I saw this on Jake Simpson's Blog this morning:
Do's and Don'ts for Cinematics - Thinking more about what Bruce Evriss touched on the other day regarding cinematics, I put together a list of do's and don'ts for their implementation...
I won't reprint the whole thing - instead, I encourage you to read Jake's point on the matter. While Jake's points are golden, they aren't specifically what I wanted to discuss. Instead, I wanted to ask a more fundamental question:
What's the point?
Cinematics in games have been around for nigh on 20 years (a lot longer than that, if you count Pac-man's bumpers between levels as the first cinematics)... They've been 3D since the mid-90's... why are we still having problems with Cinematics in games?
Some would argue that we don't have the cinematic sensibilities to do them right... and I even agree, but I suspect most people know my stance focuses on a problem that is more fundamental: Games and the cinematic form don't go together well.
To be clear: I'm not saying they haven't had their uses, but I am saying that I've really never seen them implemented in a way that doesn't totally, even if only for a brief moment, break the game experience. As a designer, that's a big part of the reason Cinematics are a bane - no matter how you implement them, they're a cluster of edge cases and special exceptions that you have to deal with and every solution tends to be a bad one.
I've worked on 8 or so games with Cinematics throughout my career, from my days as QA at Blizzard, through my time as a designer Fear Effect on the PS1 and on into Xbox games like NightCaster, where I served as both designer and producer on various installments. In every case, the cinematics were always a moment when control was taken from the player and the way the player viewed the game, changed.
Fear Effect was actually an interesting project, in this regard -- most of the cinematics in that game are "completely integrated." This is to say, the game is basically one very long cinematic (like Dragon's Lair with more interactivity). Despite this, there are moments when the game experience stops and the movie experience takes over. The result, at least for those few people I've spoken to personally (remember, I'm no expert here), was always a "cooling off" -- a moment where the player and the game disconnected. In the case of Fear Effect, we had cinematics that tended to keep the connection between story and player strong, so players didn't mind, all that much -- but there was an interesting side effect. Every time a cinematic played, the players we watched would get worse at the game. They'd have to work to regain the situational awareness they had attained prior to the cinematic.
The theoreticians of our industry will nod sagely at that and say something about "engagement theory" or "levels of excitation," but I think that misses the point. Regardless of why it happens; it creates an opportunity for the player to fail despite gained knowledge and learned skill. Every such failure takes the player one step closer to not playing again. Failure because you didn't quite rise to a new challenge isn't necessarily bad (it can even encourage people to try harder), but failure to succeed in a scenario you have already mastered almost always leads to disappointment -- and that's bad.
That being said, there are a million good reasons to want cinematics, so I'm not saying the idea should be scrapped. Instead, I'd like to point to the work of folks like Valve and say, "We should do more of that." Watching people play the various Half-life games is an interesting experience, for me. NO player I've watched has ever framed up the speakers in a "cinematic" in that game and just sat motionless while the drama played out. Some tensely guarded the windows and doors, expecting ambush at any moment, while others orbited the speakers of the unfolding drama, to watch from every angle... still others simply walked away and explored the space, looking for more elements of the game. I think this is what makes their "cinematic" implementation nearly perfect, because this allows players to experience the story and unfolding drama in a way that allows them to remained connected to -- and engaged with -- the gameplay.
It works, and it works incredibly well. This is particularly obvious and tangible to me in Half-Life 2 and its episodes, where they step back to the more traditional "player cannot interact with the world while the cinematic is going on" model a few times. Sometimes they do it to evoke a particular emotional response, which is both frustrating and interesting. For example, early in Half-life 2: Episode 2 there comes a point where you literally can't do anything but watch the cinematic play out (you can't even move) which leaves you feeling completely powerless. This is exactly what it appears they want, but as a player I felt totally disconnected from the game experience as a result. Especially after having "been" Gordon for so long that I had no less than 7 strategies in my head for how to deal with the thing that was unfolding in front of me. Instead of, "Oh no! Gordon can't do anything!" what I felt was, "Damnit, if only I had control of Gordon, I could have prevented that!" I did feel powerless, but it didn't feel natural. Instead, it felt artificial and arbitrary.
Games aren't movies -- and perhaps that's truer of 1st person shooters than it is of other genres. 99% of the time in Half-life, I feel like I am Gordon Freeman. That's the power of Valve's storytelling technique, a technique that depends on level design and gameplay more than anything like cinematics... but at that moment, when Valve instead depended on the story-telling techniques of movies to convey emotion, Gordon and I were two entirely different people. I was forced to watch Gordon's story for a time until the game would relinquish control and I could return to my story.
That moment colors the rest of the experience and I can't help but think the game - and its story - would have been more powerful, if only they had not used a traditional cinematic mode. Maybe that power would never have been lost, had we as an industry eschewed the concept of cinematics all together.
Of course, some of my favorite games would have suffered along the way, so as an observer of and participant in my industry, maybe I'm glad we didn't -- after all, you can't learn from mistakes, if you don't make them first, right?