|Posted by Gilt [18.104.22.168] on 2004/03/03 22:17:49|
|So, I was playing Prince of Persia (excellent game, highly recommend it) the other day and was struck by the use of the camera by the level designers. Through thoughtful design, their camera's always accomplished at least one of three things, and in the best cases all three at the same time:
1) they showed the current path or puzzle to take, so that the player knew what needed to be done,
2) they are placed in such a way as to facilitate the execution of the path or puzzle,
3) and finally, they showed off the sceneary. (POP actually had a Spooge button that's primary goal was to show off the visuals)
Before I go further, I want to make clear that I'm only concerned with the camera as controlled by the level designer. However it acts when it's free floating or during combat doesn't concern me. Also, in general I am talking about how the camera reacts in game, not cutscenes or anything like that.
Anyway, the first point is important because it ensures that the player is never fustrated. In POP you pretty much always know what needs to be done. Often when you'd near a puzzle or path, the camera would turn towards it to highlight it. And then when you were at the puzzle, it'd position itself to make it easier to do the puzzle.
For example, right before you get the dagger, there's a series of jumping puzzles where you leap from pillar to pillar. When you were on one, you could see the next to jump to. And they'd be more or less positioned from right to left on the screen, so that you weren't accidently jumping into oblivion. Last but not least, it was always facing a giant set piece in the background that you could admire.
So how does this all affect FPS games? Do you ever think, or even care, about the camera (ie: player's view) when designing your levels or while playing? Do you point doors towards interesting elements of a new room? Which is better: to have a message saying "The button just opened a door somewhere", or have them viewable from one another? Do you put baddies on top of that magnificent tower you just slaved away on, just so players'll actually look up at it? Have you ever layed down a path largely because of the vantage point it gave? Or are some visuals part of the puzzle/reward structure; something players should have to work for, instead of simply given?
In my current map, I fussed about with 4 intermission cameras, taking the time to capture what I thought was a unique point of view from 4 locations in non-playable coordinates(ie. not reachable without noclip.)
Good Topic Gilt...
Do you ever think, or even care, about the camera (ie: player's view) when designing your levels or while playing?
Do you point doors towards interesting elements of a new room?
Sometimes, but in general I point doors so that they maximise vis blocking with respect to available "exits" from the room.
Which is better: to have a message saying "The button just opened a door somewhere", or have them viewable from one another?
I've always found that "somewhere" (except wrt secrets a la the entrance to Zig Vertigo) to be a tad vague and sometimes downright obscure. I prefer a direct statement "The door of the goat's cage just opened...have fun". However, viewable results are my fave. Not that the result has to be simply or immediately accessible.
Do you put baddies on top of that magnificent tower you just slaved away on, just so players'll actually look up at it?
Have you ever layed down a path largely because of the vantage point it gave?
Only if doing so fits with the gameflow, except in the case of secrets. Hence, the following "or" is inclusive not exclusive as far as I'm concerned.
Or are some visuals part of the puzzle/reward structure; something players should have to work for, instead of simply given?
In my current larger works, I'm trying to have secrets give visual as well as object rewards. Scraggy's intermissions (one of which is dare I say ,beautiful) are a reward to those that finish the level. I prefer to follow the Id protocol of using intermissions to "hint" at secrets (not that I did so in Usher Recompiled =)
There Is A Technique I
use you may find usefull. When you have a lot of text for the player to read, and you don't want to fuss with the QuakeC (and you don't like the time consuming slow scroll of the Id method) is to put all of your text on a jpeg, preferably superimposed on a represenative image of the level (say, the boss baddie in the next coming level for example) put it out of the way where the player can't stumble on it and an intermission camera fixed upon it. Takes a little tweaking, but it works.
Using demo files and keygrip, you can have a virtual graphic novel (aka Max Payne) to go with your episode.
just an aside, but I've always wished for a way to make a camera mappers could control in Quake... something you could use to set up an effect like in Zelda 64 and other such games where after hitting a button or otherwise triggering something the camera would take over your vision, and fly thru the level and show which door it is you just opened.
that was in zer iirc. and custents had entities for that too... was a pain in the butt to set up though. you needed to use trigger_impulse so you had to not only include the code for the cameras but also the new trigger...
That Effect Is Done Pretty Well
in Operation Urth Magic's first level -- Fat Controler has a detailed explanation on using the custom entities for OUM.
I don't care for camera tricks in first person games. The whole idea of first person games is that you feel like you are the character. It's like in the leaked Doom3, you walk around cleaning house, turn a corner and suddenly your eyes are ripped out of your sockets and fly around the room showing you what's going on. It may be more cinematic, but it ruins the consistency.
The Legend of Zelda does something else to help you identify. It's always been a 3rd person game, but Link never speaks. So you put the words into his mouth, playing out the dialog in your mind. It helps you become Link. I think it's a brilliant move.
First of all, i just finished the new prince of persia tonight, and i agree with the comments about the helpful camera. Sure, it's occasionally in the wrong place, but most of the time it's not.
As for 'camera placement' in quake levels:
* I try to create setpieces that the player will see when they come through a doorway. Coming through a door or narrow passage is one of the few times you can be pretty sure which way the player is looking, so i take advantage of it.
* I try to arrange it so that any big event happens in front of the player. Example: if you have to go find a button to open a door, I try to arrange it so that the button is on a balcony overlooking the door. You wander around, eventually come out onto the balcony, step on the button and see the door open on-screen.
(On a side note: the gold and silver keys are a special tool that you can use when you want the player to not be anywhere near the door when they find the key/button. Becuase the gold/silver doors are memorable and unique, and becuase the player already knows what the gold and silver keys do, it's safe in this case to make them travel a distance to get back to the door.)
i agree with pushplay on the silly doom3 cinematics. That really does hurt the immersion.
I Try To Put
all of the aesthetically pleasing aspects of my level design work well to the side where the players are only likely to notice if they are slowing down and taking a breather. It would ruin my too cool to give a damn Dylanesque veneer to build set pieces for maximum visual impact.
HeadThump: Good Point
I have a favoured twist to that; I often like to hide my best pieces in really dark shadows. That way I can feel smug and superior to the people who play my maps:
"Puny Fools!! The Candy is right in front of you, and YOU CAN'T SEE IT!!! MWA HA HA!!"
Anyway, so thumper, what are some of the techniques you like to use to minimize the visual information and impact in your maps (and what do you avoid doing, so as not to maximise it too much)?
Or more to the point, what are some of the specific ways that you subtley highlight you work, without breaking that all too important Too-cool-to-give-a-damn veneer and acting like a $5 hooker swinging around her muchachas? Where is the balancing point?
I need examples, screenshots, wireframe layouts!
Got To Give That Some Thought
'I need examples, screenshots, wireframe layouts!'
they are all good questions, and except for a speedmap, some prefabs I've made for Pushplay and Tronyn and texture work I shared with LTH, I've been holding back on the community.
Soon I'll show you a full fledged map. As I've decided to take on a less ambitious side project inspired by the Dapak series which is the subject of this weeks speedmap. Right now, I'm working away with Wally and Photoshop doing a Dapak texture redux, and once that is accomplished, I'll start on the map, which should be relatively small if I keep my G*DD**N ambitions in check.
I'll have to get back to you on that, I'm not sure how I want to approach it. Most of the Dylanesque thing was tounge and cheek but there is some truth to it. When I design my maps, I tend to make sure that the detail work is evenly spread across all of the rooms, and the areas that I'm less concerned I thicken the shadow volumes. My chief concern here is if you present to the player set pieces you may disturb the emmersive experience; say, you have the player sludge up a hill and kill knights as the preced, when they get to the top of that hill they are presented with a grand view of the fortress they are set out to conquer, at that point are they likely to assume the Lord of the Manor hired some fabulous architects and designers that happen to frame the design towards the hills vantage point (and not the enclosed entrance way)or are they more likely to say to themselves, 'this level designer is showing off?'
An intellegent camera is dependant of the atmosphere of the game you are playing. I personally feel for Quake I that there should never be an "out of body" experience in gameplay. The game was designed around face paced action, with the occasional scary bit. If id at the time were to take the camera out of the players control, it would have detracted from the pace of the game.
However on the same side, for Doom3 id is taking a more cinamatic effect. The atmosphere in this game looks to be much more creepy, and slower. So taking the player out of FPS mode can really give them a sense of their presence in the map. A sense of scale, a sense of their place in releation to the rest of whats around them. I think this will add tons to the creepiness in a map.
Of course doing DM maps only, I have little experience with drawing peoples eyes to different aspects of a maps design. In xl1dm7 I concentrated on the details and texture alignment. Of course tho no one notices them while they are DMing. I like to make certain set pieces that grabs attention, but none of them really lead the player to another part of the level.
You say they don't notice them whilst they are playing, but I bet you they would notice if you hadn't bothered.
Website copyright © 2002-2017 John Fitzgibbons. All posts are copyright their respective authors.