I originally posted the following in the forums, but there didn't seem to be much interest. It took me a little while to write, so I'm copying it here for archival purposes. That, and somebody in the thread told me to do it.
I consider a lot of the music I listen to to be rather cinematic. For a while I thought that it would make great film soundtrack music. After a little thought, I decided that perhaps this was not the case. The reason the music is so cinematic, I reasoned, is that in its way it also fulfils the roll of the film; were it to be put into a film, it would seem like the two elements were competing for attention, rather than complementing one another. That's not to say that it's impossible to use it, but generally that would require some sort of compromise to be struck, and even if it didn't, the thing which made the music cinematic would not be the thing that made it suitable for use in a film.
On further consideration, I'm not entirely confident in this theory, but it did make me think that perhaps similar reasoning could be applied to the relationship between games and films. Some people seem to be eager to see their favourite franchises transplanted from one medium to another, but I'm not sure that this is always a good idea. We all know that in practice this can be disastrous, but generally this is chalked up to inferior execution rather than an underlying discordance. It's the Uwe Bolls and Paul W. S. Andersons who are ruining these franchises, we say. Now, I don't want to get into a discussion about these directors' merits, or the chasmic lack thereof (suffice it to say that my confidence in them is limited); rather, I'm interested in whether there's an inherent difficulty in translating one medium to another.
Perhaps it's obvious. Games do things that films can't. Apart from that comical sequence in the Doom movie, it's pretty impractical and jarring to have extended first-person sequences in films, yet it's common in games. In games you can influence proceedings, whereas films have to follow a static path. Games are generally much freer. On the other hand, this freedom comes with its own narrative limitations: pacing is hard to control with anything like the precision of films (if the player wants to run around shooting corpses in their groins and stacking boxes in a corner for an hour, they're generally allowed to), all the character-building in the world can be undone by a player who wants to act like a superhuman sociopath (indeed, to generate sufficient gameplay it's generally necessary for the player to repeat actions for much longer than would be acceptable in other media -- most game protagonists exceed the body count of pretty much all films), and a carefully constructed mood or emotional investment can become somewhat irrelevant when the player is restraining himself from committing acts of obscene real-world violence after dying on the same section for the twentieth time, and is instead making do with increasingly profane outbursts.
The point, I suppose, is that the things that make a good game good tend to be at least partially dependent on its being a game. This isn't necessarily a universal rule (you could have an OK game with a fantastic story, for example), but I think it probably applies to the majority. And, to a certain extent, the same is true of films. Theoretically, you can import elements from a film into a game more than you can the elements of a game into a film (as an audio-visual experience, a game can include anything a film can, whilst films, being non-interactive, cannot provide that aspect of gaming), but after a point it just becomes a film with some game around it (even the Metal Gear Solid series is unlike films, in that it has extended interactive periods). But whilst some cross-over is possible, in general the best of either form will have been specifically tailored for that format.
Perhaps it would be helpful to take an example. Half-Life (2) is a popular game, and I think quite a few people are or would be excited about the prospect of it being made into a film. And I think it's quite understandable: it's good, and it has a reasonable degree of focus on its story, something which is generally helpful in films. But, for me at least, the reason that experience is so memorable is the way you're dropped into this world, initially being told nothing about your past, and learning very little about it along the way. The silent protagonist may be a cliché, but in this instance I think it's a tool used to good effect. Knowing more about your character would only act to distance you from him; as it stands, there's an interesting tension between immersion and alienation. On the occasions that you are confronted by the G-man, you are reminded of how little you understand the world, and how alien you are to it. I suppose it's a rather post-modern device, in that it acknowledges the position of the player, although rather than breaking the fourth wall, it incorporates it into the game's fiction (incidentally, I seem to remember some of the on-screen text in the first game being more explicit about this than that in the second -- wasn't there something about assessment being terminated when you died, and so on?). Anyway, this is all a little tangential. My point is that things like the first person perspective and the silent protagonist are important contributing factors to the audience's experience. If, for example, it were in the third person, it would just be some man running around in a certain situation; in the first person, however, it's you being put in the situation, and unexplained things become that much more intriguing. If a film were to be made of Half-Life, it would become a much more standard experience. If it followed the game too closely, it would just be a dystopian science fiction action movie. Fleshing out Gordon Freeman's character would pretty much defeat the point.
So what are we left with? As far as I can see, it's the world the game is set in. It would be quite possible to accommodate the particular needs of films in this fictional world, and those of most other games. It's a pretty broad framework to work with, but that's pretty much what the scriptwriters need, otherwise you have two hours of "and then the man shoots another zombie". The problem is that games are, on the whole, very influenced by films, so what you're often left with is a fairly standard set-up. Resident Evil becomes zombies-and-corporation-conspiracy, which, incidentally, isn't a million miles from monster-and-corporation-conspiracy, which is what Alien was all about, although I wouldn't say the two are treading on one another's toes in any significant way.
Of course, you could say this about films in general. How many truly new ideas do you see these days? But if that's the case, what's the point of using the licence at all? I guess it gives you access to particular set pieces. Half-Life's Citadel, for example. It's not an entirely original concept, but it's a pretty cool execution. A lot of these little details just end up as knowing nudges to gamers, who are probably too annoyed at how much the film has deviated from the source material. Which is a problem, since it's necessary to deviate from the source in order to make a film, rather than just make an entirely linear version of the game which someone else plays for you. And the general public may well be put off, either thinking that games are "not for them", or that the film won't make sense to someone who hasn't played the game.
Returning to my theory about filmic music, another example I'd like to consider is Shadow of the Colossus. I thought it was a particularly cinematic experience, but one which would probably make a pretty lousy film. There's barely any dialogue in the game, and much of the action involves riding around on a horse. If shot well, it could be a beautiful film, but it would require real skill on the film-makers' part to keep the audience interested. People approach games in a different way than films, and this allows them to explore different areas. Even something which may seem particularly film-like, such as Metal Gear Solid 4, operates in a different way. What film could devote an hour to the ending? I don't remember how much The Lord of the Rings had after the climactic scene, but even that seemed jarring, and was obviously the result of it being based on a series of novels. Even with the extra space afforded to a television series, it's generally impractical to devote that long to wrapping up. Of course, MGS is pretty self-indulgent, and as such an unusual case, but I think that games are much more flexible in these terms than most media, other than novels.
What about Portal? It's a great experience, but I think what makes it great is, at least in part, the relationship between you and GLaDOS. If this became the relationship between a third party and GLaDOS, you would be kind of entering HAL territory. Which is fine, but, well, it's been done. There is something uniquely exciting about you being the one finding your way through the story, and, if you were lucky enough to come to it with minimal prior information, expect one thing, then gradually uncover something quite different. The way in which that's a fantastic interactive experience doesn't fully translate into a passive experience.
A classic example of the opposite transformation is GoldenEye. When the quality of film-to-game conversions comes up in conversation, I generally give that as an example of a rare success (indeed, a Rare success). But whilst some elements are very accurate reconstructions of sets in the film, I feel that the James Bond theme (I don't mean the theme tune) is pretty much incidental, and the game would have been just as good without the film licence. The way to make a good game out of a film, apparently, is to make a good game, then populate it with assets from the film. I couldn't tell you whether that's how it worked in this case, but the point is that the game needn't have been a James Bond one.
I'm not saying that films based on games can never be successful, but I think that in order to be successful, they have to take some pretty heavy liberties, and have to understand the fundamental differences between the two formats. And I think we should think twice before declaring that every game we like should be made into a film. Perhaps it's a game for good reason.
Of course, I may be overplaying the distinction. Maybe the examples I picked were unusual. I'd be interested to hear others' thoughts on the subject. As tends to be the way with these things, this all seems much less certain and concrete than it did in my head, so I'm sure there are plenty of things to pick apart.
Finally, I hope this isn't all sounding negative. I don't think that it's a bad thing. I don't mean this to be some sort of hopes-dashing exercise. If anything, I think it's an affirmation of the uniqueness and validity of gaming as an activity in its own right. And whilst I don't want to make any bold proclamations, I think that if gaming can offer interesting experiences that can't easily be directly replicated in other media, it's capable of at least a certain degree of artistry. I used to baulk at the whole "games as art" thing, but a friend pointed out that whilst games might not bear any comparison to what some might call "high art", neither do most of the films that we see. Perhaps it's not Art with a capital A, but it's . . . well, it's something. Something other than an imitation.