make almost no sense.

Finally, I will say that I like the system of rewarding the player with a title for completing a playthrough particularly well or under unusual restrictions. In particular, I like ones titles like Pacifist, which requires you to choose an ending carefully in order to earn the title (Pacifist in particular is a good one because it reinforces the central themes of the game. The only way to get through the game without harming any one is to let an innocent child die, meaning that you need to raise your sword and fight in order to protect people.).

Anyways, while I sound harsh in my discussion of the game, I still like it. I never played it to 100% completion of all endings, titles, and swords, but I rarely do that kind of thing in a game, and I still get tempted to pick it back up and try to see some ending or sequence that I haven't seen yet.

Monday, January 28, 2008

Soul Nomad and Skipping Cut-Scenes

There is one thing that has really been bugging me about Soul Nomad & the World Eaters: the way it gives you the option of skipping cut-scenes. Of course, the option to skip long cut-scenes is a valuable tool in a game where you may need to retry certain battles with lots of cut scenes preceding them. However, it is the way that the function is handled in Soul Nomad that is annoying me.

In Soul Nomad, whenever you select to go to a plot related place from the overworld screen, a dialog box appears that asks if you want to skip the next cut-scene, and presents you with "Yes" and "No" options. If you select "No", the typical menu cancel sound is played and you watch the cut-scene. If you select "Yes", you skip right to next point of controllable action. This dialog box also appears after every battle where there is a cut-scene (though oddly enough, not if you select a cut-scene from a town menu). I find this process very frustrating, mostly because it is really easy to accidentally skip a cut scene and not even realize it.

The problem is that every time I have ever seen a confirmation dialog box in a video-game, I have simply selected "Yes" without even thinking about it. Most of the time I see such a dialog box is in Final Fantasy games and other such RPGs when I am trying to save. In order to speed up the process, I got used to selecting "Yes" without even thinking. Since I see the "Skip cut-scene?" dialog box so often in Soul Nomad, I end up answering it without thinking a lot of the time too. So, there are a few times when I accidentally skipped a cut-scene. A few minutes ago, I just ran into the worst case of this yet, where I think I might have accidentally skipped a cut scene that took place right after a pretty long battle. The only way to know for certain is to fight the entire battle over again. Joy...

I vastly prefer the much more common method of having to skip cut-scenes by pausing the cut-scene, then selecting a "Skip Cut-scene" option from a pause menu. That way, a person can only skip a cut-scene deliberately, and is at least informed of the cut-scene's existence before skipping it. Unfortunately, it is somewhat more common for game developers to make it so that you can skip cut-scenes by simply pressing Start, without giving a pause option during cut-scenes. This actually led to a problem for my brother a few months back, where he wanted to pause a game during a cut-scene and had no idea whether pressing Start would pause the game or skip the scene.

There just needs to be some more consistent standards for this kind of basic thing... Oh well, I will probably post some more detailed thoughts on Soul Nomad later this week.

Old Favorites: Way of the Samurai

I don't think Spike/Acquire's unusual action game Way of the Samurai has ever been widely known or very popular, but I think it is a game which does a lot of interesting things with its story and game flow. Most notably, it is a game which embraces both interactivity in the plot and a very movie-like style, and combines both with a lot of replay value.

In terms of the main combat system of the game and other elements of being an action game, it is not outstanding in any respect. The combat system has its variations and a few good mechanics that keep the game interesting, but a few annoying ones (like the difficulty in acquiring special moves and ease with which you can break swords). As an action game, the game is mediocre to average, at best. However, the action elements are just part of the real nature of this game.

As a whole, this game is about choosing your path through a story. In each scene of the game, there are countless things you can do, and each choice has a distinct impact on the plot. Most importantly, these decisions are not merely prompts asking you for a limited number of written options, since simply ignoring an plot event is often possible, and you can often choose from many events simply by choosing where to go at any particular point in time. Also, at almost any time in the game, including the middle of many story sequences, you can just draw your sword and start attacking people, and the plot will respond accordingly. This last element gives the game an unrivaled fluidity between story and action, so the player almost never gets pulled out of the sense that he is in control of the character's actions and identity.

You can see a great example of the variety of choices open to the player in the very first scene. Here, right when you start the game, you see a gang of thugs in the process of kidnapping a young woman. The first option is to simply walk by them without saying anything, and pick one of the two different roads to take to different areas (and different scenes). The next option is to speak to the thugs, and choose from one of the dialog choices, which results either in a fight or the thugs tying the hero to some railroad tracks. Another option is just draw your sword and provoke a fight with the thugs. Finally, you can just wait around a bit until another character shows up and starts a fight (and I only know if this one from FAQs). Based on what happens here, you can get introduced to characters on completely different sides of the central conflict and see the story from very different perspectives. I consider this one scene to be one of the best moments in all of the videogames I have ever played, and it is just one of many similar scenes with different outcomes.

The plot of the game is built on the idea that there is a certain set series of events that will occur, and that the will of the player is the only variable that can alter these events. In fact, one of the characters comments on this, and asks the main character to leave the area in order to prevent him from interfering with what will happen. The player can observe and alter every major event in the plot (though not every event in one playthrough, of course), and these choices affect the rest of the plot and change the ending. This is not implemented perfectly (the final sequence, the battle against the military invasion of the area, is mostly the same in every plot path and can't be avoided), but it lets the player understand the situation and deliberately act to try to control known future events, which is greatly empowering.

The consequence of having a game with so many flexible choices is that the game is very short on each playthrough. It is rather easy to finish the game in one or two sittings. But at the same time, the game has a lot of replay value, and the game developers have introduced a few things to make multiple playthroughs more fun. The ability to collect the swords of your opponents, and then improve those swords and bring them into the next playthrough of the game, is a nice touch that lets you try new fighting styles and combat strategies, even when going through the same fights from a previous playthrough of the game.

Way of the Smaurai is not a perfect game, but it does many things to create a great sense that the player is the character he is controlling, and has the power to influence the game world just as a person in that world should be able to do. It provides an immersive game with a lot of freedom.

Friday, January 25, 2008

Random Outcomes

I finally got the chance to start seriously playing Soul Nomad and the World Eaters, a game I got for Christmas. It has an interesting mix of features from both the Ogre Battle series and more traditional Nippon Ichi tactical RPGs. I like it so far, but there is one part of the game that is already annoying me: acquiring new rooms for units. In this game, you construct battlefield units by placing characters inside "rooms", which possess in-built properties, such as how many characters can be part of the unit and in what formation. The formation is an important thing, since what row a character is placed in determines which attack they use. Different rooms can also grant special bonuses, such granting a power boost for moving further.

The frustrating element of these rooms comes from how you get them. The only way to change what rooms you have is to use the Change command, which swaps out every room that is not locked for a completely random new one. There is no way to control what properties or formation the new room will possess. This means that if the player wants a room suitable for a melee unit, they need to continuously swap out rooms until one appears that has plenty of front-row spaces and has an effect that is half-way decent for a melee team. It can take dozens of Changes to finally get what you want.

This kind of mechanic is not that uncommon in videogames, and RPGs in particular. For example, it happens quite a bit in Persona 3 when fusing Personas. While you always get the same Persona whenever you fuse two old Personas, the spells and powers that the new Persona inherits from the old ones is randomly determined. So whenever my brother fuses Personas, he always has to continuously cancel fusions and try again until the resulting Persona has the right moves that he wanted it to inherit. I have seen him cancel fusions thirty or forty times in a row until the Persona finally got the right moves. The same kind of thing can happen in older RPGs that emulated some kind of dice-rolling mechanic for coming up with stats.

These kinds of mechanics are just annoying and frustrating. The randomness of these outcomes forces players to constantly try them over and over until they get what they want. It can also force players to be frustrated when they pass up an option, hoping to get a better result, only to regret it later since that option was the best outcome they came across. I know that has happened to me. Since these kind of situations rarely spell out what all of the possible outcomes are, it can be difficult to judge whether an individual outcome is good or not.

If I was a designer for Soul Nomad, I would have argued against swapping out rooms randomly. Instead, I would have elaborated more on the "home" system of the game, where a character can purchase certain rooms. I would have allowed the player to buy rooms for every unit (not just one unit), and have expanded it to include the ability to unlock or purchase new formations. This means that the players would have a greater flexibility in designing the units they want. Intentionally limiting this flexibility with random elements doesn't really improve the experience on the player. This change would eliminate the possibility of "bad vibes" hanging out in a new rooms, but that feature doesn't really do much in Soul Nomad anyways.

Thursday, January 24, 2008

Ar Tonelico Part 1

Game Completion: Hard to say... 30-50% complete, I imagine.

As I said before, Ar Tonelico is a quirky RPG. It has a unique setting, unusual mechanics, and a very interesting battle system. I talked a lot about the Grathmelding system over the last week, so I think I will just focus on the game's battle system for today. It is fairly complex, though, so I think I will just try to dissect it piece by piece.

Reyvateils: The central focus of Ar Tonelico is its cast of girls known as Reyvateils, and the battle system is no different. Unlike pretty much every other RPG I have played, where characters are mostly differentiated by stats and abilities, Ar Tonelico makes Reyvateils completely different than other characters. They are controlled differently, are the only characters to have MP, they can't use normal attacks, they use different kinds of equipment, only one can be in the party at a time... Most importantly, they are the single most important member of the team, with every non-Reyvateil character being a glorified meat-shield in comparison. This is even taken into account in the plot, with every strong group of enemies or NPCs having a Reyvateil or two in the team. I think this kind of total separation to be a really interesting idea that could be really useful in a lot more games, and it used to great effect in this game.

Song Magic: Reyvateils are important in battle because they use Song Magic. The song magic system is very different from typical RPG special ability systems, since it is based on MP being a restriction of power and duration, rather than being a simple limitation of how often a spell can be used. The MP cost of a spell is constantly counted down in small iterations of time between character actions, so that a Reyvateils MP reserve, which can easily be thousands of MP, can be completely drained by a single spell with a cost in the range of 30-70 MP. What is more, it is not a simple system of spells taking a while to cast; it is a system where you can freely control the chanting time, and as such the duration (for support-type Blue Magic) or power level (for attack-type Red Magic) of songs. You can cast a quick and cheap attack spell to do a light hit, or take a lot of time to build up a powerful attack spell that will defeat every opponent at once. Also, whenever you are not chanting a spell, the Reyvateil's MP quickly regenerates, so MP is not something that has to be conserved (though MP restoration items do have uses, and there are limits to how often you can use many songs).

The system is very fluid and fun, and the different powerful spells open up a lot of different strategies for the game (work towards a single finishing blow, work to resist damage, activate healing so that you can freely use the attack skills of the frontline fighters, etc), and it is very easy to switch strategy in the middle of a battle. This is easily one of the most innovative and fun systems I have seen in a game in quite a while.

The Ambience Field: This is the point where the glorified meat-shields come back into the game. As a whole, non-Reyvateil characters have three important functions: protecting the Reyvateil (a minor system, but a good one), another one I will describe below, and empowering the Ambience Field through their regular attacks. Essentially, each time a character attacks an enemy with a weapon, that enemy's Ambience Field value will rise (to a max of three), and that number will fall at other times. The higher the Ambience Field, the stronger the attack of a Reyvateil's Red Magic will be. It is a system which rewards a dedicated attack strategy and good timing with very powerful Red Magic attacks. It reinforces the central importance of the Reyvateil in battle (and thus in the story). It is both minor enough to not be necessary in minor battles, and powerful enough to be invaluable in boss battles, and I like it a lot.

The Harmogauge: The other important thing that non-Reyvateil characters do in Ar Tonelico is boost the all-important Harmogauge. It is a meter which is filled from the left with a blue line (based on the actions of the normal characters) and from the right by a purple line (based on the actions of the Reyvateil). When the two lines connect, a Harmocrystal is filled and the gauge resets (though it can be emptied to the point that a Harmocrystal is lost). Filling Harmocrystals has enormous benefits: the chanting speed of Song Magic increases, the power of normal characters increases and they gain access to stronger Skill attacks, powerful counterattack moves become available, and enemies will drop more items at the end of battle. That last benefit is very important.

It is necessary to get a high Harmoguage level in almost every battle, simply because it is the only way to get enough items for grathmelding and for sale (since selling dropped items is one of the few ways of getting sufficient amounts of money). This turns every random battle into a struggle to build the Harmogauge as high as possible before the enemies die. While that might sound like a chore, I find it is a rather good way to make normal battles more interesting, so they don't become an exercise in just dispatching enemies without any thought or excitement. However, it has a drawback...

The problem with the Harmogauge system is that an element of it, the Limit Gauge which controls your maximum number of Harmocrystals, is a highly flawed system, because the only efficient way to build that gauge to its maximum is to fire two reasonably charged shots of Red Magic, and it takes far too long to fill the gauge with Blue Magic. As a result, trying to get the rarer item drops of weak enemies (who won't survive two shots of red magic) is incredibly difficult, and the game rewards the offensive playstyle of using Red Magic a lot more than Blue Magic, despite their parity in longer boss fights. The game would probably be a bit better if the Limit Gauge system was altered somewhat.

As a whole the battle system of this game is a lot of fun. The only real drawback is that this game is not terribly difficult, so it seems that it is never really necessary or even possible to use the system to its full extent. This is certainly a battle system that should be used as inspiration for game mechanics of the future.

Wednesday, January 23, 2008

Exponential Defense Growth

I was talking with my brother about the Legend of Zelda series quite a bit over the last couple of days, and I realized that we had very different experiences in the final battle of The Legend of Zelda: Phantom Hourglass. While took serious damage in the fight and had to heal up using my potions, he managed to beat the entire final boss using up less than a third of his life meter. The difference? My brother was equipped with the power that cuts enemy damage in half. It made me realize how incredible defensive abilities can become in a Legend of Zelda game.

For example, look at The Legend of Zelda: Ocarina of Time. In that game, the main way of increasing the survivability of Link is to find Heart Containers and Heart Pieces, which increase the number of hearts Link has. However, that is not the only way. It is also possible to effectively increase Link's total amount of life by carrying Red or Blue Potions, which fully restore the life gauge. On top of that, there is a special blessing Link can receive near the end of the game that reduces all damage taken in half. Link starts with 3 hearts, but that can grow to twenty. With four potions, it becomes 100 hearts. With the special defense on top of that, Link effectively can take 200 hearts worth of damage before dying. In order to even challenge a fully stocked up Link, an enemy would need to dish out 8-10 hearts worth of damage with each hit.

The end result of this is a difficulty curve in the game that is skewed towards the beginning. The early stages in many Zelda games tend to be the most lethal. The final parts of the game tend to be the easiest, since it is hard for Link to actually die. This is backwards. The game should be easiest in the beginning, so that a new player can become accustomed to the gameplay and drawn into the story. The later parts should tend to be more difficult, so that the player feels challenged and engaged. As it is, many Zelda games can be frustratingly hard early on and anti-climactically easy once you get to the final battle.

This is sometimes justified by saying that it lets the player choose their own difficulty setting, but I don't think that actually works in practice. This setup forces the players who want an easier time through the game to undertake what could be the most difficult and time-consuming task in the game. It makes the dedicated completionists get stuck with an easy trip through the game. And it makes the early part of the game, before these radical power increases can be acquired, equally hard for everyone.

The problem would not be that bad if the increase in defensive power is more linear. It is easy for a game developer to take into account a gradual, steady increase in character life by gradually stepping up the strength and numbers of enemies. The really problematic element is the doubling of defense. A doubling of defense is the same as doubling the size of the player's life meter. It is something that has a drastic effect on the character's longevity. When game designers balance the difficulty of the game to assume that the player has not collected every item or power, a defense doubling item can throw that balance completely out of whack.

There are plenty of other game series with similar defense mechanics to the Zelda series. The Mega Man series has plenty of games where the player collects Life Upgrades, armor that cuts damage in half, and Life Tanks that fully restore health. The Metal Gear Solid series also has life increases, healing items, and optional equip-able body armor. So, this kind of problem can potentially happen in many other series than the Zelda series.

One way to keep this problem in check is to avoid straight multipliers. If the healing power of potions was a simple set small number of hearts (like how a ration in MGS heals only a set amount of health), then having a potion or not wouldn't have as much effect on game balance. Eliminating the huge effect of the defense doubling would also help limit the wild difficulty swings.

Tuesday, January 22, 2008

Using Items

Many RPGs let you use consumable items to attack enemies, create defensive barriers, heal allies, and various other effects. However, as I mentioned in my last post, I find that the extent to which I use such items varies quite heavily depending on which RPG I am playing. In some RPGs, I might use items regularly, while in others I might not use them at all. I am not certain that my experiences are universal, but I do see specific reasons for why I use items more often in some games than in others, and some types of items more than other types.

A major factor that determines how often an item gets used is the rarity and cost of the item. If I can only get a few items of a type, and there are no easy ways to get new items of that type, then I will probably never use that item. I will always "save it for later when I will need it more", and that means never using it at all unless there is a particularly difficult moment in the game in which the item is needed (which has only happened once or twice to my recollection). If an item can be used an infinite number of times, or can be acquired in great numbers cheaply, then I will use it freely.

Another fairly obvious factor is the overall usefulness of an item. I am fairly likely to use a powerful item (such as Final Fantasy's classic "Megalixer" item), especially if it is stronger than other items for its cost. An item which is weaker than other items for its cost is a lot less likely to be used. To use an example from the Final Fantasy series, if I can afford large numbers of both Potion and Hi-Potion, then I will use more Hi-Potion items simply because they heal more than normal Potions.

From this point on my argument gets a bit more complex, so first I will list some specific games (mostly taken from those I mentioned lately) in which I tend to use items, and some games in which I do not. In Chrono Trigger and Final Fantasy games, I only used HP restoration items (mostly only outside of battle), MP restoration items, status restoration items, and Megalixers (only in the final battle). In Atelier Iris 2, I used items of all kinds sporadically. In Ar Tonelico, I only use MP restoration items (and only rarely in boss battles). In Odin Sphere I used many items in every boss battle, and often used items in difficult normal battles. Taking all of this into account, there are a few things I can determine.

First, the opportunity cost of an item is extremely important. In the Final Fantasy series, healing items do not heal more than one person, while other magic spells do, so it is often better to use magic spells to heal many people at once in order to save time that could otherwise be used to do other things. On the other hand, in Atelier Iris 2 only one character can heal the whole party at once (and only with mild effectiveness and at high cost), and there are many healing items that can heal the whole party very well, so healing items are more likely to be used in a pinch than character-specific abilities. Players will tend to go with the most efficient strategies, so items will only be used if they are part of those strategies.

Second, items will be used more if there are effects that can only be created with an item. In many RPGs, this is mid-battle/mid-dungeon MP restoration. In these games, MP restoration items are some of the most important. Even if items are otherwise pointless, items with necessary unique effects will still be used by the player.

Third, the cost of an item is always relative to the cost of other options. In the Final Fantasy series, HP restoration items tend to be useful because it saves MP to be used on other things (such as attack spells), which means that using HP restoration items saves you from using rare and more expensive MP restoration items (since those are the only way to restore MP in a tight spot). In Atelier Iris 2, all character-specific abilities are drawn from a meter that is filled as the battle progresses, so using those abilities has no cost on any permanent resource, making even the cheap and common items of that game have a relatively high cost. Since using the most powerful effect at the lowest cost is the most basic of all strategies, players will tend to use the lowest-cost method to getting any effect.

Finally, there is a psychological cost involved in the ease of use of any particular method. Often, item menus are large messes which require far more tedious navigation than other menus. If it is annoying to sort through, then many players will not bother with it, and use options other than items whenever possible.

As a whole, the reason I used items more in Odin Sphere than in any other RPG I have ever played, even Atelier iris 2, is because everything in the game is designed to encourage item use. Items have a lower cost than character-specific abilties (it can take countless Phozons to refill the Phozon Gauge, practically as many as will be absorbed by the weapon in the stage). Also, items can be replenished easily through the Alchemy Mix system, and have powerful, unique effects that are necessary to winning major battles. Finally, the item menu is rather easy to navigate. Meanwhile, Atelier Iris 2 does many things that discourage item use, most important of which is the fact that items are relatively costly (despite the Mana Synthesis system making every useful item easy to acquire) and unnecessarily powerful (so strong that the basic Flame item can wipe out a whole enemy group).

It is important that games match their systems to the way the game will be played. Players should spend more time and effort on things that will be necessary and useful, rather than on things which will not be useful or are unnecessary. Developers should not waste effort making countless items and systems that will not be used by the player. Items should not become Fake Rewards because they will never be needed due to contradictions in game balance.

Monday, January 21, 2008

Game Ideas: Some elements that Naruto games really should have

I have to admit that I am a huge fan of the Naruto manga and anime series. Ever since the last set of three volumes was released in the US last December, I have been kicking around in my head some ideas for what I would create if I had the chance to make a video game based on the series. Since I think about this kind of thing anyways, I thought I would put some of my thoughts down here. Since tie-ins to popular properties are a common and sometimes lucrative part of the industry, and there are plenty of Naruto games already, I don't think it is a pointless effort.

First off, I should mention that I have not yet played any of the Naruto games that have been released so far, though I have watched a few trailers for many of them and know a thing or two about the Naruto Ultimate Hero series. I think I have some differences in opinion with those games anyways. Anyways, Naruto is a Shounen Jump series that is all about one-on-one duels between super-powered ninja. It is a natural fit to be turned into a fighting game.

Now then, there is plenty of material inside the series itself to help flesh out the mechanics and movesets of a fighting game. I sometimes wonder if the the author, Masashi Kishimoto, began to intentionally pattern some of the moves of the characters on video game mechanics starting in the middle of the Chunin arc. At the very least, he provided many setting details that help define how a Naruto game can be played.

First off, there is chakra, the energy that fuels all of the special techniques the characters use. Chakra is thus a natural choice to form a special gauge that the characters deplete to perform special moves. The idea of special meter like this is very common in fighting games, such as the Tension meter in Guilty Gear or the Ki meter in the DBZ Budokai series, though its use is often limited to the most powerful kinds of moves. Characters in Naruto can often be seen gathering chakra (often with a certain identifying pose too), so it would make sense to give the characters the ability to charge up the chakra meter by holding down certain buttons.

The kinds of moves that Naruto characters tend to be easily grouped into useful catagories:
1) Regular attacks: The basic punches and kicks that every fighting game character ever seen uses.
2) Ninja weapon attacks: Throwing shuriken and kunai is something that most of the characters in the game would be able to do. Paper bombs, smoke bombs, wires, and giant shuriken also appear. So, there needs to be a mechanic for these kinds of attack.
3) Special attacks: This is where the chakra meter comes in. Every character has an array of special attacks that use chakra to execute. There are big flashy ranged moves like Sasuke's Pheonix Flower, and there are multi-hit combo-attacks that are initiated with a single opening blow such as his Shishirendan. There are also hard to dodge physical attacks such as Lee's Konoha Hurricane. These moves can either seriously weaken or finish off opponents.
4) Special Counters: It is also a common thing for characters to have special defensive or counter attacks, such as Neji's Rotation, Lee's Floating leaf kick, or the Substitution jutsu that just about everyone uses. The flashiest example is Kakashi's ability to mirror his opponent's Special Attacks when they are used on him.
5) Transformations: Between Naruto's Demon Fox forms, Sasuke's curse mark, Rock Lee's 8 Inner Gates, and Choji's special reserve, a significant fraction of the cast shares the ability to change into a super-powered mode where they have way more chakra and power than before. These transformations share some common features that make it easy to design mechanics for them. First off, they give the characters more chakra than normal, which can be emulated by either increasing the size of the chakra meter or letting it regenerate faster than normal. Second, they usually grant more speed and attack power. Third, they are often tiered in sets of two or three. Finally, they all have adverse affects on the bodies of the transformed, which might take the form of slowly draining the life points of the transformed fighter or something similar.
6) Alternate modes: Different than transformations, most characters have special powers that they can turn on or off separately than their transformations. Another Sasuke example is his Sharingan, which would give him passive benefits and open up additional special moves, at the cost of a constant Chakra drain. He also switches the Sharingan on and off independent of his transformation. This also includes other persistent abilities that change a character's fighting style, such as Naruto's shadow clones or Neji's Byakugan. Most would be a constant drain effect that can be turned on or off, or a one time cost effect that can be easily lost.
7) Ultimate attacks: A common thing in Naruto is for a character to have a single attack that is designed to finish off an opponent in a single massive blow, much like Instant Kills from the Guilty Gear games. These include the Rasengan, the Chidori, and Lee's Lotus attacks. They can finish an opponent off instantly if they connect (or at least do some serious damage), tend to power-up and change form when the user transforms, and can only be used a very limited number of times or at some significant risk to the user. However, they also can be easily dodged or countered if used recklessly.

There are a few things that should be mentioned about the nature of combat in Naruto that are both tricky to emulate, but are the most important part of the series' feel. First off, combat takes place in a complex multi-dimensional space. Battles tend to be in huge forests where the fighters are constantly ducking behind cover and moving between layers. The entire cast can walk on vertical surfaces and water. Thus, a traditional flat, one-dimensional space is too limited for the kinds of movement seen in Naruto. At the very least, more complex terrain, like what is seen in Super Smash Bros. or DBZ Tenkaichi Budokai is necessary.

Second, combat in Naruto tends to focus on characters trying to land a small number of powerful blows, where most attacks are dodged or avoided. Even big attacks like the Chidori or Rasengan can be easily stopped by a character catching the arm of the attacker. I never liked the button-mashing/minigame approach to deciding the outcome of this style of attack that is seen in many anime titles, and I think a pervasive dodge/parry system would be more appropriate to the overall feel of Naruto.

The biggest problem with a Naruto based fighting game is that many important members of the cast do not have fully realized movesets. While the developers could try to find things to flesh out the move-lists of these characters with more obscure moves, I think there is a different approach. It might be possible to let the characters have customizable movesets, where you could equip characters with generic moves (such as various elemental jutsu) that some characters can use better than others. While many of the major characters don't have fleshed out move-lists, often their general strengths and weaknesses are known (and sometimes plotted out on handy graphs).

If I ever get around to playing a Naruto game, I plan on critiquing it based on this outline.

Sunday, January 20, 2008

Item Creation Systems, Part 2

In my last post, I compared the the item creation systems in Ar Tonelico and Atelier Iris 2. As a whole, I found Atelier Iris 2's system to be a fun and effective system. However, while Atelier Iris that the basic premise I discussed, which I would consider the traditional system (item creation based on combining different basic items into more complex items, which are then fused into more complex items) is a viable system, it is not the only system that is good. In many ways, I consider the item creation system of Odin Sphere, an action RPG for the PS2, to be superior in some ways.

First, I need to point out a fundamental similarity between Odin Sphere's Alchemy Mixing system and Atelier Iris's Item Synthesis system: both systems rely on finding recipes, and following those recipes in order to make an item. Overall, I think this is a superior system to the alternative (not providing recipes and instead asking the player to experiment to learn anything), simply because it is easier for the player to use, and involves less frustration.

Despite that similarity, it is at the level of recipes that Odin Sphere makes one of its most important departures from more typical item creation systems like the one seen in Atelier Iris 2. In Odin sphere, all items are created from mixing one of five certain types of the item "Material" with one of the five "Mandragoras" or a "Troll Molar" or "Milk". As such, there is an absolute limit on the number of possible combinations, and in the game every possible combination has a recipe. It is a very limited set of items, but because the number of ingredients for mixing is small and there are only a few recipes (often following elemental themes based on Mandragora), it is very easy to memorize the entire system. This makes it very easy to use. At the same time, it means each mixable item is more unique and important, rather than a minor thing that will become obsolete or forgotten.

Another elegant aspect of the Odin Sphere system is found in the Material item. This item is essential to all alchemical mixes, with the value of the mix determining the resulting item. However, rather than these items being static ingredients, they are incredibly flexible. Any value of Material can be created by mixing a Material with common items like seeds, bones, bread, or fruit, or by mixing two Materials. Because this system is explicitly mathematical with no hidden processes it is easy to calculate, and creating a Material does not depend on finding particular items or special ingredients. All you have to do is combine a Material with items you don't need (which you need to do because inventory space is terribly limited), and you will eventually get the Material you want. Also, the game rewards mastery of this part of the process by giving you lots of "Phozons" (the functional equivalent to Experience Points in the game) if you create Material with really high values.

The item required for the Alchemy Mix other than the Material is another interesting side of the system. Here, the game takes a fundamental problem of more traditional item creation systems, the rarity of necessary component items, and turns it into something more interesting. In Odin Sphere, it is pretty much always possible to find a Mandragora when you want to Mix something, but what Mandragoras you can find depends on the stage. you will always be able to find the Mandragoras you need to complete a stage, but it may be difficult to find the other Mandragoras. This forces the player to think about what items to stock up on based on which stage he wants to attempt, and makes resource management more interesting.

The Odin Sphere system does not have the advantages of Atelier iris 2's complexity, which allows detailed item customization and the creation of equipment, but it is a very simple and easy to use system that is very well integrated into the game. It is one of several reasons that, even though Atelier Iris 2 is built around an item creation system, I actually used the items I created a lot more in Odin Sphere. However, I will need to discuss that another time.

Friday, January 18, 2008

Mario & Sonic at the Olympic Games (Wii) Revisit

After spending the last couple of days trying more seriously to win gold medals to unlock stuff in Mario & Sonic Olympics, I have to admit that I was wrong about some of the events. Part of some of my earlier problems with Table Tennis had to do with some misconceptions on my part of how the controls worked. Previously, I tried pulling the controller to the side then slamming it forward, like how you would if the remote was an actual table tennis paddle. However, upon closer examination of the control demonstration, I realized that you just hold the remote in front of you and make small movements to the side. Once I had that straight, it took me only two matches to get the timing down and go from utter incompetence to total domination.

I also was able to have a really fun time on one fencing match. Since I was playing a character who seemed to have a slow thrust, I put some effort into mastering the parry. I discovered that if you have a really good reaction time, it is actually possible to react to the glint in your opponents eyes and execute a successful parry. For the first time, the feint option actually looks useful to me, but only if you are fighting a human opponent who likes to parry.

From what I have seen (since I haven't tried them myself yet), rowing and archery are solid events. Rowing is a pretty complex event, but the instructions are pretty clear at all times, so it is possible for someone to make a really good showing on their first try. Archery, like some other events, is an event that takes two or three tries to master the controls. The lack of position tracking on the nunchuk attachment makes the game trickier than it should be.

I still enjoy the swimming events. I was able to break the world record on my very first attempt at the 100m Freestyle relay. Goes to show that being successful at a mini-game is often what can make someone like it. If someone's first exposure to an event is inexplicable failure, the event can quickly feel frustrating and not fun. The frustrations I alluded to in my last post actually put me and my brother off the game for a while.

What brought the two of us back was the burning desire to unlock more events in the game. While having unlockable content in a game like can be frustrating to someone who just bought the game, my experiences with this game have demonstrated to me that it has a definite use. People just seem more willing to put up with a frustrating experience if they have to do it to unlock something. While that is no excuse to let a gameplay experience be frustrating, it does help in this game. Many of the events in Mario & Sonic at the Olympic Games require playing through them a few times to get a feel for the controls. Forcing people to play them long enough to get good at the controls lets the players get over the initial frustration and discover that the game is more fun than they might have originally thought.

Of course, I wish the game would hurry up and let me play more Dream events already. The Dream Race is the coolest event in the game so far. It is Mario Kart as a foot-race. You could build an entire game around the concept.

Currently, the only problem I have with the gameplay is that it is not clear in any given event what each of the four character statistics actually does. More information in the instruction pages would help character selection out a lot.

Item Creation Systems, Part 1

Since I have finally completed many of the games I have been playing over the last few months, I am going back to a few old ones I want to play some more. Right now, I happen to be playing Ar Tonelico, an RPG for the PS2. This is a quirky game, which makes it fun to write about, so I will certainly have a lot to say about it, but for now, I can only think of one thing: I hate Grathmelding.

Grathmelding is the name for the system that lets you combine old or useless items into new and useful ones. It is a great concept, but the execution in this game is just terrible. What is worse, Ar Tonelico is made by the same developers as the Atelier series, an RPG series which uses item creation as the central mechanic to good effect, and it begs the direct comparison because Ar Tonelico and the Atelier Iris series seem to share the same basic game engine and a lot of artwork. It seems entirely likely that the Grathmelding system is in the game simply because Ar Tonelico is an offshoot of the Atelier series, even though its inclusion is glaringly forced into the setting and plot.

For the following comparison, I will be drawing examples from Atelier Iris 2, mostly because that is the only Atelier game I have played.

The Grathmelding system is actually very similar to the Synthesis system of Atelier Iris 2. Both are based on combining common junk items into basic items, then combining those basic items with other junk into more complex items, and continuing that cycle until the end of the game. Essentially, in order to create a single powerful item you need to combine several items, each of which was made from several items, some of which were made from a few more items. In other words, unless you have a stockpile of pre-combined items, you may need to go through several stages of item combination in order to get a finished product. Also, this entire process breaks down if you are missing a single component of any of the combination steps. Creating simple items is easy enough, but making complex ones can take a lot of tedious effort. Essentially, such a system will quickly get bloated and unplayable unless measures are taken to remove some of the complications and middle steps.

The major differences between Ar Tonelico and Atelier Iris 2's item creation systems lay in the different ways they try to solve the complications of the basic system they share, and the different ways they reward mastery of the item creation system.

The main method that Ar Tonelico uses to simplify the item creation process is to make some of the items you create available in shops. This way, instead of having to remake an item every time you need it in a later combination, you can just buy it in a shop and use it without the middle steps. However, the usefulness of this system is flawed, simply because the prices of the items you need to purchase are far too high, and money is too scarce. As a result, the only reasonably economical thing to do is to make each item from scratch, including every component item. The game makes no effort to solve the problem of needing specific items for each step that may be rare or hard to find.

Atelier Iris 2 simplifies the basic system through two methods: Mana Synthesis and allowing several possible ingredients for each recipe. Mana Synthesis is a mechanic which lets you recreate certain types of items very cheaply and easily, without needing to buy them from shops or recreate them through tedious effort. Also, any particular item can be made with many different possible ingredients, so you can choose to make it with either rare ingredients found in chests or bought from shops, or it can be made with cheap Mana Synthesis Items that can be acquired reliably, or some combination of Mana Synthesis Items and normal items. Even some of the most complex items can be made with the same effort as a low-level item this way, so the system is less tedious and frustrating.

Even though Ar Tonelico's Grathmelding system is difficult, heavy use of the system is not very rewarding. Every item can only have one variable ingredient, the Grathnode Crystal, and this only affects the "Quality" of the item, which only matters if you turn that item into another Grathnode Crystal. As such, mastery of the system will allow you to get a few rare Grathnode Crystals, but the money and effort required for this would be better spent just buying items from shops.

In Atelier Iris 2, system mastery lets you actually make the items and equipment you make even stronger. You can add more range or power to an attack item like the "Flame" item depending on what ingredients you choose for it. Then, you can add those same properties to any item that can use "Flame" as a component (such as "Mega Flame"). It is even possible to build recursive loops using different ingredients, using "Flame" items to make "Megal Flame" items, and then using "Mega Flame" to make "Flame", in order to vastly more powerful items. This is much more fun and rewarding than having items you create always remain the same.

As a whole, Atelier Iris 2 has a system that makes it fun and rewarding to mix together items, and Ar Tonelico's system is a chore. Even though the two systems are based on the same framework, the differences have a huge effect on the playability of the game. However, I'm not done with examining this kind of system yet, but more will have to wait for another day.

Wednesday, January 16, 2008

When you give in and check a FAQ

Sooner or later, you give in and look up something about the game you are playing on GameFAQs. Preferably, I do so after I have already beaten the game on my own efforts. After all, FAQs are usually the den of the worst kind of spoilers for a game. However, every once in a while, a gameplay element comes by that demands that I look it up. In my case, it is often involved with collect x  number of items quests or puzzles that are too hard for me to solve on my own. I don't think there is any shame on a player's part for having to look something up (though I hate the spoilers). However, I would not want to design a game where the vast majority of players need to look something up on GameFAQs.

A player reaches for a strategy guide or FAQ when the fun-factor of a challenge is overcome by frustration. This can happen most frequently when a collection quest becomes monotonous, or when a puzzle or challenge becomes so difficult that the player stops having fun trying to figure it out on their own. As with most things having to do with fun-factor, personal tolerances can vary wildly between individual players. However, there are general concepts that are worth looking at.

100% completion of collection quests has to be one of the biggest causes for looking up a FAQ. There is a reason that Zelda and Pok√©mon games have permanent spots on the top ten FAQs list on GameFAQs. The problem with completion quests is that sooner or later, you have scoured every place you can think of, and are inevitably missing 2 or 3 of the items you need. In any collection quest where you are asking someone to look for more than a handful of something, you can expect that no matter how easy it is to find the objects, someone might just miss one and assume that they thoroughly swept the area.

There are two ways to control how much someone needs to consult a FAQ for a collection quest. First off, a developer can limit frustration by requiring the player to collect less than 100% of the items. This doesn't stop the problem completely, but it does put off the problem of needing to track down the couple you missed until later, usually the post-game period. Of course, there should be an additional reward for 100% completion, to avoid the player feeling like their effort was fruitless. However, this approach doesn't work for all kinds of collection. The second way is to limit the benefit of a FAQ by giving the player hints of some kind in the game. A pair of examples of this are in The Legend of Zelda: Twilight Princess. In that game, you can purchase clues to the general location of heart pieces from the fortune teller. This makes tracking down elusive heart pieces much less tedious. However, I thought its explicit hints might have made the quest too easy overall. I actually prefer the implicit hints found in the golden bug hunt. The implicit hints for the golden bug quest came from their pattern: the bugs came in pairs that were always found in the same general area, and were scattered somewhat evenly across the game world. A pattern like this can work just as effectively as a blunt hint with the added bonus of being fun for the players as they uncover the pattern. The trick is to make the pattern consistent and logical enough to figure out and be extrapolated.

Making it so that puzzles don't send players scrambling for a strategy guide is a trickier problem to balance. If you make the puzzle too easy or the hints too obvious, then the puzzle doesn't challenge the player, which means the player probably is bored. However, if the puzzle is too hard, then it sends players straight to GameFAQs. Figuring out how to put a puzzle in the butter zone is tricky. However, the key often involves a gradual stepping up of difficult of puzzles. Start with simple puzzles that teach the concepts to the player, and then move up to increasingly complex variations. If the player is familiar with all of the basic elements of a puzzle from past experiences, then they are more likely to be going to be able to solve it on their own.

Hmm, I should probably talk more about some of these topics in later posts.

The Items of Legend of Zelda

Well, I am pretty much at the end of Legend of Zelda: Phantom Hourglass, so I guess I will write about one of the most important elements of any Legend of Zelda game: the various tools you use to do stuff. These are the treasures you get in dungeons that let you find more treasure, and thus play a central role in both gameplay and the reward mechanics. However, not all treasures are equal, and there are a too many to describe all of them in a single blog post, so I will just list some of my favorites and least favorites, in no particular order.

Favorites:

1) The Boomerang (original and Phantom Hourglass versions)
This item has been in pretty much every Legend of Zelda game, as part of the omnipresent Bomb, Bow, and Boomerang trinity. However, in most of them it is completely useless outside of a limited number of situations. It tends to be limited by its role as a stunning weapon, which is of minimal effectiveness in many Zelda games (only being essential in the difficult battles of the NES games), and by the fact that other weapons tend to kill or stun more easily than the boomerang does. However, the ability to guide this weapon's trajectory in Phantom Hourglass, not to mention its quiet impact sound, makes it a useful tool throughout the game, for both puzzles and battle against multiple opponents or opponents you need to hit around corners or from behind, so it never gets replaced by the more lethal Bow.

2) Bombs (Twilight Princess and Phantom Hourglass)
If any Legend of Zelda tool was perfected on its initial introduction, it is would be the Bomb. It is simple and effective, is essential in puzzles, and can be useful in combat throughout the game. The way Twilight Princess lets you choose special types of bomb for each Bomb Bag is a nice addition to the classic tool, and Phantom Hourglass is the first game where using a bomb as a weapon is very easy (a side effect of making it very easy to hit anything with a thrown object). This is pretty much the benchmark for a good Legend of Zelda item.

3) The Grappling Hook (Phantom Hourglass)
The Hookshot and its many variants have been in many games, but they are outclassed by this new item. Beyond the simple gimmick of pulling Link towards something, the Grappling Hook can also be used to create bridges, launch Link like a slingshot, and pull two connected objects towards each other. What is more, it does not rely on just set targets, but can be used in combination with mundane objects like rocks, stone tablets, and torches. Like the best of the hookshots, it doesn't just open up obvious paths, it makes you look at the terrain in a completely different way.

4) The Ocarina of Wind (Link to the Past and Minish Cap)
Yes, the Ocarina of Wind, not the Ocarina of Time. Teleportation between different places in the game is not important for puzzles or battle, but it is important for avoiding tedium. Of all the methods of teleportation in the series, this one is my favorite simply because it is quick and easy to use.

5) Empty Bottles
The simple usefulness of just being able to carry various liquids (such as healing potions!) and small items, and especially various odd things that are just floating around or laying on the ground, makes these simple glass containers a great item. I guess I should have said Bomb, Bow, Boomerang, and Bottle earlier.

6) Wolf Senses (Twilight Princess)
The various "truth" items that reveal hidden things are all god items, but the Wolf Senses manage to surpass them by just being much more cool and fun, and by letting you follow scent trails.

Mixed Opinion

1) The Bow
Amazingly, the bow is not on my list of favorite items, despite being essential to every Legend of Zelda game. I guess the problem with the bow is that it tends to either be difficult to use and underpowered (in the 2D games) or overpowered to the point of pushing aside other items (in the 3D games). The Bow also tends to have a less consistent niche in puzzle-solving (other than the relatively boring role of target shooting) and no outstanding examples from any game. I want to rate it highly, but I don't think I can.

2) Ice and Fire Rods/Arrows
I do like these special arrows, and they are useful, but their combat role overlaps with the role of the Bow (direct liner attack) far too much, and few games actually use these two items to their full potential in puzzles. The Ice Arrow/Rod in particular has had a history of poor showings, and few good puzzles, despite having a solid concept. I guess they just need a slight re-imagining like the Boomerang received.

Least Favorites:

1) Roc's Cape, Zora's Flippers, and Power Bracelets
In the end, these two items just enable Link to do things he should be able to do normally and is able to do freely in other games. I am glad these have been appearing less and less as the series progresses, though they still turn up.

2) Iron Boots (Ocarina of Time)
Twilight Princess showed that these boots can be fun in some situations, but in the end they are wildly implausible and any situation in which they make sense would be better served by another item (like proper swimming gear or some other way of anchoring to the ground).

3) The Spinner and the Rod of Domination (Twilight Princess)
These two items are very fun and useful, but are completely useless outside of the dungeon they are found in. What few places they are useful are few in number and blatantly obvious. I think they are just poorly thought out items for a Legend of Zelda game. Both would be better off as unusual terrain features of particular dungeons, rather than as items.

4) The Clawshot and some Hookshots
While I praised the Hookshot and its variants above, there is a darker side to those items, Namely, any variant in which you can only target very specific and obvious targets. One of the greatest strengths of the good versions is their ability to open up routes for creative players, but these versions don't have that potential for creative use of terrain, and become boring as a result.

I think that is it for now... I might talk some more about the other items that I did not list here at a later date.

Monday, January 14, 2008

Phantom Hourglass: Temple of the Ocean King

On various message boards I frequent, I have been hearing multiple people complaining about the Temple of the Ocean King, the central dungeon of Legend of Zelda: The Phantom Hourglass. I briefly mentioned it before in my post about the map feature of the game, but I think I should talk about it a bit more since I have gone quite a bit deeper into it over the last week.

Despite some of the complaints I have heard, I think that the dungeon is actually a lot of fun. It presents a very different experience than the typical Zelda dungeon. First, the Temple is filled with a curse, so the player is forced to play through the dungeon on a time limit. Furthermore, it is populated with the Phantoms, nigh-invincible sentries who can kill Link in a single hit, draining away his precious time. In order to balance out these challenges, the dungeon has numerous safe zones where the Phantoms can't reach Link and the timer doesn't drain. So, Phantom Hourglass plays like a tense stealth game like Metal Gear inside the Temple. However, another dungeon in the game uses the same curse/safe zone set up, and plays much more like a regular dungeon. It is something else that sets the Temple apart.

In a normal dungeon, once the player clears a puzzle or challenge, he usually never has to clear it again. However, the Temple resets itself every time Link leaves, and thus the player has to go through the entire dungeon over again every time he enters. To many players, this is apparently the most frustrating aspect of the dungeon. However, it is also the greatest strength of the dungeon in my opinion. Even though the dungeon resets every time Link enters, the experience is never the same twice. Ever floor has multiple ways to clear it, depending on what tools Link has at his disposal. Furthermore, there are numerous hidden treasures scattered throughout the Temple to be found by a dedicated player. No other dungeon in the history of the Zelda series has had this many short-cuts, mysteries, and hidden treasures. With the right short-cuts and fore-knowledge, it is possible to clear through the entire Temple in a matter of a few minutes.

However, while I find the dungeon fun, I can't ignore the fact that there are other people who find the dungeon frustrating and annoying. Why is this the case? I have to admit, I always feel a little hesitant to start a Temple expedition. Part of the problem is that the Temple is very large. A dive into the deeper parts of the Temple can take half an hour or more if the player isn't making good time. In a console game, this aspect isn't very uncommon. The Legend of Zelda: Twilight Princess has a 100 floor Cave of Trials after all. However, Phantom Hourglass is a portable game. Portable games should be easy to pick up and play for a few minutes before stopping suddenly. A super-dungeon like the Temple of the Ocean King is too big for that kind of play. Nintendo should have added a better save feature to let players keep their progress between game sessions (or at least a suspend option). More than one mid-point return warp at the very least.

I think that increasing the number of paths, and varying the challenges presented to the player more, would have helped. The strength of the Temple is that it is a dungeon that the player will play through multiple times across the length of the game. Often, floors are limited to the "original hard route" and the "later easy route". Increasing the number of routes, particularly on the most-commonly seen first several floors might have helped.

I am hoping that Nintendo revisits this kind of dungeon in future Zelda games. Preferable console ones.

Sunday, January 13, 2008

Super Mario Galaxy: Two Player Mode

Right now, my brother is going through Super Mario Galaxy. Since I like cooperative games and I would be watching him go through the game regardless, I decided to try out the two-player mode in that game.

The two-player mode in Super Mario Galaxy is an interesting novelty. It is very different than the alternating cooperative/competitive gameplay of the much older Super Mario games, and is a very refreshing change from not having any multiplayer support in the previous 3D games. The only other game with a cooperative multiplayer mode that really compares to it is Rare's old Jet Force Gemini for the N64. It is rare to have a multiplayer mode in which one player has a completely secondary role to that of the other player. While other games may limit the control options for any player other than the first (such as multiplayer RPGs), this is usually only for certain sections (such as movement between battles) and all players will have equal control over the game during other times.

Having one player be simply play a supporting role has a number of advantages. It removes the need to have two autonomous characters running around at the same time (which can be very hard to implement well, especially in 3D games). It makes it easier to simply add a multiplayer mode onto an existing single player mode, without resorting to the old system of alternating play. It also is a way of rewarding cooperative play by giving a slight increase in the character's ability; two players can do more with one character than a single player can.

Still, this implementation of cooperative games has a major drawback: it can be hard to make it fun for the second player. If the support role is not interesting, or feels redundant, then it becomes just a chore for the second player, and the value of having a cooperative mode is lost.

Overall, the way cooperative multiplayer works in Super Mario Galaxy in particular is imperfect. The second player functions as a hand to grab and throw Star Bits, and can also make Mario jump or spin midair. The biggest problem is that, despite the fact that the floating hand is the most important part of the second player's abilities, the first player still has that full potential at the same time. What is more, the first player can grab pull-stars, shoot Mario with those sticky pod things, and feed Lumas with his floating hand, as well as a few other things, but these options are unavailable to the second player. Rather than being essential, the second player is mostly redundant even in the cooperative mode, simply because the first player never has to depend on the second player to do something. It would be much better if the second player had full control of the floating pointer with no control over Mario, and the first player had full control of Mario with no control over the floating pointer. That situation would make the second player feel more useful, and make his role more consistent.

On a side note, I don't particularly like the feature of letting Mario jump higher if both players make him jump at once, simply because coordinating two people to do something at the same time is nearly impossible, especially in a pinch. The first player is better off just doing a backwards somersault or triple jump, rather than try to pull off such tricky timing. On the other hand, just having a second pointer on the screen is a great tool for trying to point out something that the first player didn't see, or to help point out a direction to go, which has often been difficult to do with words alone, so I approve of that feature.

I have had a lot of fun playing the game with my brother, but a lot of my fun with the game is because my brother voluntarily leaves a lot of things to me that he could do on his own otherwise.

Saturday, January 12, 2008

Power-Ups: Limited Use vs. Free Use

One thing I have been noticing about Super Mario Galaxy, and to bit less of an extent its predecessor Super Mario 64, have a significant departure from the old 2D Mario games in terms of how power-ups work. In the old Super Mario Bros. games, power-ups are what I will call "free-use". Most of the major power-ups, such as the Power Mushroom, Fire Flower, and Feather, appear all over the game. Any given stage might have them, and the player can keep these power-ups indefinitely as long as Mario isn't hurt, and between stages. On the other hand, Super Mario Galaxy uses a set-up that I will call "Limited-Use". In Galaxy, power-ups are only found directly next to places where you need to use the power-up to progress, and cannot be carried on to other stages. Of course, these terms are a simplification of a much larger number of possibilities. Free-Use power ups can also refer to powers or weapons in a game that the player can use at any time or anywhere, while Limited-Use can also refer to powers that can only be used at specific locations.

The advantage of Limited Use power-ups is that it gives the level designer the ability to have almost unlimited control over how the player approaches a stage. For example, the game gives the player the ability to fly using a certain power-up. Flight is always an extremely powerful ability. If a player is allowed to fly at any point in the game, it might become possible for the player to bypass most obstacles or even reach places where the designers didn't want him to reach. However, if the player can only fly by collecting a certain item that appears in a specific place and limiting how long the player can use the power, then the designers can prevent the power from becoming a problem. Movement powers such as high jumps or flight tend to be the most common examples of powers that are limited.

However, while the design philosophy of Limited Use has its uses, I have never liked it that much. First off, if most powers in a game are designed to be Limited-Use, then it often becomes obvious as soon as the pertinent power-up appears that you need that power in order to progress through the stage. For example, if you find a flight pad and Red Feathers in Banjo-Kazooie, you obviously need to fly (usually along a route that has Red Feathers) in order to get to your objective. While this can be a useful tool for helping players along in easy games, it can be very predictable. I find it boring stage design myself. In some cases, such as the use of limiting abilities such that they can only be used on specific pads, it can become intrusive to the game and obnoxious.

The greater problem with Limited-Use is that it prevents the player from having fun just paying around with powers. Sometimes it is just fun to be able to play around with an ability, or to use it in unexpected areas to see if you can find something unexpected of hidden. Often, when there are powers that are limited use, the game designers won't make them usable except in places where they are essential to complete a task. However, this means that designers will often to forget to give the player access to some abilities most of the time. For example, Fire Flowers are apparently a very rare thing in Super Mario Galaxy. However, they would not cause any problems in the form of letting the player get into places he should not be, nor would it upset the difficulty of the game too much. Their rareness is inexplicable, and is a bit of a lost opportunity.

The reason I prefer Free Use design is that it opens up the possibility of the player having more than one way through an area. For example, say that a certain power-up can be found in Stage A, but not in stage B. If the player can manage to not lose the power-up after clearing Stage A, he might be able to find a short-cut or hidden reward in Stage B using it. In this sense, Free Use design opens up more design space and rewards the player for playing around with an items powers in various areas. There are actually quite a few examples of this in various Super Mario Bros. games. 

While Limited Use does make design work easier, I think Free Use creates a more rewarding game experience. While it is easier for the designer if you don't have to worry about a player jumping or flying over a wall you don't want him to cross, there are always ways to solve these problems without eliminating the freedom of the player.

Thursday, January 10, 2008

Megaman ZX Advent Part 2

Game Completion: Completed, with only a few optional things remaining.

Overall, this game is a solid entry in the Megaman series. It won't change the mind of someone who doesn't like the series, but it is very satisfying for a fan of the series.

The plot of this game is adequate. It is not a great piece of literature, but it has a few good characters and the plot suffices to explain the events of the game and maintain the player's interest. Certainly, focusing a lot of the game on the interesting characters of Prometheus and Pandora helped it a lot, since those two are very different from the typical enemies of the Megaman series. I guess here is a list of a few good things and a few bad things.

1) Legion and the Sage Trinity are a nice addition to the series, though it would have been better for consistency's sake if they were at least mentioned in the last game.

2) The main villain is interesting, though he probably could have had more presence in his role as a supposed ally before his big revelation that he was the evil mastermind.

3) The portrayal of Model W in ZX Advent is very different than in the last game in many ways, and this is not adequately explained. Of all the things that seem like plot holes, this is the only one that is problematic for me.

4) The idea behind the four Megamen fighting in the Game of Destiny is a good one, but it wasn't executed perfectly. Their motivations and goals were not presented clearly, and more importantly they did not have enough presence in the game. It would have helped a lot if they simply appeared in more scenes, and had clear objectives whenever they appeared. As it is, they seem more like common minions than a group of idealists fighting some Highlander-style battle for ultimate power.

5) The power of Model A is clearly based on Axl's abilities from Megaman X7 and X8, but the connection between Axl and Model A is obscured by a lot of unusual stuff, and it seems that a few potential plot threads tracing back to Megaman X8 are still unresolved. I suppose more might be explained later, but it just doesn't make a lot of clear sense. Oh well, I guess this is fairly typical for the Megaman series, actually.

6) Prometheus and Pandora are great characters, so much so that the one fight against them is not enough. Fighting them each at least once more would have been nice. I fact, it would have been better if there were battles against all of the other Megamen.

Finally, there is one last issue with the game's plot: Ashe's story does not make sense. It was a good idea to differentiate the two possible main characters, but in this case the plot, particularly concerning the character's main motivations, just make so much more sense with Grey than they do with Ashe that it significantly changes the enjoyability of the story. Even the villains have better dialog when talking with Grey.

On one final note, I really like the system of collecting different medals for defeating the Pseudoroids in specific ways. It is a fun challenge that makes me want to fight the bosses several times. I vastly prefer it to the system in Megaman ZX, which punished you for fighting intelligently against an opponent and did not really reward fighting a boss any time other than the first.

Wednesday, January 9, 2008

Mario & Sonic at the Olympic Games (Wii)

I have been playing a fair bit of Mario & Sonic at the Olympic Games over the last week. Still not enough to unlock anything, but I have gotten a fair feel for most of the basic events. I think the game is a bit hit and miss in its controls and difficulty so far.

So, without further ado, is a breakdown of my thoughts by event:

100m Dash- This one is pretty basic. The first time I tried my hand at it, I lost badly, but I have since won some races, so I think it has a decent enough difficulty curve.

100m Freestyle Swimming- Now this is much better than 100m Dash. I like the variety of control schemes that each correspond to a different swimming style. It allows the player to find a control they are comfortable with. I, for one, prefer the butterfly so far. The addition of a timing element with breathing to control stamina recovery also adds a bit more to the event. It is the only event I have a world record in. I think it is one of the best events in the starting set.

Long Jump- Not bad, but it seems rather difficult to get a good score in it. I like the simplicity of the concept behind it.

Javelin Throw- It is a lot like Long Jump in control, but the greater precision to its scoring based on the differences in distance between the sports gives the player a better idea of how well they are doing. Its one weakness is that it is hard to tell how close your character will land to the foul line, so it is way too easy to foul out.

Hammer Throw- My brother has an Olympic record already. I like it the event overall. The only trick is that it takes a few tries to get used to swinging the hammer right.

Trampoline- Another good event. The mix of input timing and inputting sequences of buttons works really well. It makes me wish for a greater representation of similar gymnastic or diving events overall. A simpler event where you only had to worry about the command input and not the timing would have been nice to have. Maybe floor gymnastics?

Skeet- This event can humble someone who thinks he is a good shot. The skeet are fast. However, it is a good challenge for someone to get better at shooting with the Wii remote. I like the addition of the timing element to represent the shooter trying to stay calm. I need to try my luck at this one more.

Table-Tennis Singles- This is where the trouble starts. Table-tennis is hard. I am someone who has played videogames for years, and was passably good at Wii Tennis, but this game seems impossible. Unfortunately, bad controls may be at play. While the basic idea behind the controls seems to be good, they are not responsive enough. Particularly, the backhand seems to not work well. Trying to do an accurate rendition of a backhand hit causes the character to hit way too early. Compared to Wii Tennis, there is little time to react. And the opponents are too strong for a beginner to face and hope to win.

Individual Epee fencing- This is the one event where the game design just seems poor. The designers tried to create a system based on a rock-paper-scissors approach, where feint beats parry, parry beats thrust, and thrust beats feint. However, only thrusts score points and feints are useless. Most of all, a system like that doesn't really work in an environment where the players are moving around and can attack freely. It devolves into an exercise of spamming parries until you opponent attacks, or just spamming attacks, trying to futilely guess what your opponent is thinking. Luck seems to be a bigger factor than skill.

If I was designing this event, I would have based it more on the old quick-draw minigame from the old Kirby games. Once the signal is given, both players thrust their remotes forward, and the faster player wins a point. From what I have seen, real Olympic fencing matches seem to go that fast. Variations using different movements could be used in other fencing events such as saber.

I will post more thoughts some other time.

Tuesday, January 8, 2008

Old Favorites: Chrono Trigger's Dual Tech System

Since I can't think of anything from the games I am currently playing to write about, I guess I will borrow my brother's idea of writing about older games that I like. In this case, one of Square's last RPGs for the Super Nintendo: Chrono Trigger. I can probably write a lot about the many things I like about that game, but I will start with the Dual and Triple Tech system.

In Chrono Trigger, each of the seven characters has a list of only eight special techniques they ca use in battle other than the basic attack and item commands. These lists are fairly specialized for each character, with each character being (with a few minor exceptions) being limited to a single element and a limited number of roles (such as healing, physical attack, or magical attack). Also, it is impossible to alter this list or customize the characters in meaningful way, so your only choices of how to customize your team is through choosing team members. However, because of the way Chrono Trigger embraces that concept, the system is not limiting at all.

The biggest strength of the Chrono Trigger system is the way it makes the classic idea of choosing a few people for a battle team from a larger pool of characters and turns it into something far more than just picking characters to cover every role that you need. In a typical RPG that allows character choices, so long as you have the basics of survival (healing and defense) and attack covered, character choices do not depend on each other. For example, in Breath of Fire 2, choosing between Nina and Bleu, who are both mages who use attack spells, is simply matter of personal preference, and won't affect the rest of your team. However in Chrono Trigger, while both Robo and Ayla are primarily physical attackers, the choice between them is decided by more than just personal preference, because of the Dual Tech system.

With the Dual Tech system, combination of character affects your team's abilities just as much as character choice itself. For example, the choice between Robo and Ayla when Crono (lightning attacker) and Marle (ice healer) are already in the team. Choosing Robo not only adds his own physical attack abilities to the team, his Dual Techs with Crono enables powerful wide-area magical attacks, and his Dual Techs with Marle brings some very strong full-party healing moves to the team. On the other hand, Ayla's Dual Techs with Crono enable powerful single target lightning attacks and wide area physical hits, and her Dual Techs with Marle give a powerful stealing ability and strong ice attacks. The choice between Robo and Ayla affects the abilities of the other characters. With certain combinations, this lets characters do things as a team that they can not do individually under any circumstance (such as ice mage Marle and fire mage Lucca using Shadow elemental attacks).

In essence, Chrono Trigger sacrifices individual character flexibility in exchange for much greater flexibility of the party as a whole. Choosing the party is the very same thing as customizing your characters. It embraces the idea that choosing members for a team is an important choice, and has an elegant system for rewarding good choices in party composition. I think this is a huge improvement over many other RPGs, where the choice of team members is redundant because of the freedom to customize characters individually (such as Final Fantasy 7 or 8, where you choose a party of three characters out of a pool of 6-8, even though each character is infinitely flexible in equal ways), or where the choice between characters is no different from merely choosing a list of abilities you want to bring to a battle.

Unfortunately, the concept of a deeply integrated system of character synergy affecting team selection has not been imitated much by later RPGs. A few RPGs have used it to a limited extent (such as the Unite attacks in the Suikoden games), but they do so more to encourage specific plot-related combinations of characters rather than provide a way of making every character combination unique. Even Chrono Trigger's sequel, Chrono Cross, failed to effectively make use of such a system. However, the quality of the concept is good enough that I imagine it will resurface eventually.

Monday, January 7, 2008

Memorable Characters in Megaman games

Just half an hour ago, I managed to beat Megaman ZX Advent, after sixteen hours of play. I am very impressed so far with how ZX and ZX Advent continue to build on the cast of memorable recurring characters in the series. Prometheus and Pandora in particular are very interesting characters that I have looked forward to fighting in both games in the ZX series so far. Between both ZX and ZX Advent, the Megaman ZX series has a pretty sizable cast of memorable recurring characters. This certainly has not always been the case in the Megaman series.

In both the original Megaman series and particularly for most of the Megaman X series there has been an unfortunate lack of good characters. In the original Megaman series, the good guy cast was actually very fleshed out. There was Megaman, the main hero, Protoman, his friendly rival, Roll, the female supporting character, Dr. Light, all round good scientist/father figure, and a large assortment of minor helper characters with varying degrees of traction. On the other hand, for most of the series, the only character on the villain's side is Dr. Wily, the mad scientist and recurring Final Boss. It took until Megaman 7 for another villain character, Bass, to be introduced. Yet, Bass's role is an important one. As soon as he was introduced, he became a wildly popular character. So much so that he shared main character status with Megaman in the final entry in the Megaman series, Megaman & Bass.

The Megaman X series was a lot worse off in terms of characters. For the first three games, the only characters of note were X, Zero, and Sigma. For most of the X series, the only other characters aside from these three were various villains that challenged X outside of the regular stage progression who were all killed off by the end of the game they were introduced in (usually as Fortress stage bosses). This list includes Vile, the X Hunters, Bit & Byte, Dr. Doppler, Colonel, Iris, Double, Dynamo, Gate, High Max, and Lumine.  And yet, the characters in this list are not equal, even though they generally played the same role in their individual games. Vile, the first of these characters is still popular even today, while Dynamo is a joke to fans. When Vile was introduced, he defeats X in the intro stage of the game and forces Zero to sacrifice his own life to save X in the first fortress stage. Vile also has a unique schtick: he rides around in a big mech. On the other hand, Dynamo never did anything more than than throw out a couple taunts before you fight him, and never did anything more than be a nuisance. Dynamo also fought in an unoriginal manner, using only beam sword attacks and energy blasts, the same as numerous other characters in the series. Colonel and Iris also deserve special mention. While their role was similar to other one-game characters from the X series, Colonel and especially Iris were able to win a lot of fans thanks to their tragic plot in Zero's side of X4. Good character designs and good story won them fans and appearances in the Megaman Battle Network series, while Double, the more typical opponent from X's side of the plot in the same game, was forgotten.

The X series also had a spotty history with good-guy supporting characters. Amidst the limited number the first four games was Dr. Cain, who was only mentioned in the manual of the first game, appeared in two scenes of the second, only appeared in one optional scene in the third, and was once again only mentioned in the manual of the fourth. The only other supporting character early on was Zero, who twice saved X's life in the first game, was brought back in the second game with a cool optional boss battle and hints at his mysterious past, was a limited use character in the third, and hit main character status that rivaled X's in the fourth game and onwards. Megaman X5 was where some decent supporting characters were finally introduced. Chief among them were Alia and Sygnas. Alia became a major support character for the rest of the X series, handing out hints and advice in the middle of stages, giving out information during cut-scenes, and doing mission de-briefings. Alia (and a few other characters based on her) even became an unlockable character at one point. Sygnas filled out the missing role of commanding officer and gave context to game missions in cut scenes. He was eventually used to give the player mission rankings in X8. Other characters introduced in X5, such as Lifesaver and Douglas, did not do so well because their role was more minor and less clear-cut. However, Alia has had to deal with the stigma of being an annoying character because of some poor game mechanics from the first game she was introduced in.

The creators of the Zero series payed attention to the lessons learned from the original and X series. From the get go, the Megaman Zero games have had a memorable cast of supporting characters in the form of Ciel and her resistance fighters. Ciel was very important to the plot of the Zero games and was always around to talk to, while the recurring cast of resistance fighters was colorful and entertaining. On the other side of things, the Zero games also introduced the Four Guardians, a  group of recurring opponents for Zero who all had interesting personalities, distinctive fighting styles, and grew ever more developed and interesting as the series continued until they eventually became Zero's allies. Even in the ZX series they are still important characters.

So in summation, here are some tendencies from across the Megaman series about what qualities tend to make for memorable and popular characters:

Supporting good-guys:
*Has plot significance
*Appears in cut-scenes in a prominent role
*Fills out a niche or role such as commander
*Is involved in some kind of relationship with a main character
*Has a likable or quirky personality
*Doesn't annoy the heck out of the player

Villains:
*Has plot significance
*Appears in cut scenes on a recurring basis
*Has a distinctive look and fighting style unlike any other character
*Is fun to fight
*Does more than just fight the hero
*Has more motivation than just fighting the hero
*Has an interesting or fun personality

Of course, good character design, theme music, and writing are also important, but those traits are somewhat harder to quantify past "I know it when I see it."