Cinematic Platformer

[mode7]

Hi Guys!

I recently started thinking about designing a so called cinematic platformer. For those who never heard of this term you might now Games like Out of this World/Another World, Flashback, Heart of Darkness or Oddworld and ultimately Prince of Persia (the old one). Also the (brilliant) AGS game 1213 would fall into this category (I wonder if there are any more of these here?)

From the technical side this is coming together quite well. Actually coding the basic controls took me only a few hours (the stuff I learned when making Abduction really helped). And with the rotoscoped animations it looks already pretty cool.
Story and art-wise I already got (at least) a concept.

Whats puzzling me is the gameplay. When I think about these games mentinoned before I've mixed feelings.
I love the degree of immersion they create compared to classic point'n'click games. Of course the graphics and animations and that they actually try to balance action and adventure unlike the so called "action adventures".
On the other hand I don't like the trial and error gameplay, the fact that you die on every corner, the sluggish controls and the lack of adventure style dialogs.

So I'd like to know from you what you liked and what you hated about these games. To avoid mistakes other games made before.

The other thing is, when doing a platformer you really have to design stages in a functional way (very much more than in a classic adventure).
So maybe someone has some article, link or advice concerning 2D level design.

[GarageGothic]

I suppose Limbo on the Xbox360 could be placed in this category as well. I like the term "cinematic platformer" - never came across it before, but it's very fitting.
Not sure I would call the first Prince of Persia "cinematic" though, since there's very little story development and the levels are more or less interchangeable, but for the second game it's certainly true.

Another genre description I've heard for this type of game was "acrobatic platformer" - perhaps more often used about their 3D successors like Assassin's Creed and Tomb Raider. But I think that label makes the very good point that the player character is more agile and mobile than in ordinary platformers - and this aspect could be a possibly way of avoiding the trial-and-error gameplay, and even more so when combined with the idea of "cinematic".

For instance, imagine a mechanic where if you fail a jump you may still survive by grabbing onto the ledge and drag yourself up, this is already considerably different from the binary make-it-or-die approach of, say, Super Mario. Now, add on to that the "cinematic" aspect - we want the action to feel choreographed, meaningful, not just dependent on pixel-perfect player control. Let's assume that the developer, in the role of "director" of the film the player is the main character in, not only designs the levels but is also allowed to subtly guide the action, much like the gamemaster in a tabletop roleplaying game - unless he's a total sadist - will try *not* to get the player characters killed just because they made a poor decision. We want the player to have a good experience and not get frustrated, this isn't a game with multiplayer or high scores, so player skill isn't a core element - of course he should feel challenged, but not at the cost of pacing and immersion.

So to return to the jumping mechanic example, say we allow a fair bit of error margin to allow for a perfect jump and an even larger margin for failing but catching onto the ledge. Now there's a good chance for the player to feel success from a perfect jump and a feeling of first terror then relief if he falls short but saves himself at the last moment. You can add other director controlled factors to this to create more excitement. Since you know the speed and default jump distance of the player, you already know where he's going to land the moment he has pressed the jump button. But that works in reverse too - you can fudge the jump distance if it helps the drama, for instance having him land closer to the edge making him feel he "just barely made it" rather than "that was easy". By changing such a "simulated peril" setting you could even make different areas feel more or less threatening without *actually* putting the player in danger or changing the difficulty of the challenge.

Imagine a chase scene with the villain closing in from behind - let's try make it feel like a real action movie and have the player fail a jump and struggle to safety as the bad guy closes in. If we played this fair, the only way to achieve it would be to make a chasm almost as wide as the player's maximum jump distance, but that'll only mean a lot of players falling to their deaths first time around and starting the sequence over with some of the excitement already lost. Instead, force him to fail the jump - wherever he takes off from (within reason, not ten meters away) he'll always get across just far enough to catch on to the ledge. Yes, it's inconsistent game physics, but few players will analyze the scene enough not to think they themselves fumbled the leap, or that the gap was wider than they estimated.

Another archetypal scene that comes to mind is the opening sequence of Raiders of the Lost Ark where Indy during his escape the Peruvian temple rolls under a closing stone door, loses his fedora, reaches under the door and grabs it a fraction of a second before the door slams shut. To achieve that in game form would be very difficult under normal circumstances - most players would either fail or make it out easily, never feeling threatened. Applying the director/choreographer logic, you could adjust the speed of the stone door very closely to the player's actions. Let's assume the door doesn't move at a smooth constant speed since it's thousands of years old, so the moment Indy starts rolling underneath the door (the game already knowing he'll make it and by what margin), we can accelerate the closing speed to leave just a wide enough gap by the time he's through. Same thing with reaching for the hat, however confident he feels, he'll always manage in the nick of time, and the moment he's retracted his arm... we slam the door down with a loud bang.

Now that's cinematic!

Sorry for rambling, I wrote a bit about this approach to game design (which is one I would personally employ for most types of single-player games) for a not yet posted blog entry, so these ideas have been running through my mind for a while. I realize I didn't really address any of your questions, but I think the main point I wanted to get across was to use scripted events vs. player control together to create more immersion while allowing players with a wider range of skill levels to enjoy the game.

Also, you should totally watch the Making of Another World video - I actually wanted to recommend you an article on Another World that I believe was printed in Edge Magazine. I read it online, but I can't seem to find it on their site now. Maybe they put it in the subscription only section.

[mode7]

Thanks GarageGothic. This was really helpful. You bring up a good point and a classic game movie dilemma. In a game the player is no super hero. Unlike in a movie it's possible that he or she wont make it. There are two aproaches none is perfect.
1. Let the player die and replay the scene which is somehow unsatisfying and makes some trial and error (or try and die gameplay.

2. The second, your aproach, is using so called game time. E.g. give the player a feeling he or she is just in time. Many modern games do this. The big problem is it completely removes challenge.

The modern day equivalent to the first one would be active time events like in Resident Evil 4. Here they worked because they were rarely used, well choregraphed and you were able to replay them right away.
On they other hand they didn't work in Fahrenheit because they were 80% of the game.

I think a solution lies in between. Have them were its necessary and only there. Have auto save points. Give the player a fair chance to master. And have some "scripted" moments in between to keep the action high and the frustration low.

[GarageGothic]

Of course it should be possible to die, otherwise all sense of danger is lost the moment the player figures out the trick. I remember a scene in The Longest Journey at the witch's house that seemed very scary and had a great sense of urgence - that is, until you realized that even if you remained passive, the witch never got any closer to you. And yeah, that kinda ruined the terrifying atomosphere

What I was advocating though, wasn't zero danger or no challenge, but rather a way to make the game a bit more lenient in terms of quick reflexes and perfect handling without losing the more experienced players by coming off as too easy.

[Anian]

There's nothing more numbing and basically atmosphere killing than making a game (platformer) so precise and unforgiving that the player has to remember every every move they have to make. A lot of older platformers had this and I think it's just silly, there's no appreciation for art, level design or fun if you're basically playing A, A, B, UP, RIGHT etc. You can go memorize the phone book if you like that stuff.
But that's where going in the other direction is good, and why I agree with GG, if you let the game be just a tad forgiving and having a "game time" at some parts, it'll allow the player to get more into the game itself.

[mode7]

Thanks guys. This really helps. Actually here's a video of what I've done so far. Still too early to post this in the production thread. It works out really well:

http://www.youtube.com/watch?v=M8IePkB9GCw

[Snake]

There's nothing more numbing and basically atmosphere killing than making a game (platformer) so precise and unforgiving that the player has to remember every every move they have to make.

Like for example...
Skip to 7:08. It is all memorization.

Great start mode7! It is VERY Prince of Persia.

\\--EDIT--//
I remember the lightning part of that level being a lot longer than that

[arj0n]

Looks verry cool Mode7, awesome.

[mode7]

There's nothing more numbing and basically atmosphere killing than making a game (platformer) so precise and unforgiving that the player has to remember every every move they have to make.

Like for example...
Skip to 7:08. It is all memorization.

Great start mode7! It is VERY Prince of Persia.

\\--EDIT--//
I remember the lighting part of that level being a lot longer than that

Ah Battletoads. Always liked it. Bot on the other hand I never made it that far ^^. Many classics rely on memorization e.g. the brilliant Megaman 2. But these days are over - and this is a good thing.
Yeah it is very prince of persia - But unlike PoP these parts will not be the main gameplay element. Still till they shouldn't be redundant. So I'm still looking for advice in 2d level design.

[Wonkyth]

I've not read most of this thread, but the advice I can offer on level design is to get a bunch of games of a similar kind, play them through, and make note of all the bits you enjoyed the most. You might be surprised by which bits you actually enjoy more, since often it seems to be the simpler levels that allow more immersion.
This if what I've been doing while planning a couple of platformers I have in the works, and the best part is that it can all be done on paper or it's digital equivalent.

[cianty]

This look very promising, mode! I really like it!

I couldn't help noticing that the door seems a bit too small as the character is exactly the same size - shouldn't the door be slightly taller? Maybe not...

Anyways, it looks very cool.

[Dualnames]

This looks GREAT! Well done!

[Matti]

Wow, this looks great indeed, very Flashback-esque.

Another World (aka Out of this World) is one of my all-time favorites and inspires me a lot. Flashback was something else.. with more focus on jump'n'run elements and action. I found it a bit too boring after a while but it was cool nevertheless. Heart of Darkness was fun too but the story was crap and the game too childish.

I really these kind of games and I'm sure it's quite possible to mix platformer and adventure elements together to create something fun and unforgettable. I'm very displeased that developers so seldomly try out something innovative and mix genres together. Many old classics give perfect examples of great non-standard gameplay but few people dare to try something new (or old)..

Good luck with this, I'm quite interested how it will turn out.

[ddq]

Looks legit, though I would recommend making the way the camera follows the player a bit tighter. Not 1-to-1, but less waiting for the screen to shift slowly to where you want to look.

[theo]

Your test video looks great. Nice flow!
Very curious how this will turn out. Best of luck to you!