Analysis: Final Fantasy XI

Also known as "Final Fantasy Online", FFXI is, in many ways, a standard MMO. It contains many of the concepts that can be found in other MMOs, such as a linear, static storyline and a reduced focus on "the individual". However, unlike a lot of other online games, FFXI has an undercurrent and a background that I really think could have used some development and focus. It's a game that, like Lost Planet, would do better with a wider range of games to explore the different facets of the setting.

The unfortunate fact about FFXI is that the things I'm about to discuss are largely irrelevant to the game itself. The need for an infinite swarm of players and monsters means that the political and historical setting of the game are irrelevant. Still, I feel that there is enough material to analyze it even if it is set apart from the game it's meant to represent. To that end, much of my information comes from supplemental material, rather than the game itself. In fact, I believe we'll discover that "the game itself" is restricting some otherwise believable content that exists only as lore.

Setting
Final Fantasy XI takes place in the world of Vana'diel, which contains both the "races of light" and the Beastmen. The races of light are the player-character races - the standard Humes, the proud Elvaan, the burly Galka, the cat-like Mithra, and the petite, but magical, Tarutaru. The Beastmen are, by and large, also fairly standard for their role as "human-like antagonists", with the warrior Orcs, the crafty Goblins, the theocratic Yagudo, the hive-themed Antica, and so on.

Initially, this seems like a very cut-and-dry setup, with the "good" races united against the "evil" races. However, both groups are more than they seem. The races of light, for example, have only recently united, and in the past they were fractured and divisive, warring amongst themselves for national glory. It is only the uniting of the Beastmen against them in the Crystal War that led to their forced alliance. In addition, the races are hardly perfect, with the same amount of backstabbing, infighting, and personal flaws that you'd expect from real humans. The "Beastmen" are hardly pure evil, either. In the past, they interacted with the other races at roughly the same level - to an Elvaan, a Yagudo was no more "beast" than a Hume was. However, offenses committed by the "races of light" have led to these Beastmen to unite against them. For example, the turtle-like Quadav were a fairly isolationist race until the Humes and Galka aggressively mined into their territory. The Yagudo, similarly, believe the other races to be encroaching on ground they think is rightfully theirs.

In essence, this is a story about ten or so different races that came together into two major groups. On the other hand, the way the story is presented initially suggests that it's the other way around - two groups that are made up of five or so races each. In terms of believability, it seems far easier to accept the former than the latter, because the former is built up on relationships and alliances, and the latter is a label applied to barely-developed cultures. One is a label applied for game reasons, and the other is a label applied for more natural reasons.

It's the same with any historical alliance. In World War 2, you could say that you've got the "Allies" and the "Axis", each with their own member nations - or you could say that a bunch of independent nations came together for their own personal reasons and formed two major powers: the "Allies" and the "Axis". The former suggests that their group identity is more important than their individual cultures, and the latter suggests the opposite. The former is better for gameplay, because it creates two solid and unassailable factions that can war against each other, and the latter is better for a believable world, because it doesn't make sense for things to be so cleanly divided. It's a pure game construct to justify two wholly static sides.

Background and History
The Crystal War is the defining conflict of FFXI's recent history, and as mentioned above it shapes the player's view of the conflict by making things very black and white. However, delving a bit further back into Vana'diel's history, there are a lot more "petty conflicts", or conflicts for purposes like greed and expansion rather than good and evil. Each race has had its own empires and its own wars, for whatever reason. This comes back to the black-and-white concepts described above: are these nations just there to be part of the "good versus evil" concept, or are they composed of individuals with their own desires and ambitions?

For example, Vana'diel's history has included:
- infighting among Aztec-themed Taru tribes in ages past.
- an alliance between Elvaan and Quadav to defeat Tarus and Mithras.
- Humes using superior weapons to soundly defeat the traditional Elvaan knights.
- a Taru version of Alexander the Great who was accused of insubordination and stripped of his rank.
- a Galkan captain who led raids both against other "civilized" nations as well as Beastmen and pirates.

This is the kind of stuff that never comes up; it's comparatively natural politics, with different motivations and cultures being involved. Compared to the "Good guys versus bad guys" main plot, it's far more intricate and believable. It's comparable to human history in that way. The whole "every race is its own solid empire" thing is a bit troubling, but there's at least hints of dissent that force cooperation and unity. Wars are fought over territory, over different belief systems, and over pure material wealth, but they're rarely fought for an enforced "right versus wrong" concept. Even the alliance of the "light" races is more comparable to the Holy League, with the Beastmen taking the role of the Ottoman Empire as foreign invaders from a different culture/religion. While the invaders are in the picture, everyone's united under their shared faith and belief, but you know as soon as that threat's gone they're going back to killing each other for selfish reasons.

Also, like Final Fantasy Tactics, the wars in FFXI are meant to be on a much larger, more conventional scale than the "adventurer" concepts established by the actual gameplay. Wars, at least as represented by background material, exist on a much more grounded level than combat in the game. One battle consisted of Hume musketeers holding a pass against Elvaan cavalry, which is grounded in logic that simply doesn't exist in the game. Things like "levels", "special abilities", and "stats" are not addressed in the same way that they are in gameplay. This is best represented by the game's opening movie, which is meant to be more like "real combat" than the gameplay is in terms of movement, strategy, and the amount of punishment an individual can take. This isn't reflected in gameplay, but if that movie reflected what was happening in gameplay it would look preposterous. In the same way, life for citizens in the game is not the same as life for player avatars - compare the description of Bastok and its economic/political situation to the infinite playground of the player character.

In short, from a historical perspective, FFXI wants to be believable and grounded - not, perhaps, "realistic", but fantasy is fantasy. In comparison, the game itself is bound by its own nature, relying on a lot of meta-concepts and additions that are simply not acknowledged by the world at large. Now, I'm being a bit unfair here, because things like "good and evil are not that simple" are technically brought up in the game, but the issue is that for gameplay reasons they're never important. The Beastmen, with the exception of some goblins, are always hostile, forever. This is because they need to serve as sources of experience and money for players. The background, in contrast, indicates that the situation is a lot more morally grey, leading to concepts like interaction and diplomacy that just don't fit in the confines of the gameplay.

Presentation
Weapons and armor in FFXI can range from fairly reasonable sets of cloth and mail to more unrealistic designs, whether in terms of material or in terms of...everything. It's hard to say where one ends and the other begins, so in general it's easier to say that FFXI's aesthetic is "vaguely realistic" and leave it at that. However, a lot of the more grounded designs, I feel, can be ascribed to Mitsuhiro Arita, who provides a lot of the artwork for the game's supplemental material. For example, this article from the Vana'diel Tribune establishes believable armor layers, including different parts, and generally suggests a much more believable design (in terms of materials and in terms of its sensibility) than is established in the game.

Mitsuhiro Arita also provides the illustrations for the Vana'diel Profiles, the historical biographies that provided a lot of the information I analyzed above, such as the pre-Crystal War stuff. The combination of writing and artistic style is what really sells them, for me - it feels more like a historical record than a JRPG plot. They operate on different rules: FFXI is power-hungry high fantasy, while the background material is much more grounded and connected to some sort of reality. Again, it's all the necessities of gameplay. Besides obvious things like graphical and content limitations, FFXI as a game has to deal with the fact that it's essentially a theme park. Players need to have unlimited, unhindered access to content, they need to advance in a very drastic and direct way, and they need to have things to kill so they can move up in the world.

Essentially, that is the issue. Final Fantasy XI has an interesting world, but that world cannot co-exist with the necessities of gameplay. Perhaps if it had been done as something else - a tabletop RPG, for example, or even a miniatures wargame - it would end up differently. Of course, the parallels with Warhammer Fantasy become more clear at this point, but while Warhammer had more grounded content as the majority of its material, with Warhammer Online being the exception, Final Fantasy XI has the opposite situation. In essence, the world that the game presents is defined by its gameplay.

This was also an issue with Final Fantasy Tactics, but in general I felt that FFT's squad-based approach and potential for perma-death made it at least a little less of a problem. In FFXI, the game's setup is so alien to its background material that they might as well be entirely different. Like FFT, the background material of the game relies heavily on the mundane, using simple, understandable historical events in addition to a larger backdrop of magic, gods, and demons. The problem for both is that the RPG gameplay does not correspond with a world where people behave, and are treated like, people.

So, To Sum Up:
1) Final Fantasy XI is a primary example of gameplay influencing and limiting story and interaction.
2) Its good ideas can be ascribed to things that exist outside the purview of the game's mechanics.
3) Its bad ideas can be connected to the necessities of MMO gameplay.
4) If it was set up in a different format that was less reliant on non-realism, it would be better able to explore its mundane elements and eschew its "grindy" elements.

Risk, fear, and mortality

Combat is a major aspect of media. Games bear the brunt of combat-related events, but films and books have their fair share of it. In some cases, combat is reasonably non-lethal (based on unarmed melee combat or something similar), but in the majority of cases, combat is meant to be lethal, either by guns, swords, or magic spells. Injury and death are two very different things. Injury can be healed or recovered from - death cannot. Death is a lesson that cannot be repeated; if a character dies, then they are done.

This is a topic I've covered before, but that was largely based on a concept of avoiding death through different actions. Today, I'd like to talk about reactions to death, or how it's depicted within the universe. There's two major concepts I'd like to establish: "Risk" and "Fear".

Risk refers to the likelihood of a character's death, either in statistical terms (for games) or narrative terms (for books, movies, and cutscenes).  A character with a lot of plot shielding is low risk, as is a character who's not usually in lethal combat situations. A powerful character is at less risk than a weaker character, and a high-level character is at less risk than a low-level character. The importance of risk is that it determines the plausibility of a character dying, and affects how they ought to react. Risk can best be described as the meta-aspect of character death.

Fear refers to a character's reaction to the potential of dying. A character who is doing everything he or she can do in order to avoid death has high fear. A character who quips and snarks in the middle of combat and clearly doesn't give a damn has low fear. Fear is important because it is an aspect of their characterization - if the character feels like he or she is going to die, then the audience should become drawn in. A lack of fear can be seen either as being "badass" (i.e. they don't care if they die) or "irreverent" (i.e. they know they're not going to die) depending on how it's handled. Fear can best be described as the in-universe aspect of character death.

So with those concepts established, let's take a look at this chart I whipped up. "F" and "R" represent fear and risk, respectively, and "-" and "+" refer to an absence or presence of that element. The interaction of these elements can affect the audience's understanding of a character and their condition. Low amounts of either of these elements can result in something being "cartoony" or "rule of cool", while higher amounts makes a work more "serious" (as in, "something that should be taken seriously").

No Fear, No Risk (-/-): This is a character who is immune to everything and they know it. This is the standard "snarky protagonist" who beats up every enemy soldier without a second thought. There's no drama, just a visceral "awesome" feeling from watching a guy be objectively superior to everyone they encounter. The Star Wars prequels were overloaded with this, culminating in Episode 3 when an entire elevator full of battle droids holding Obi-Wan and Anakin at gunpoint were sliced through without even a brief pause for peril or danger. Many people will defend this aspect as being "rule of cool" or some similar concept, but there's absolutely no room for drama here - it's a dolled-up version of shooting fish in a barrel, and its only value is as some sort of revenge or power fantasy.

No Fear, Risk (-/+): This is a scenario where a character is technically in danger, but continues to quip and snark nonetheless, or charges out into the open with both guns blazing. In a fairly balanced game, this character is either lucky or dead. In a non-interactive medium, their death will generally either be a comical waste or a dramatic reveal. "Risk" naturally, can come in several forms - from sudden death in combat to "cursing the author for totally killing that dude off" - but the end result is the same: a disconnect between what's happening and how the individual treats it.

I should clarify that when I say "no fear" I mean that there is no acknowledgment of the risk of death. The individual proceeds as though combat was a game, and they're generally surprised if they're severely hurt (as is the audience, more often than not). A character who acknowledges the threat of death but chooses to ignore it is not the same as a character who treats death as something that "happens to other people".

Fear, No Risk (+/-): This sort of character can be a coward, a self-preservationist, or perhaps even a grimly determined soldier - but they're also not in any danger, regardless of their response to the situation. This is a difficult one to judge; on the one hand, they're displaying believable responses towards combat and the risks it entails, but on the other hand they're not going to be exposed to it for meta-reasons. They might be a protagonist, a title character, or a viewpoint character - for whatever reason, it's obvious they're just not going to die. However, depending on how the situation is handled, a "Fear, No Risk" character can suggest at the possibility of that death happening anyways. In addition, the character in such a case doesn't know that they're not going to die, so maintaining the illusion of risk can be a major factor in whether something is plausible or not.

On the other hand, "Fear, No Risk" can also lead to melodrama if there is a gameplay/story (or action scene/story) mismatch. A character who takes a billion hits in gameplay can be afraid of a man with a pistol in a cutscene. This creates a sense of false drama, where what should be a trivial obstacle is made into a dramatic issue because they're ignoring the thousands of wounds the character has already shrugged off.

Fear, Risk (+/+): The most believable combination, with a match-up of character risk and character reaction. These are characters who could die at any moment and who know it. Their reactions may vary, from cowardice to resignation to defiance, but the drama of the series comes from not only their reactions to a perilous situation, but also the fact that it could actually happen and they're not arbitrarily protected by the plot. Therefore, their sacrifice (though still fictional) has much more emotional attachment to it. Their reactions are made more poignant because, for them, the risk is present and plausible.

Now let's take these concepts and apply them to some existing franchises.

Warhammer 40,000: Because of the grand scale of the setting, all four points are touched upon. "No Risk" is generally reserved for protagonists and book characters (like the main characters of Dawn of War 2), while "No Fear" is generally much more ubiquitous, applying to a broader group. It's probably best to look at it faction-by-faction:
Necrons: No Fear, No Risk. The Necrons have no real humanizing elements to them - they're an unstoppable force that can briefly be delayed, but never fully countered. They're not meant to be sympathetic, they're meant to be boogeymen. The same can be applied to the Tyranids, who may suffer defeats, but whose hivemind status means that the loss of individuals is unimportant.
Orks: No Fear, Risk. Not a huge amount of risk, mind you, but an ork can still die. Orks also happen to lack any fear of death as a racial trait, so while they might get shot or blown up, they don't really care. This is why they're the "comedy race": because they're operating under the same combat rules as everyone else, but the consequences don't matter to them.
Space Marines: Varies. The Space Marines are among the most schizophrenic factions due to their varying levels of power. Protagonist Space Marines are generally "no fear, no risk", while lower-level space marines are "no fear, risk". I pointed out DoW2 because that gap exists: the main characters can be resurrected, but the helmeted generic space marines that serve as their squad members are just sort of left where they die.
Imperial Guard: Fear, Risk. The IG are the "normal human" faction, and as such have the most human response. The IG are thrown into countless unwinnable battles, and their soldiers know it. Thus, the IG are the most "serious" faction, because there is consequence to their decisions and actions. This is in contrast to the Orks, who are more physically powerful and also do not care about losing. A few special characters might fall under "no risk", but the average guardsman is definitely expendable and aware of it.

Berserk: Berserk is a grim, bloody manga - but it still manages to have a cast of characters who are largely immune to death (though they can still be harmed). Therefore, it is "Fear, No Risk" for the protagonists (at best) and "Fear, Risk" for everyone else. Soldiers are slaughtered en masse by monsters or demons or each other, but while the main characters can be pushed to their limits, it's pretty clear that none of them are going to die off unless it's in a major climactic scene. Still, one thing I appreciate about Berserk is that it manages to make the soldiers' deaths horrific and disgusting enough that the "risk" and "fear" are both warranted.

Dragon Age: Dragon Age would like to be "Fear, Risk" judging by the focus on mature concepts and themes - blood, betrayal, death, foul sorceries, and so on. However, it comes out as "No Fear, No Risk" because the game centers on a team of protagonists who can't be killed outside of very specific player choices (no risk) and have conversations like this or like this (no fear). The snarky quips prevent any feeling of trepidation or fear, and never forces players to leave their comfort zones. In exchange, the game offers us romance subplots straight out of a Joss Whedon flick. It's almost impossible for me to take it seriously, because the "seriousness" of the setting primarily comes from blood and gore, not from any real logical framework. Most of this analysis applies to Mass Effect as well, with the semi-exception of the final mission of ME2 (a partial-Fear, partial-Risk situation).

Metal Gear Solid: While MGS generally maintains an action movie mentality, MGS4 features a lot more of Snake getting hurt and mutilated by things in an attempt to gain sympathy for his character. This creates a "Fear, No Risk" situation: he gets shot a million times in gameplay and barely cares, but then he gets stabbed in a cutscene and "oh no he's hurt badly what do we do". The characters try to act like things are super-serious, but the fact that you can heal up in gameplay by eating a ration sort of undoes the pathos of the situation. The guards, on the other hand, are "No Fear, Risk" - they can die easily, but with the exception of the militia in MGS4, they'll charge at you despite the odds. You can blame mind control and nanomachines for this, but you'd think they'd at least try to regroup or reorganize or something instead of continually charging at the guy who's mowing them all down.

Heavy Rain: A "Fear, No Risk" situation for some characters, a "Fear, Risk" situation for others. Finding out which characters have risk and which don't is a matter of experimentation, but on the first playthrough most people will treat it as though every character is equally at risk, which makes it a partial "Fear, Risk" masquerading as a full version.

Bleach: The ratio of "people looking shocked at some major wound" to "good-guy deaths" is currently hovering somewhere around 500,000 to none, so Bleach is a pretty solid "Fear, No Risk" situation. This does not apply to the main character, Ichigo, who doesn't really care about the potential for dying, and thus is solidly in the "No Fear, No Risk" category. This is meant to make him seem more powerful and courageous. Whether or not it works is arguable.

Every Superhero Comic: "Oh no, [x character] is dead!" If [x character] is a main character, resurrect him or her after a short period of mourning. If [x character] is a minor character, leave them dead so you can illustrate how totally awesome whoever killed them was. Rinse and repeat as needed.  A "Fear, No Risk" scenario because of the reactions of the characters in-universe.

To Sum Up:
1) Whether combat is "serious" or "lighthearted" is based on potential for death and reactions to death.
2) A situation where neither fear nor risk is present is an exercise in "rule of cool" at best.
3) A situation where risk is present, but fear is not, is an awkward setup that suspends disbelief regarding the character's personality unless some explanation is offered.
4) A situation where fear is present, but risk is not, will vary depending on how obvious it is that the character is not going to die.
5) A situation where both fear and risk are present matches up the best in terms of conveying drama and the potential for death, because both the meta-universe and the character acknowledge danger.

Analysis: Final Fantasy Tactics

Probably the most serious and down-to-earth Final Fantasy game, Final Fantasy Tactics is a thematically divided concept. On the one hand, it's a very grounded story about a civil war of succession tearing apart a country, with a side helping of religious corruption and intrigue. On the other hand, it's a Final Fantasy game with chocobos and mages and adorable noseless character sprites. Essentially, I think the development team did the best job they could with the tools they had, which is why FFT remains my favorite Final Fantasy game to this day.

Thematically, FFT can be divided into the "story" and the "game". These are not very different, as is the case for many other games, but it's still different enough to be noteworthy. In essence, both are considered to be part of the "canon", and this can cause some minor contradictions and issues.

Background
Final Fantasy Tactics takes place in the Kingdom of Ivalice, which consists of multiple duchies not unlike a real kingdom. Ivalice's most important recent event is the Fifty Years' War against their neighbor Ordallia, which was costly to both sides. Following this, payment could not be delivered to the soldiers who had fought in the conflict (an event not dissimilar from the Shogunate's inability to pay samurai after the failed Mongol invasion of Japan). This led to rebellion and an increase in banditry throughout the kingdom.

This trouble came to a head when the weak and ineffectual King Ondoria Atkascha died, leaving the throne torn between his infant son-by-blood Prince Orinus and his adopted daughter / half-sister Princess Ovelia. This dispute, largely used as a way for nobles to gain power themselves, resulted in the War of the Lions. Behind the scenes, however, the state church of Ivalice has its own agenda, and is manipulating things to its advantage.

In essence, FFT's plot is largely political. The protagonist's actions deal not with these disputes, but instead with the dealings of the church, who are attempting to gather ancient stones and unlock a sealed demon. Everything  described above essentially exists in the background except where protagonists and their actions are concerned. In fact, most of the War of the Lions is only shown when it intersects with the "find all the macguffins" plot. I can see why they did this, but it also feels like a bit of a waste. Still, the fact that they don't really discuss either conflict in-depth allows the player to imagine what must have happened given the descriptions in the game's libraries.

The background is one major thing that I like about FFT. It's a war of kings and successions - the mundane nature of the strife makes it more believable. Even the "sealed demon" plot is a bit more mundane than most, connecting more into the intrigue and drama of a tangled social web rather than simply being "get all the stones, kill the demon, everything's fine". However, the actual low-key elements are ignored in favor of the comparatively high-fantasy, high-magic plot.

Setting
Unlike the later incarnations of Ivalice, the world of FFT holds humanity as the only "sentient" race. Monsters and other creatures still exist, but they are dealt with primarily in the form of random battles (i.e. they waylay travelers). Humans are the only species who have any sort of culture or infrastructure. To me, this gave it more of a direct historical bent. There were no wacky shoehorned races with a single defining characteristic - it's just human interaction.

FFT works on a sort of waypoint map system. It takes a day to travel between "waypoints", and a waypoint represents a general area - a forest, a marsh, a city, a castle, or whatever it needs to be. This suggests a great deal of distance, meaning that 99% of the game world isn't visible and can be assumed to hold farms, mines, and all the other necessary components of a functional society. This is further reinforced by the inclusion of "propositions", or mundane jobs like mining and salvage that your subordinates can be sent off to complete.

One thing that I don't feel is explored enough is the role of magic. In FFT, magic is fairly common - anyone can become a mage given enough practice, and the only restriction on spells is based on "MP" (no material costs and so on). However, there's no real sign of this being a big deal within the setting. For example, despite the presence of mages who can cast healing spells, the country's recent history includes a deadly plague that killed many people, including the parents of several protagonists. Still, there's no explicit detail given either, so it's possible that there were other issues at play.

For obvious reasons, the "party versus party" aspect of combat in Ivalice gets the most coverage. Each class is able to pull its weight fairly well, but like many other games the "normal" humans become unrealistic to make up for this, eventually becoming able to give and take far more punishment than a real person could. In essence, every class is equally "magical", and there's no normal humans within the gameplay. Larger scale combat is given much less attention. It is mentioned that the wars have produced many casualties, but other than that the details of warfare are left to the player's imagination. Still, the presence of magic users in both gameplay and cutscenes suggests that there is at least some level of mage-based warfare in addition to the standard medieval setup.

A special note should go to the role of gender in the game. There doesn't seem to be a patriarchal or misogynist culture in Ivalice - women can take any job a man can (with the exception of Bard, because they become Dancers instead), and there are several prominent female characters like the royal knight Agrias and her subordinates. However, the "important" parts of the plot are largely carried out by men - i.e., most of the nobles and high priests are male. There's no point where gender is overtly an issue, but the undercurrent somehow remains.

Design
Final Fantasy Tactics is the most "serious" Final Fantasy game in terms of its design, through the use of color, lighting, and costuming. This does not make it "realistic" - it's just less cartoony than, say, Final Fantasy Tactics Advance (compare the black mages on the right to the black mage from FFTA). As discussed in the last update, the simple use of different palettes can radically affect the feel of a setting.

The armor of the setting is quasi-realistic, but nowhere near "actual" realism - look at the knights above and you'll notice that they basically have a small chestpiece and arm/leg armor. The difference between it and, say, Fire Emblem is in terms of perception, not of design. It's easy to notice Fire Emblem's armor issues, because they're bright and in-your-face. In contrast, FFT's armor design is more muted and thus less overtly ridiculous.

One thing the style does manage to accomplish is to make the materials feel more realistic. Look at the fur lining on the female wizard's coat, or the leather of the male wizard's boots and gloves. I would say that metal is the unfortunate exception to this, especially in the case of dragoons, whose armor doesn't really seem that solid (perhaps due to its coloration). For the most part, though, mundane materials like cloth are well represented. A problem arises when this can call attention to the sillier designs, such as the female squire or geomancer (aren't they cold?).

Generics
The part of Final Fantasy Tactics that really makes it my favorite Final Fantasy game is the "generic soldiers". These soldiers consist only of a name, statistics, and (in the original Japanese and re-released PSP update) a single quote connected to their name. However, their nature as dynamic participants in the gameplay means that they endeared to me more than their "unique character" counterparts. FFT does a pretty great job of taking the few lines they get and trying to take as much characterization as they can get from them. In addition to standard combat roles, generic characters can be sent out on missions (as described earlier), and when they return they offer a short report detailing their success or failure.

Generics are not simply the player's tools, however. They can be dismissed and will be upset by this, or they can leave on their own if they become too cowardly or too pious. Each of these possibilities has a set of quotes associated with it that I found to be fairly impressive (skip to "00quotes" on this FAQ):

Dismissal attempt: Won't you rethink this? We've come this far together.
Dismissal attempt: Are you certain about this? I'd thought us faster friends.
Dismissal attempt : I beg you, do not say such things! I'll prove my worth to you, I swear it!
Bravery threat: Fear gnaws ever at my heart. I do not wish to die!
Bravery threat: I...I'm afraid. I do not mean to be so craven, but...I am.
Bravery threat: I beg you, will you not send another in my place when next we face battle?
Faith threat: I've lost all faith in humanity. Are there none I can trust but the gods?
Faith desertion: I had rather obey the will of the gods than yours.

What I like about these quotes is that they illustrate a consciousness outside of the direct control of the player. These are not pawns to be used, although the gameplay would certainly have you think so. These are human beings who either support the protagonist's cause or value his coin. The player can lead them to their deaths if he so desires, but these simple quotes exist to illustrate that in both self-preservational and moral terms, these "generics" still have their own beliefs and desires. This makes them people, rather than tools, and it would have been interesting to me if the game had been able to expound more upon this.

Conclusion
Final Fantasy Tactics is, to me, a game where a lot of very small, finely detailed touches can change the larger influence of the game. The artistic choices make it feel more serious despite the clothing design not being entirely realistic, the political background makes it feel more plausible despite the main plot revolving around magic stones, and the minor quotes and personality traits of generic characters makes them feel more realistic than a standard "blank slate" character. Its virtues are centered largely around these very small details - the game itself is standard in a lot of ways, but the way it's presented is what it differentiates it. It should serve as a lesson to other games about how these simple-to-implement aspects can totally affect the feel of a setting.

To Sum Up:
1) FFT's "mundane" plot makes it feel a lot more plausible than the fantastic-but-disconnected plots of other Final Fantasy games.
2) While many aspects of FFT are industry-standard, their presentation makes it feel a lot more "serious" than comparable games.
3) The fact that so many of FFT's good qualities are found in subtle touches should serve as a message to people who want to design believable worlds.

Design versus imagination

Image used with permission of Jack Monahan

In a previous update, I talked about the nature of books and text as opposed to other medium. The important point in that discussion was that most mediums are limited in how many senses they can portray directly, and for the other senses things like descriptions and clues are necessary to convey what's going on. In the case of movies and games, it's sight and sound. In the case of graphic novels, it's sight alone. In the case of books, no senses are "directly" portrayed, which makes it both more creative, more abstract, and more of an exercise for the author and the reader.

However, in many cases books are not entirely text. An occasional illustration here and there may exist, providing a vague but definite framework to the series - what a character looks like, what the environment looks like, and so on. One non-book example comes from Half Life's Gordon Freeman, who is never seen (there's one picture of him but it's not identified as being him) in the game, allowing players to imagine their own ideas of appearance. This shifted with the release of Half-Life's Game of the Year edition, which put him fairly flagrantly on the front cover (compare with the original cover). This simple change turned Gordon from a faceless player-avatar (with an established background) to a definitive, recognizable individual who just happened to never talk.

The point, then, is this: even if something is largely an "imagination-based" medium, like a book or a pen-and-paper RPG, the visual design and illustration of the series establishes a framework for people to think about the game. Obviously this has positive and negative effects. It's positive because it's evocative and helps to establish what the setting is supposed to be like, giving definitive shape to characters, places, and events. It's potentially negative because it can curtail the role of imagination. This can be especially problematic with RPG sourcebooks, which are meant to represent a wider range of concepts and settings beyond what the book sets out.

This means that some very subtle differences in artistic tone and direction can totally shift the nature of a text-based medium. While these things tend to be regarded as less important than things like "good writing" or "solid rulesets", they're still pretty important in establishing a mood as long as the art styles within a given product are relatively consistent. A shift in art can make or break a mood, and while each art style is going to have its own fans, it's also going to be a different group of fans.

Compare the art of Karl Kopinski to the work of Mike "Daarken" Lim. They've both done work on Warhammer Fantasy and Warhammer 40k, but Kopinski represents the "old school" and Lim represents the newer editions. The thing to notice is that while they depict the same things, and have a lot of similar themes and tones - dramatic battles with lots of emotion and conflict - simple things like color composition and lighting make them very different in tone. It's a markedly different style, and as such has different associations in terms of the tangibility and believability of the images.

This is a difference that results even when the things being depicted are actually the same. Now let's ramp it up by comparing D&D's different editions, from AD&D to 2nd Edition to 3rd Edition to 4th Edition. You could say that the artwork has improved in technical terms, but it has also changed thematically as well. 1st Edition's artwork was very basic and medieval-derived. 2nd Edition was obviously more fanciful, but still fairly realistic - it was exaggerated and heroic, but still based in reality. 3rd Edition became more about the glossiness and stylishness of the design, going for a more "punk" approach. 4th Edition was similar to 3rd, but took the general "glossy" feel even further, making things feel more cartoonish (at least to me).

No matter which of these editions and styles you prefer, there is an obvious difference between them. The style reflects on the gameplay, because those images are there to provide a framework for the player's understanding of the people, creatures, equipment, and other things within the universe. Even if everything else is the same, the art style is going to play a major role when it  comes to a reader's baseline understanding of the universe.

I could talk about a lot of other franchises that shifted art design (Warcraft would be the biggest one), but the point I'm trying to make here is that even when the design is not explicitly connected to the product, the art choices still influence how people think about the game. The same is true for books - something as simple as a chapter illustration or cover art can provide a reference point. If it's turned into a movie, it can be even more severe, as now every part of the universe is given a distinct visual appearance.

TVTropes uses a term called Inkstain Adaptation, and while not entirely the same, it gets across a lot of what I'm trying to illustrate. If you have a vague concept, people can do with it what they like. If you have a definite concept, it's going to color their viewpoint on the franchise as a whole. Therefore, establishing that definite concept is a major concern when it comes to creating a new franchise, or expanding an existing one.

To Sum Up:
1) Visual design and illustrations can serve as points of reference even in a heavily text-based material.
2) The audience is going to extrapolate from the images they see what the rest of the universe looks like based on the style and composition of the illustrations.
3) Changing the style and illustrations can have much more far-reaching consequences than simply being "new art" because of this.

Analysis: Assassin's Creed

Assassin's Creed is a prime example of an overcomplicated game - a game that could work well with a simple premise, but throws on a bunch of questionable elements to try to make itself more "unique". In this case, the fairly believable universe of "the Crusades" and "Renaissance Italy" are connected to the much less tangible concepts of a world-wide history-spanning conspiracy that includes basically every historical figure worth naming.

But I'm getting ahead of myself. To sum up the story, Assassin's Creed puts the player in the shoes of Desmond Miles, a modern-day man who uses a machine called the Animus to relive the memories of his ancestors, Altaïr Ibn-La'Ahad and Ezio Auditore, both of whom were assassins. In essence, this story exists to justify the "gamey" aspects and creates a framing device: you are playing Desmond, who is in turn playing the game's protagonist. This means that there are two major aspects of the game to review: the "in-universe" historical periods, and the "meta-contextual" modern conspiracy stuff. Let's start with the former.

Location/Setting
Personally, Assassin's Creed's representation of historical locations and cities is one of its major redeeming factors for me, not because they're "accurate" or "detailed" or anything, but for sheer presentation value. It's one of the few games where cities feel "real"; the streets are bustling, the landscape is sprawling, and despite the eventual repetitiveness of the environments, it generally feels like stuff is going on. There's a lot of what could be called "detail" in the sense of little touches and design choices that make the world feel more natural.

Of course, a lot of the design is there for convenience's sake - whether it's Damascus or Acre or Venice, there will always be boxes and signposts to free-run your way across, and most of the crowds exist for the sake of having crowds rather than establishing an actual infrastructure. I don't blame the game for that, because that's what it needs those things for. It's impressive enough that it has crowds that large and uses them for the gameplay, whereas in comparable sandbox games like GTA the crowd ends up being just sort of an obstacle. In AC, negotiating the crowd either subtly or overtly is a major part of the game.

What ended up compromising the "historical believability" of the whole thing, for me, was how clean-cut everything was in terms of game mechanics. Every high point had a bale of hay underneath (more on that later), there's always people to help or shops to open or whatever. It feels too segregated - less like Altair or Ezio is actually making a difference in the community, and more like he's checking off the next section of his big to-do list (which is probably how the player ends up feeling about it). So much stuff felt the same that there was no part of the city design that really surprised me or felt "new". In all the games, the only place that really impressed me with its uniqueness was the city of Acre in AC1. It was "unique" because it used a totally different architectural style, group of people, and color palette from the standard Middle Eastern cities like Jerusalem or Damascus. In AC2, going from city to city barely meant anything except guards wearing different colored uniforms.

Guards and Combat
The guards were another thing that I really liked about Assassin's Creed, specifically in AC1. I liked the natural progression from "low-level" to "high-level", going from light armor to heavy armor without being unrealistically flashy. I also liked the relatively simple costume design, using things like surcoats to identify allegiance without being overly flashy or colorful. The general "improvement" trend continued in Assassin's Creed 2, but the colors are much more obvious and the armor is more decorated. While this reflects changes in the environment, from the Middle East to Italy, it also feels a bit less subtle. Still, I thought the guards did a pretty good job of establishing sensible uniform concepts while still distinguishing allegiance and class.

As far as their reactions and behaviors go, though, it's a bit of a mixed bag. Assassin's Creed is neat in that enemies will panic or flee from fights if things are going downhill for them. They'll chase after the protagonist if he runs away, they'll throw rocks at him if he tries to climb to safety, and a running escape will usually end with a lot more guards after them. This is balanced by the fact that they're really comically easy to kill, all the time, for the entire game. Enemies attack one at a time and telegraph their attacks in a really obvious fashion, and a patient player can just counter-kill them easily without ever really being in danger.

It's not even really just a question of game mechanics in this case. It rarely feels like the protagonist is in actual danger, because no matter how many people have him surrounded they are bound by rules of chivalry to attack him one-on-one. The funny part is how many ways the game tries to give you to escape - diving through market stalls, hiding in hay bales, hiring mercenaries or using vigilantes to hold them off - but there's no real point. You can run, and they can chase. If they catch you, they're going to attack you one at a time. It's a situation where the player is given a lot of options, but no reason to actually use them.

Essentially, it's a trade-off. Assassin's Creed intends to make the player feel like a total badass by letting him slice his way through guards like a classic swashbuckler hero. On the other hand, it also wants to create a dynamic where he should escape from the guard, making for exciting free-run chases. These two things need to be balanced: the guards need to be more of a threat, or the player shouldn't be expected to want to run from them.

Improvement and Upgrading
In Assassin's Creed 1, the upgrade system was simple and linear: kill a target, get a piece of equipment. It wasn't inventive, and it didn't really reward all the side-quests (flag collecting, templar killing) but it made for distinct and obvious advancement. You also had a whole period to experiment with your new stuff, although really it wasn't that big of an improvement to begin with (better sword, more throwing knives, etc). It was a simple system, but it was balanced pretty well.

Assassin's Creed 2 changed this in two ways. Firstly, money was introduced, providing an overarching concept of purchasing, rather than "earning", upgrades. Secondly, the player received control of a villa that could be upgraded (at cost) to provide more money, making it basically a combination of "thing to spend some money on" and "way to get more money than you know what to do with". While actual upgrades were still unlocked as the story went along, it instead unlocked the option to buy them.

Money in AC2 is spent on three things: personal upgrades, supplies, and Villa upgrades. The first group is a one-time deal (you buy a weapon or piece of armor, you've always got it), so the constant influx of money from the Villa pretty much makes it trivial. The second group (ammunition, medicine, and so on) is more substantial, but remains relatively cheap throughout the game. The third group only exists to bring you more money. Sure, it makes the villa look nicer too, but essentially the only interactive part of the villa involves you getting more money.

Assassin's Creed: Brotherhood mixed it up a bit by making the entire city of Rome littered with shops and stores that can be re-opened with a relatively small investment. This ended up being less an issue of income and more about convenience: "I want to buy something but I don't want to run all the way over there. I'll just run to that convenient shop window and fix it up into a blacksmith or something". It felt really modular - all you see of the shop is a window, so it feels less like you're fixing the town and more like you're, again, just checking off a list.

AC:B also introduced the concept of assassin recruits, who have a minor RPG system associated with them. These guys would have made a great money sink: buying them armor and weapons, customizing their look, and giving them diverse gear options. Unfortunately, they have a pretty simplistic upgrade scheme: you put points in attack or defense, which upgrades their weapons or armor. Every recruit ends up with both attack and defense maxed out, and when they reach "full assassin" status they all get the same outfit and weapon without any sort of customization. It feels like a real wasted opportunity, especially when the trailers and screenshots were full of diverse groups of assassins.

Story vs Gameplay
Assassin's Creed is a prime example of story/gameplay division. The "game" is about being a stealthy assassin who blends in with crowds to take out targets without being seen. The "story" is about a guy who tends to walk right up to his targets, kills them extravagantly, and then comforts them as they die (even in a case where you specifically have to use a ranged weapon because he's too far away to stab!). The decisions the player makes when controlling Ezio has nothing to do with what he ends up doing. This is hardly unique when it comes to games, but it's still frustrating when being sneaky is the entire point of the game.

This is also one of the areas where the whole "Animus" thing starts to become annoying. It's obvious that the linearity of these sequences is justified by the fact that it "already happened", but it's a really shallow excuse because you have so much freedom in the rest of the game. The game gives you so many tools and tactics that it's really annoying when Altair or Ezio takes the dumbest route possible to complete his objective. Aren't these guys professional assassins? Why is the player allowed to do so much better than them?

AC is also full of a lot of "suspension of disbelief" mechanics. When Altair takes damage, it's "de-synchronization" rather than "getting wounded". This is to explain the fact that he can take a huge amount of damage, and also to reinforce the Animus mechanic. In AC2 it's changed to actually being Ezio's health, since it's reinforced by armor and replenished by medicine, which brings it back to being silly. In cutscenes, things seem to operate under real-life logic - you get stabbed, it hurts.

To be frank, though, the gameplay is pretty schizophrenic too, largely for reasons of simplicity. Jumping 300 feet onto solid rock kills you (as you'd expect), but landing on a hay bale from that height means that you're perfectly fine. There's some justification of this as the assassins being superhuman, but it's a question of impact. They're not immune to falling damage, because normal falls kill them. It's the hay itself that makes them not just okay but fine. One's gravity-based acceleration onto a hard surface, while the other makes hay seem like a cloud made of pillows and marshmallows. It's just a cart's worth, it's not that soft.

Basically, Assassin's Creed should pick a side: is it an unrealistic "rule of cool" setting where assassins can pull off crazy stuff all the time with no effort, or is it a grounded setting where stealthy killers must use tact and guile to avoid an untimely demise? The part that bothers me is that it doesn't choose, it says "I want both" and then screws them both up.

Framing Device
Conspiracy nonsense aside, the Animus is probably my least-favorite part of the game. The Animus, and everything about it, suggests the following logic to me: "A game about being an assassin in a historical setting isn't interesting enough by itself. We need to throw in some meta-elements to make it worthwhile".  It introduces a lot of incongruities of the type described above: it makes the game more "real" by justifying the HUD, loading screens, and so on, but it also holds the game to a higher standard because of it - and it can't actually reach that point.

I'm fairly sure the gaming public would have been okay with a suitably thematic HUD - I mean, we put up with it in almost every other game. If you want to make things unrealistic, then go ahead. Just say "well he's like a swashbuckler and that's what we're going for" if you don't want things to be super-accurate. That's fine. What bothers me about the Animus is that it wants to be realistic and fails, rather than not caring about realism in the first place.

What's worse is how many things are justified by the Animus - but then you get to play as Desmond and it's still in third person and he doesn't get hurt by falls and there's just so much wrong all the time. Why did you bother with this, Assassin's Creed? Did you think the Animus would justify anything? It doesn't! Just let me play the game, I can deal with thematic unrealism! What are you even doing??

Also, let's just get this out of the way, the game's pretty much a joke historically speaking. It's a big hobnob of famous names tied to an arbitrary conspiracy based on two 12th-century factions in the Holy Land who now extend to the beginning and the end of time itself (using the same name). That could be applied to "rule of cool" too ("wouldn't it be crazy if there was a giant worldwide conspiracy responsible for orchestrating every event in history??") but it's just so haphazard that it seems like a self-insert fanfiction. It's an attempt to "totally blow the player's mind" by making them question everything except none of it makes any sense. It's not just a "conspiracy" or a "secret war" or something semi-plausible, it's an all-encompassing super-conspiracy that should have fallen apart under its own weight.

Conclusion
 The parts I like about Assassins's Creed are largely immersion-based. I like the look and feel of the cities, I like the mundane designs, and I like the little details. To a limited extent, I'm willing to connect to the protagonists in terms of doing cool stuff, because it's presented in a way that tries to give it more impact, whether it's through visceral combat or death-defying acrobatics. I like the fact that it uses history as a backdrop, because, well, I like history, and I think it's cool when it gets leveraged in an accessible way for modern audiences.

What I don't like is all the fake stuff, not just because "it's not real" but because it's sloppily done. The framework makes no sense, the plot seems like an overreaching attempt to be deep and meaningful, and even the basic aspects of cutscene-based decision-making is frustrating. What really gets to me about AC is that there is such a divide between these two things. It can't be just "rule of cool" because it has the Animus and wants to be taken seriously as a realistic thing. It's not realistic enough to pull that off because it wants to be a cool and fun game. I would have been happier if it tried to stick with one or the other, but doing both just pulls it apart.

So, To Sum Up:
1) AC does a pretty good job at conveying a world that feels populated, even if it gets a bit repetitive.
2) While the combat and free-running are stylish, the dual nature of "fight" and "flight" are undermined by how weak the enemies are in both fighting and chasing.
3) There's ways to interact with the larger world and attempt to affect the environment, but they're so simplistic that it becomes unimmersive - they're just a game mechanic.
4) The fact that Assassin's Creed can't make up its mind on whether it's "cool" or "real" does more damage to its plausibility than anything else about it.

Grounding fantasy

Wizard courtesy of Jack Monahan

One justification for "lack of realism" that I hear a lot is this: fantasy is escapism, so doesn't need to be realistic. Obviously, I try not to focus just on realism in this blog - that's why it's "believability", after all - but the essential gist of my counter-argument would be that even fantastic things are based on amalgamations of understandable "real" elements. From human elements like knights and castles, to magic spells like fireballs and lightning bolts, to monsters like gnolls and dragons, there is always an undercurrent of "something real". They may be used in unrealistic ways, but they're still based on real things. Connecting the audience to those real things brings about the sort of sensory understanding and tangibility that we've discussed in the past.

Magic and Technology
One of the most contradictory elements of fantasy is that it often attempts to be "separate" from realism, and yet relies on realism in terms of things like weapons and armor. This is because of divergent aesthetic choices: fantasy is associated with "medieval stuff", but as things get more and more magic-oriented, it makes less and less sense. If magic is common, people who still use swords and spears need to be justified. AD&D, for example, balanced them out by making fighters tougher than wizards, eventually to an unrealistic extent. Other games take similar measures - making normal people unrealistic so that they're not outshadowed by wizards, who are meant to be the "fantasy" element.

To quote Warren Ellis' Crécy: "We have the same intelligence as you. We simply don't have the same cumulative knowledge that you do. So we apply our intelligence to what we have". Powerful magic should change the world. Why are people still using swords and armor - which are pretty hard to make - if there's magic available through vague and indeterminate methods? The more powerful the magic is, the more "normal people" have to be scaled up to deal with it. Suddenly, a fighter is no longer a regular human with a sword - now he is "magic" in his own non-magical way. Believability has been sacrificed for the iconic image of "a warrior with a sword" or "a knight in shining armor", which is fine if that's all you want, but not if you want to establish a universe that makes sense.

Even simple "low magic" would be a big deal for a fantasy universe if it was reliable. There are so many relatively mundane uses for something like a fire spell or a wind spell that it would change the setup of a culture. Technology is based on exploiting available resources in a reliable and effective way, and magic is, essentially, a new resource. Of course, many settings don't bother to show the "resource" aspect of medieval life, except in the sense that, for example, there are smiths and people buy swords from them. The intricacies and time investment of making a sword is overlooked in favor of a "new weapon every level or two" mindset so that there can be constant improvement.

Medieval technology is not an enforced aesthetic choice, it's an attempt to use what works. In contrast, fantasy often has knights and armor for their own sake, rather than "because they make sense". Can you imagine an actual smith in the universe of World of Warcraft, Record of Lodoss War, or Fire Emblem? There's such a difference between "metal" and "whatever faux-plastic cartoon armor is made of" that it's difficult to even imagine connecting the two. Yet those settings still have blacksmiths, because it's an expected fantasy concept. It becomes an issue of drama when people are forced to either (a) use "natural" abilities that aren't representative of normal humans or (b) use weapons that are inferior to magic in every way.

This can be balanced out if there is a reason not everyone uses magic. In the Warhammer and 40k universes, magic can have critical backlashes and summon demons, and thus is not trusted by the majority of people - hence there is more reliance on technology, with magic kept on the sidelines as an oddity. If magic is clean and reliable, there's no reason not to incorporate it into the world's technological systems. If it's questionably safe or overtly hazardous, there is.

 
Spell Effects
It's true that spellcasting is, by definition, unrealistic. It operates on its own set of rules. It generates something from nothing. It messes up a lot of physical laws in ways that aren't immediately visible (semi-permanent changes in the environment and so on). However, one thing about most magic is that it exists in an understandable state, such as fire, water, lightning, rock, and so on. Unfortunately, this can be abstracted to an almost meaningless level due to poor or unrealistic representations of it.

For example, would you ever connect fire in a video game to a real fire in terms of "being hot" and "burning things"? A character can be set on fire, but it's just a flickering orange aura that doesn't affect their character model at all. Lightning consists of big, cartoonish bolts and don't convey the actual power and effect of lightning. Even something as simple as an earth-based spell tends to throw a clod of dirt, rather than a convincingly heavy stone. It's stuff that audiences and players are used to, because those elements are common and poorly represented in so many games. It's fine for a mage to hold a fireball in their hand and throw it because the audience/player can't see the heat, and as such the heat doesn't exist. When it explodes, it does damage, and that's it.

This is true in non-magic situations, as well. Compare the damage done by a flamethrower in Far Cry 2 (one of the only games with actual spreading fire effects) to a flamethrower in real life. The former feels more like a set of effects, rather than actual heat and fire. You're just spraying a fire effect, things don't react to it like they would to an actual flamethrower. Obviously the limitations of technology have a lot to do with this, but even when it comes to peoples' imaginations, those representations are going to affect their perceptions. It's more of a concern for spells because unlike a flamethrower there's no way to see a "real" spell.

Despite their failings, at least things like fireball and lightning bolt are appealing to an existing concept. Spells like magic missile that do "arcane" damage are even more difficult to connect to. In some situations, a magic missile does "force" damage, which is at least connectable to an impact of some kind, but what is the audience supposed to make of "arcane" damage? In-game, of course, it's perfectly serviceable, because all you need to know about it is how it applies to the other numbers in the game. In-universe, though, it's difficult to get a sensory understanding of what just happened.

Basically, a spell does something - it burns, it freezes, it shocks, it melts, and so on. Connecting the fictional and abstract concept of a spell to a tangible feeling like heat or cold can make it more intense and immersive for the audience or the player. Underplaying these concepts may be a necessity of the medium at times, but repairing those concepts would do a lot of good in terms of tangibility, especially when it comes to how audiences understand and imagine those things in other venues.

Monsters
Monsters are a questionable element in fantasy because they're not biologically sound. Generally, a fantasy/mythological monster is either "some animals mashed together" (Chimera, Manticore, Pegasus), "an animal that's weird and different" (Unicorn, Cerberus), or "an animal crossed with a human" (Minotaur, Mermaid, or any of the countless animal-themed races found in fantasy RPGs).

One reason that animals show up a lot in mythology is that they're conceptually familiar. It's the kind of term that can be used to verbally describe an unknown concept - it's been suggested, for example, that a unicorn is based on descriptions of rhinoceroses as being essentially "a horse with a horn". Animals provide a baseline that can be understood and connected to already-known things. They also provide a ready-made design - just throw a bull's head onto a human, bam, you've got yourself a monster. Even Cthulhu, a beacon of interstellar incomprehensibility, was basically described as being "squid-faced", and that's the part that people remember.

Even with less directly "realistic" monsters, there are still a lot of sensory elements that can be used to make them feel more tangible. A monster's skin can be slimy, leathery, scaly, or coarse; their breath can be foul, fetid, rotting, or noxious; their roar can be piercing, deafening, or rumbling. The use of such adjectives connects the fictional construct of "a monster" to real, understandable concepts and allows them to aggregate together to form a larger whole. The more the description and art can convey those senses, the more "real" the monster will feel. The monster itself doesn't need to be realistic - that's not what we're concerned about right now - but it should feel tangible and comprehensible.

Basically, any given monster - from a troll to an ogre to a gelatinous cube to a beholder - is made up of recognizable parts. In some cases it might be a bit of a stretch to connect the "real-life" equivalent to the monster, but it's still helpful to have a point of reference. Making use of real descriptors can overcome the barriers between having a monster who exists only as fantasy and a monster that feels a lot more "real" to the player. It allows the player to bring their real-life experience to bear when trying to comprehend a largely unrealistic world.

Conclusion
What fantasy treats as "magic" isn't really that abstract. It's generally a reliable, observable phenomenon due to the mechanics and rules of the game system. If you combine elements x, y, and z, you can throw a fireball. In addition, what magic actually does is generally based on real things, whether it's freezing someone or creating a gust of wind or making someone stronger. Magic is based on concepts that exist in real life, and by leveraging that general fact sensory elements like touch and smell can be reinforced.

When people say "it's fantasy, it doesn't need to be realistic", they're undermining the senses that cannot be portrayed. A cartoon has no smell, a film has no taste, a game has no touch (at least not yet). Even these fantastic elements, though, can be easily connected to real-life things through these simple methods, and this will give the audience and the players a greater appreciation for what's being shown. This should aid in their immersion and their suspension of disbelief.

So, To Sum Up:
1) If magic is reliable, then it's a resource or tool just like any other part of the environment, and should be used as such. If people aren't using it that way, there should be some explanation of why that is.
2) Spells generally try to simulate some real, understandable concept even if they do so in an abstract or "unrealistic" fashion. By evoking the intended reaction or material, the spell's effect can be made more impressive and tangible based on the audience's understanding of it in real life.
3) Monsters are almost inherently unrealistic, but by taking aspects and concepts used in real animals, a connection to the monster's sensory nature can be created.

The player/character relationship

A key element for writing a convincing character is to convey the idea that their universe is real to them. Their decisions have weight and impact - their losses are tragic, their victories are triumphant, the sensations they feel are connected to what's happening to them and around them, and so on. This comes from a direct, fully-sensory link to the universe: what's happening to the character is, well, what's happening to the character. If they get slashed with a sword, then all the associated baggage of being slashed with a sword should occur to them unless there is a reason not to. The drama of the story hinges on the idea of the character actually caring about what happens to them - if they aren't interested, then why should the audience be?

The player, on the other hand, cannot be as invested, due to the combination of lacking sensory detail and "real-life" consequences. A player, like any member of an audience, can be deeply invested, and they can roleplay, but it's never going to be the same for them. There are a bunch of ways to try to make the player care about what happens - likable characters, events with consequences, all the aspects of believability discussed in previous articles - but ultimately "the game" is not reality.

However, one interesting correlation about this is that dissension usually arises in cases where a character cares and the player who controls them doesn't. In Grand Theft Auto 4, Nico Bellic switches from "concerned individual with friends and comrades that he cares about" to "guy who just ran over a sidewalk full of people for no apparent reason". In other sandbox games, like Saints Row 2, the sociopathic nature of the protagonist matches the player's usual actions much more closely. Of course, this isn't absolute - there are plenty of people who play games in a "character appropriate" way, but the issue is that it's not important or notable to the story. There's "the character" and there's "the player", and ne'er the two shall meet.

There are a couple different ways to approach this situation, going from most-divisive to most-immersive.

Independent Character / Controlling Player
This is the scenario described above. The character is their own person, and when not being controlled by the player they make decisions as they see fit. When the player is in control, they may act totally contrary to the way they see themselves and choose to act. The player is the character, for a given definition of "is", but the character is also their own person. One of the issues of believability regarding this setup is the fact that there's no repercussions for the player-controlled sections of the game. Perhaps the character will quip or remark in response to a situation, but there's no real commentary on what's happening. The rules tend to work differently, as well - the usual mix of "cutscene bullets kill you in one hit" and "dying is a big deal" and other gameplay/story segregation concepts.

An approach on this that I feel would be interesting is the idea of this divide existing in-universe as well as in the meta-sense. That is, the player represents a spirit or entity with no connection to their realm. Occasionally, the character allows the player to control their body and imbue it with superior speed, strength, etc. This addresses the issue of motivation (the player acts as a spirit that acts according to their own whims rather than consequences) as well as justifying the difference in action. Perhaps the player and character could even come into conflict depending on how their viewpoints differ. "Deadly Premonition" plays with this concept, but does not create an adversarial relationship out of it. It is less "direct possession" and more of an advisory role.

Dynamic Character / Choice-Making Player
Not all games make a character "independent". Some games, such as the majority of Western RPGs, make characters variable in differing ways. These are limited by the reactions that the developers include, as well as any existing meta-systems like a good/evil meter. In short, though, it seems like having decision-making aspects should make the game more believable - after all, now the player is invested in the character, and the character isn't simply an immobile, unresponsive part of a linear story.

However, I'd say that there is a downside to this concept, and it is that "there is no longer a protagonist who is part of the universe". The protagonist is now motivated by the player's decisions, and the player remains unconnected to the universe in terms of consequences and senses. It's more like the character is now an avatar for the player; even if the player is establishing roles and making decisions based on their conception of the character, their motivations are still based on whims, rather than consequences. Death is "Oh, I lost" and not, you know, death. Again, there's nothing you can really do about that - it's just that extending choice to the player character includes them in that disconnected state.

Essentially, the "RPG character" is a strange hybrid. On the one hand, they exist in the universe and are supposed to have all the sensory/consequence issues that real people have. On the other hand, their choices are made by a person who doesn't have any of those. This means that their choices aren't going to be made for the same reasons that real choices would be made except as a coincidental bit of roleplaying. Proper immersion can aid in the player making decisions that the character would be motivated to make, but this is by creating sympathy or evoking similar emotions. It's a complicated scenario, but "making the player care as though the world was real" is the goal of a lot of RPGs.

I should note that, theoretically speaking, there are two kinds of DC/CMP setups:
1) The character exists as their own person, but the player influences the kind of person they are (think Shepard from Mass Effect). This is often used to try and make up for the limitations of video games: you can't really make an "open-ended" protagonist, so the next best thing is to make a linear one with branching paths. You have to be Shepard, but you can choose how Shepard acts and what he or she looks like. You can't choose their voice though. The voice is there forever.

2) The character is created by the player (think any pen-and-paper RPG). This allows for the player to adopt a far broader range of roles even within the confines of the setting, but is difficult to execute in a game that relies on premade dialogue. In a P&P game, the "dialogue" is naturally generated, so it's okay to do that. In a video game, it has to be written out beforehand (with some room for dynamic generation), so a few "personalities" develop (Good Protagonist, Evil Protagonist, etc) which is the most likely cause for the first category's existence.

"The Player As A Character"
Similar to the scenario I described above where the player is an advisor or spirit, some games go with the idea that the player and their interface directly represents a character in its own right. For example, in "MechCommander", the game display and HUD and so on are explained as being the MechCommander interface. The player is a Mech Commander, and the "game" is the program they're using to command their mechs. This can be seen in the opening movie, where the (non-player) commander uses an interface similar to the game's interface to direct his units. The hacking-based game "Uplink" is another example: the interface is meant to be directly equivalent to an in-universe program. Even fantasy games can get in on it - in "Dungeon Keeper" and "Black and White", the player represents an ethereal spirit disconnected from the "real world" but able to influence it in various ways. When the player is "damaged", what's being damaged is their connection to the world.

In terms of immersion, this can be the best concept (if properly explained). There's as few unexplained game things between "the player" and "the player's role" as possible. The lack of connection might even be noted - what risk is there to the commanding officer throwing their soldiers' lives into combat? The relationship between a player and their troops might well be the same as the in-universe relationship between a commander and their subordinates. Like in previous examples, they may care and take things seriously, or they may not - it's all up to their whim. Unlike previous examples, this is something that makes sense in-universe, because now there is a reason for all the disconnected elements. Of course, there are still limitations on content and dynamic generation, but in general the link between the player and their role is much stronger.

So, To Sum Up:
1) The character cares about things in-universe because they're happening to the character. The player cares about things in-universe in a detached or isolated way - at best sympathetic, at worst sociopathic.
2) This divide means that players and characters make decisions for different reasons - even if both have a reason to do things in "the most optimal way", the different factors mean that "optimal" results in a different set of choices.
3) One way around this is to make the character the same as the player, sharing the same limitations on sensory information and the same disconnect from events. This explains the player's decision-making process in-universe, because they're a powerful, untouchable being whose whims determine their course of action.