Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Game Design Theory / [STICKY] Level Design Theory:Tutorial/Tips

Author
Message
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 2nd Oct 2005 01:35
Level Design: Some Quick Tips


Introduction
Hi all. Level design is a funny-old job/activity, one that a lot of people forget about, and that there are few books on learning about. I thought I'd mention a few tips here to get people thinking about certain aspects of level design. Am I an 'expert'? Maybe I am, maybe I'm not, but I have some experience and I hope I can share it with people.

The Theme
You probably already know what your game type is (if not, why are you doing levels?) - whether it's an RPG, or a 3d platformer, or an FPS or an action/adventure or something else, you should know what the 'challenges' your player will face, and what 'tools' you have to use against the player.
If it's an RPG for example, maybe you can't jump. So have no gaps.
If it's a platformer, have lots of jumps, but maybe no NPCs.
If it's an FPS, you can have areas for fights, but maybe no hand to hand.
If it's an action/adventure, maybe you'll be pushing blocks and punching baddies (no long range combat).

Anyway - know your theme and know what tools you'll have to play with. Be it ranged combat, lockpicking, block pushing, button pressing, NPC talking, melee combat, short jumps, long jumps, swimming etc.

The Environment
Again, this may already be set out by your game. Maybe it's entirely set on floating islands, maybe it's underwater, maybe it's in a junkyard. Or maybe due to your story, the level must take place in the back garden of the main characters childhood friend. You should know roughly where it'll be. But even if you're left with a seemingly dull environment, don't despair - there's fun to be had in almost any environment. (Except offices, they're the bane of my level designing life. Why people think it would be a good idea to set a level in an office...)

The Three Mantras: Gating, Landmarks, Freedom
These are three things you should try pay attention to.

Gating
Gating is the art and science of stopping the player from going where you don't want them to go (yet). This may be walls around your level, locked doors, cliffs (walls, textured differently), parked cars, roadblocks, magic forcefields, level restrictions, door fees, fences and even large barking dogs with pointy teeth.
There's two kinds of gating - puzzle gating, and permanent gating.
Puzzle gating is where the aim of the player is to get past your gate - it's the way they progress. They may have to pay the guard 400 gold to progress, or they may have to pick the lock to the door. These points are often 'bottlenecks' - places the player must go through to progress. These are points where plot is most often fed to the player - think of boss fights, mountain passes, Hallmark quests, hub levels and scary old men in taverns handing out jobs. To progress, you MUST go through these areas.
Permanent gating is just the walls of your world. If you never stop the player, they'll go on forever and either hit the edge of your content and drop off the end, or get very bored. Walls, fences, cliffs and other blocking devices are your friends. However, they are also quite hackneyed. Try to create new and interesting ways to stop your player - burning forests, radioactive waste, quicksand, exploding prison-collars, even really strong winds can be a new way of saying to the player gently but firmly - "This is not the way to go."

Landmarks
Players (usually) don't like getting lost - NEVER underestimate the players ability to get lost in levels you think are the most straighforward things ever seen. One branch in a forest path and they can get turned around - worse if there's a fight that disorientates them. Make sure you have a nice big landmark that can be seen from most parts of your level as a good orientator. Remember the film Labyrinth? Often she could see the Goblin Kings Castle from a long way away - she wasn't near it, but she knew the right way to go.
Also, landmarks are a good way of telling the player over and over - "This is where you're going."
So whether it's the evil witches tower, the last escape rocket launchpad, the mighty pyramid or the magical fountain in the middle of the forest - it helps to keep the players mind on the ultimate goal as well as the immediate goal.

Freedom
So I've just told you to keep the player hedged in between walls, cliffs and fences - well here's the tricky part: do it without the player realising. Games like Half Life 2 do it marvelously - it's a blindingly linear game, but because of some clever level design, it seems very free. Make sure that the way that the player should go is interesting looking - make non-essential routes look boring. Place clever lighting, put signs, add movement to draw the player's eye. Learn the direction the player will take through your level. Maybe the right way is straight on, but a fight might turn them around - have some enemies retreat the way you want them to go - and they may chase them! You're leading the player by the nose, and hoping they won't notice.
Don't get me wrong - have multiple paths, or even free roaming landscapes if you like, but most players like some guidance. They like to be told gently where the 'best' place to go is. Even roaming games like Morrowind guided you down the path of most interesting plot - but you could ignore it if you pleased.


So there you go, a few tips from a crazy Tinkergirl. If you find these interesting to read, and post accordingly, I might try to give away a few more. If people know all of this already, then I'll try not to waste any more of your time
Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 2nd Oct 2005 01:59
This is more of actual "game design" rather then "level design". (Hell forget it, both are interconnected I guess)

Still, some good tips there Tinker. Will brush on a few of these in a tutorial of my own I'm writing.


The future is here, and I can't afford it.
Chenak
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 2nd Oct 2005 02:26
I'm very poor at level design, but this should help quite a lot, thanks Looking forward to more tips
The admiral
21
Years of Service
User Offline
Joined: 29th Aug 2002
Location:
Posted: 3rd Oct 2005 04:56
What sort of tools would you suggest for making levels??

The admiral
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 3rd Oct 2005 12:39
First and foremost: pen/pencil and paper.
Sketch, scribble, write and scrawl any and all you can first. Sure, it's the stupid answer, but it's the most important one. The brainstorming stage is the most exciting, but you'll never remember it all. Write it down fast!

Secondly: Wordprocessing program.
Take all your work and make a document about it. You want it to make sense and there's nothing like trying to write a full document to show you the holes in it.

Third: More paper (or Visio).
Draw it, draw it 3d if you can, or plan view if you can't. Some people use Microsoft Visio for this, but basically you're mapping your level. Beware that if you map in 2d, your level may suffer from flatness. Vertical interest is a buzzword for a reason - it's more interesting.

Lastly: Your level editor.
Personally, I use whatever I've been given. Max is good, and industry standard, however it is very expensive. GMax is a nice start, but once you've 'tasted' Max it can be hard to go back. I've also used UnrealED but not for DB - that's nice and can get your stuff up and running nice and quick.
But as the song goes - if you can't be with the one you love, love the one you're with. (I.e. - just try to get the best out of the tool you have. It may be slower to learn a new one, than to do it with what you have.)
Short answer - Max if you can afford it. It's ace.
Bizar Guy
19
Years of Service
User Offline
Joined: 20th Apr 2005
Location: Bostonland
Posted: 3rd Oct 2005 22:15
How much does max cost?

Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 3rd Oct 2005 23:02
Max costs 2000-3000$ dont bother with it.

USE CSHOP PEOPLE.

Geez.


The future is here, and I can't afford it.
Chenak
21
Years of Service
User Offline
Joined: 13th Sep 2002
Location: United Kingdom
Posted: 3rd Oct 2005 23:23
You can purchase the student version from around $300 i think, I use my uni's version from time to time and it is brilliant. But 3k$ is a good investment if you are planning on selling 3d models, not such a good investment (unless you happen to be rich) if you are only planning on using it for hobby games in my opinion.

3d world studio (cshop) is good for the hobbiest though, it only lacks some decent uvmapping.

Personally I got 3d world studio for maps, milkshape and lithunwrap for meshes, gimp for 2d art and characterfx for animations, its all you need really. Could use gmax but theres some very unclear rubbish about licensing.

Anyway back on topic, the advice is awesome thanks it should help me a lot.
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 4th Oct 2005 01:07
FPS Level Design (Single-Player)


Introduction
FPSs are a breed unto themselves, and require a particular kind of level design. Whether your game is a 'run and gun' or a tactical, thinking-man's shooter, chances are your level will have things common to other FPSs levels - or at least, considerations.

Level Structure
There are a couple of overall level structures that you can consider for your level, some are easier to make 'half decent' than others, though all can be done well.

The Hub

The Hub, is a style of level where you return to a particular area more than once during the course of the level. It will probably have branching areas off this central hub space, these may be sealed off initially, or you may leave them open for the player to explore. Note - there is a difference between a true Hub level, and one that just happens to have lots of non-essential branches. A true Hub level will require the player to revisit the area more than once.

Example of 'The Hub'
From the central hall of the pyramid, explore the three crypts and bring back the three fragments of "Ra's Sun Disc" to open the underground chamber of the Mummy.
Here, you have the central chamber being the 'Hub' with four offshoots - one is sealed from the player, but the other three may be entered at any time.

Pro's
By reusing geometry (level space) you can get a lot more play time out of a smallish level because you backtrack.
You can mix it up by having areas you've already visited change on further visits - maybe a pillar collapses, or zombies break down the doors.
You can give the player a 'base of operations' at the Hub point, somewhere to stash or collect ammo and other pickups.

Con's
It can get dull - players like to see new stuff, so unless you're careful to add changes, or change the way enemies attack while you're backtracking, they can quickly get into a "Seen it before" mentality.
Care must be taken to avoid a specifically "Collect three items" syndrome as seen above. Unless that's how your levels work, of course.

Linear Levels

A Linear level is one where the player goes forward all the time. The level, though it may make many twists and turns, never actually branches and if stretched out, would be a straight line from start to end. There's no backtracking, no choices of which way to go, and no real thought required.
Note, however, that Linear levels may (and should aim to, for vertical interest) loop back on themselves to see where they've already been - good examples being fighting through a canyon, up through a cave system, then across a bridge you went under earlier. It's still a straight line, but it's always nice to see where you were, from a different angle.

Example of 'Linear'
The Sergeant needs you to fight through the trenches, blow through the barrier on the other side and secure the bunker for the dropship to land next to. The only way is onwards!
With trenches providing a simple 'corridor' for the player to make their way through, even blockages can be gone around by going through simple bunkers and rooms off to the side. It's still a simple path all the way to the bunker at the end - the corridor just looks different.

Pro's
Easy to do - this is probably the simplest kind of level to do reasonably. Pick your challenges that you want to throw at the player and arrange them in a line - then twist that line, add some interesting geometry and you've got a level.
You can do lots of 'one shot wonders' with level events - collapsing bridges, exploding statues, crashing spaceships - you know which direction the player is coming from, so you can set these up to be superdramatic.

Con's
Can seem restricting. Modern gamers are getting very wary of linear gameplay - they don't like it. But if you do it well, they might not notice. HL2 was linear, but their 'corridor of gameplay' was so well hidden that most people never felt restricted.
No reuse of assets. Players don't get to learn an area, they are always new. This is both good and bad. It's expensive in level building terms - it's cheap to just use the same level parts over again

Branching Levels

The branching level has several ways to your objective. Maybe one way is a stealth way, maybe one requires explosives, maybe if you keep your AI buddy alive he'll open a door, maybe "You take the high road and I'll take the low road". Whatever way, branching a level gives options to the player and depending on the strategic worth of those options, may cater the gameplay to their style.

Example of 'Branching'
"Here, listen up - if you want to get into the castle there are two possible routes: you could try to get in through the front gate, even though it would be suicide, or I hear that there's a secret tunnel through the moat and into the kitchens. Of course, you'll have to actually get TO the castle first..."
In this scenario, the player gets the option of a frontal assault (and told that it'll be hard) or told that there may be a sneakier route available. Some players will jump at the chance of a big fight, whereas others will prefer to stealth in. This means that you can be really vicious on the frontal attack, and really really sneaky on the stealth approach. Because that's what they've chosen, and that's what they're expecting.

Pro's
Lets the player feel clever - if there's a secret entrance, or a stealthy way, or even just a not as obvious path, the player will feel smart if they find and use it.
Lets you cater the gameplay to their style - they 'choose' the type of gameplay, so give it to them!
Keeps it interesting, and choices are almost always good. Avoids dreaded 'linearity'.

Con's
Harder to balance. It's not as easy to ensure that both (or all) paths are as challenging - if you have one path harder, make sure there's goodies or another reward for the player who takes that route.
Some geometry is 'lost' - the average player, who plays your level will never see the other path. They'll never see the pretty graphics, your cool effects, your funky eventy. It's heartrending, but it's the way it is.


So there's the three simplest level types - some others are 'The Arena' - where it's pretty much one big area with many fights (drop your enemies in where the player can't see them), 'The Web' - kind of like the Hub, but your player works around a central area and only gets in when they've completed an objective, then gains access to all the previously visited areas.

Also, you can probably see that a lot of these will work with other genres - platformers in particular are fond of Hub levels and Branchers, while Crash Bandicoot was pretty much all Linear levels.

Once again, I hope I've helped, not hindered and I really hope I haven't preached
Peter H
20
Years of Service
User Offline
Joined: 20th Feb 2004
Location: Witness Protection Program
Posted: 4th Oct 2005 01:53
wow, thanks for the tips!

has enlightened me.

"We make the worst games in the universe."

Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 4th Oct 2005 02:31
Nice updates Tinker.

Quote: "3d world studio (cshop) is good for the hobbiest though, it only lacks some decent uvmapping."


Not if you're a retard who doesn't know what a level's supposed to look like.


The future is here, and I can't afford it.
Baggers
19
Years of Service
User Offline
Joined: 31st May 2004
Location: Yonder over dem dere hills
Posted: 4th Oct 2005 11:21
Some very solid advice here, thanks. We need more of this kind of theory in this board.

Peter H
20
Years of Service
User Offline
Joined: 20th Feb 2004
Location: Witness Protection Program
Posted: 6th Oct 2005 04:22
Quote: "Some geometry is 'lost' - the average player, who plays your level will never see the other path. They'll never see the pretty graphics, your cool effects, your funky eventy. It's heartrending, but it's the way it is."

...oh man... that branching thing was sounding real cool until then

i mean...i just can't stand the thought of not displaying every painfully created work of art to the player

that would be like... like... well... it would stink

"We make the worst games in the universe."

Benjamin
21
Years of Service
User Offline
Joined: 24th Nov 2002
Location: France
Posted: 6th Oct 2005 12:33 Edited at: 6th Oct 2005 12:49
Quote: "Some geometry is 'lost' - the average player, who plays your level will never see the other path. They'll never see the pretty graphics, your cool effects, your funky eventy. It's heartrending, but it's the way it is"

Actually I believe this is a pro, as it will make them want to play it again to see what is down that path. So it increases playability.

Tempest - P2P UDP Multiplayer Plugin - 80%
Want to try the beta? E-mail me.
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 6th Oct 2005 13:20
Ah, but the average player, never will.

Your completist player certainly will, but if I remember right, it's something like 70% of players never get to the end of a game.

Don't get me wrong - the proportion of your players who are completist will love every non-essential nook and cranny you can fit in, and they'll look for your hidden secrets and alternative paths. And replayability is an excellent thing to strive for - things like time-challenges, randomised levels and multiplayer being the big winners here.

But if you're taking the hard line - bang for buck, then many players will only play a level the number of times it takes them to get to the end and on to the next one.

Depends on the playerbase you're aiming for
Peter H
20
Years of Service
User Offline
Joined: 20th Feb 2004
Location: Witness Protection Program
Posted: 6th Oct 2005 17:49 Edited at: 6th Oct 2005 17:51
Quote: "completist"

lol, so that's what i'm called...

i'm the person who (if i like it) plays the game 3 times over to find every little location i haven't seen

"Indiana Jones and the Infernal Machine" was a great game for this...

i don't know why but i'd have to say that was my favourite game...played it through 3 times at least

Quote: "Actually I believe this is a pro"

i agree, the thought of doing it is painfull but it would be worth it to increase the quality of the game...

"We make the worst games in the universe."

Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 10th Oct 2005 17:56
Game Mechanics, Puzzles and Tutorials.


Game Mechanic vs Puzzle
The difference between game mechanic and puzzle is a relatively subtle one, but an important one. You could class a puzzle as a subset of game mechanics.

Game Mechanic
A game mechanic is a tool that you and the player use. It might be something like jumping in a platformer, or it might be 'collect gold keys to open chests' in a cartoon adventure game, or it could be 'smash the boarded up door with the crowbar to progress' in a FPS.

Puzzle
A puzzle is a 'one off' of a particular example of one or more game mechanics. Examples of puzzles could be 'time your jump over the lava geyser' in a platformer, or 'collect 3 platinum keys to open the special chest' in a cartoon adventure game, or 'smash open 5 windows within the time limit to let the poison gas escape' in an FPS.

Note: If you have a puzzle, but it does not use a well known game mechanic, the player may be confused, annoyed, and likely to abandon your game. They'll feel like the game has been teaching them all the mechanics, only to then spring a challenge they haven't been prepared for. They usually don't like that.

Double Note: The worst example of this has to be end-levels or baddies that use completely different game mechanics to the rest of the game - a classic example being Sly Cooper. The game was a cartoon stealth-platformer in third person, but the last level was a driving/shooting 'mini-game'. Why 'train' your player throughout the game, only to not give them a final test of their practiced abilities?

The important thing, is that a game mechanic should be used frequently throughout your game, taught early on, and its complexity increased as you progress. Otherwise, it's not a game mechanic, it's a puzzle, and a confusing one.


The Tutorial
Having a tutorial is often almost essential to a game. Next to no-one* will read your manual (if you have one), they'll want to jump straight into the game to play.

Some games might not even seem like they have a tutorial - but they do. They'll sneakily introduce the game's main mechanics in a safe, non-game critical environment. These ninja-tutorials (as I'll now call them are some of the best - people learn best when they're not being patronised.

A classic tutorial is like what you may find in, say, Call of Duty with the seperate rooms and instructions for each weapon (also the seperate tutorial/training for HL). A Ninja-Tutorial is more like the HL2 first few levels, and the first area of Jak and Daxter.

The introduction of game mechanics is the second most important feature of a tutorial - the first being familiarisation with the controls. Each of your game mechanics should be slowly introduced to the player it its simplest form.

Eg.
Simple jumping up and down stairs, in a platformer.
Collect the wooden key next to the chest, to open the chest, in a cartoon adventure game.
Have the crowbar pickup in a room with only one new exit - boarded up with a single plank, in an FPS. (May as well put some breakable crates in too, to get the player used to the idea.)

Introducing the controls
Firstly, don't reinvent the wheel unless you really do have a new kind of terrain. If your game is a similar genre to another well known game, then think about being inspired (copy is such a dirty word) their control setup. Is there really any use in having a new jump key? Be aware of the players assumptions of how the controls will work - if random people who try to play your game instantly try to use WASD and you're using the arrowkeys, maybe you should think about changing.
But in the end, you're going to have to introduce the controls anyway - in case complete newbies pick up your game. A simple sprite on the screen with "WASD to move" written on it that fades off after the first minute of play, will go a long way to help.


Summary

A Game Mechanic is a tool that the player uses throughout the game to progress.
A puzzle is a specific instance of one or more game mechanics.
A Tutorial will introduce controls and game mechanics.
Ninja-Tutorials are harder to do, but nicer to play.
Classic-Tuturials are easier to do, but can be patronising.
Be Inspired by the genres leading games and their control setups.
Always Introduce Controls as if the player has never used them - they may not have.


*Except that rare breed of people who actually like reading manuals. I am one of these mutants, but we are in the minority. That's why so many manuals for games these days are 50% developer and publisher credits, 30% adverts for other games and a mere 2-3 pages of actual instructions. *sigh* Gone are the days of character bios and backstory.
Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 10th Oct 2005 18:14 Edited at: 10th Oct 2005 18:14
It's kinda lost it's "level design" orientation, but still good basic advice.


The future is here, and I can't afford it.
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 10th Oct 2005 18:22
Thanks Megaton. I often think of the Tutorial as the first level of the game, you see.
It depends on the fine line between Game and Level design, as you sort of pointed out much earlier.
Zergei
19
Years of Service
User Offline
Joined: 9th Feb 2005
Location: Everywhere
Posted: 10th Oct 2005 18:30
Quote: ""Some geometry is 'lost' - the average player, who plays your level will never see the other path. They'll never see the pretty graphics, your cool effects, your funky eventy. It's heartrending, but it's the way it is.""


An option is to let the player decide how to enter the castle (following the example above), but that the escape route must be the one not chosen to enter, so if he chose to get in throught the sneak in mission, they escape would be throught the front door.
However this shouldn't be done always as it will end up being predictable.

Oh, and not to brag about, but i play at least 20 times games i really like, such as Zelda and the Zero series. I know, quite insane
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 10th Oct 2005 19:05
In the case of the escape route being the alternative entrance, this makes it even harder to balance in most cases. Puzzles that work in one direction, may not work in another (a wall becomes a simple cliff, a door locked from one side becomes easy to get past, jumping up ruined stairs becomes a simple scramble down them). As you say, Zergei, it does depend on the level, and should never be overused.

Games companies often don't mind if you're completist or not - you pay the same for the game either way

Some people think completists are sad, others see them as the only people who truly complete a game.

The real view is, as long as you have fun, it doesn't matter
tpfkat
18
Years of Service
User Offline
Joined: 1st Sep 2005
Location: lancashire/uk
Posted: 10th Oct 2005 20:30
i like games like the theif series,always seems to be hidden entrances and exits,an idea im wanting to implement on but can make designing a level a pain in the aaaarrrrrrr jim lad
db programmer 12
18
Years of Service
User Offline
Joined: 7th Sep 2005
Location:
Posted: 16th Oct 2005 01:01
I can make my own levels personally. I have a brain and will that's all I need.

Warning Pointless Signature Below:
Runescape - Current Level 37 EA Isn't Evil They Just Buy Everything And Plan to Rule the World! 2D Shouldn't Die, It Rocks!!!! DBC Level - Advanced Beginner
Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 16th Oct 2005 04:31 Edited at: 16th Oct 2005 04:31
Kudos for you then.


The future is here, and I can't afford it.
Zergei
19
Years of Service
User Offline
Joined: 9th Feb 2005
Location: Everywhere
Posted: 16th Oct 2005 04:31
Lucky you
DrewG
18
Years of Service
User Offline
Joined: 25th Aug 2005
Location:
Posted: 17th Nov 2005 06:49
Nice stuff Tinker, MODS MAKE THIS A USEFUL STICKY!!!!!!

Wiggett
20
Years of Service
User Offline
Joined: 31st May 2003
Location: Australia
Posted: 17th Nov 2005 15:10
very informative tinker, I will pass this thread on to my level editor programmer.

Quote: "*sigh* Gone are the days of character bios and backstory."


gone are the days of decent games ruling the market.

dared
19
Years of Service
User Offline
Joined: 20th Nov 2004
Location:
Posted: 28th Dec 2005 12:39
Hl2 Was kinda freeform: you never HAD to go to the red barn, you never HAD to fart during gameplay either.

Who Loves coding?
Any1?
Oh forget it!
jimenex
18
Years of Service
User Offline
Joined: 21st Nov 2005
Location: New York City
Posted: 28th Dec 2005 22:28
Thanks Tinkergirl, just giving names to these in-game dynamics is important (Even if they seem obvious to some), as the terms catch on, it makes it easier/quicker to talk about complex problems and challenges without having to define everything repeatedly. I think of the time saved and immediate understanding a simple acronym like "FPS" facilitates. I hadn't heard the term "hub" used in a game design context. I'll use it from now to indicate: "...a style of level where you return to a particular area more than once during the course of the level..." instead of typing endlessly.

"Friends may come and go, but enemies accumulate."
Deadwords
19
Years of Service
User Offline
Joined: 2nd Feb 2005
Location: Canada
Posted: 29th Dec 2005 05:03
Thanks Tinkergirl. These tips are fundamentals in Game/Level Design, and these are forgotten often in amateur games, and a few times professional games.

Skalex - Nobody can ear you scream ... you're on a forum!
Medieval Coder
18
Years of Service
User Offline
Joined: 31st Dec 2005
Location:
Posted: 1st Jan 2006 15:15
Wow nice job this should be stickied!!!
This helped me a ton!
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 4th Jan 2006 22:32
The Funnel - A Combat Setpiece


Right, though hard graft and luck, I've been priveledged to 'discover' (not on my own though) a successful fight setup for FPS's. It's just one example of a fight setup, but used sparingly, it can create a very intense, climactic faceoff. I give you, the Funnel. (Red is enemy cover, green is player cover).



When it's appropriate:
Useful in an FPS that has cover using AI (or AI who can be scripted to go to points, and then fire from crouching), and that has a good mid-range gun play. If your game cannot use cover, or your game is mostly a sniping or handtohand game, then this may not be for you.

The geometry:
The general area is shaped like a funnel, with the end that the player enters from having only one entrance, and the enemy end having two (or possibly three) entrances. One entrance from each edge, and possibly one leading to the 'balcony' area at the rear.

The cover:
Very important, this. The player has some cover at the end that they enter from. This should be fairly good cover - enough that giving it up should be quite a risk. The cover that the enemies have comes in two flavours - the low, long wall in the middle of the top of the funnel (with two larger cover areas at either end) and the pillar cover on the balcony.

The setup:
The player enters from the narrow end, and enemies feed in from the two top entrances (left and right), with some enemies on the balcony. (You may choose to have a sniper or equivalent). Enemies should keep to their respective cover, popping out to keep the player pinned down.

What happens:
Usually, the player sticks to their cover for some time, pinned down by fire from three directions - left, right and above (the balcony). Usually, the player doesn't push forward until the flood of enemies has slowed down, because of the large 'killzone' of empty space between the two cover-rich areas. This results in the player dashing into cover between barrages, and to reload. Picking off enemies as they pop their heads up.

Why it's good:
One thing you should note - while it's often VERY dangerous to have the player attacked from several angles (nothing more annoying than being shot in the back) - this situation gets away with it because all the enemies are within the players field of view. If the player faced forwards, they could see all the enemies - there's no 'where did that come from??'. However, they can't just keep one doorway or corridor covered and shoot anything that approaches - they must keep on their toes. Result: a tense, but never unfair pitched battle.

How do you use it:
Obviously, the image above is the Funnel in its most basic, vanilla form. Think instead what could use that setup - maybe the Funnel is reception area of a office building, with the lower cover the reception desk and the balcony the cafe area above. Maybe the Funnel is set in the engine room of the alien spaceship, with the balcony accessed by service ladders and computer stations forming the lower cover. You get the idea. You can put the Funnel in many environments - maybe have the player fighting uphill towards the balcony for an added sense of difficulty. Throw in some smoke grenades, a sniper on the balcony (or someone with a rocket launcher) and maybe a few exploding barrels, and you have yourself a tough but fair fight that your players should remember

*Yes, I've used the Funnel before, in a commercial product. It was used twice and both fights were consistantly rated the most exciting and memorable fights of the game. (Your milage may vary)
Joh
19
Years of Service
User Offline
Joined: 5th Jul 2004
Location: Malaysia
Posted: 5th Jan 2006 00:51
Sweet. More! More! MORE!!!

Mind if I upload a pdf of this up here, for easier offline reference ?
Deadwords
19
Years of Service
User Offline
Joined: 2nd Feb 2005
Location: Canada
Posted: 5th Jan 2006 00:59
More would be great. Thanks Thinkergirl!

Skalex - Nobody can ear you scream ... you're on a forum!
Megaton Cat
20
Years of Service
User Offline
Joined: 24th Aug 2003
Location: Toronto, Canada
Posted: 5th Jan 2006 05:03 Edited at: 5th Jan 2006 05:03
Great stuff Tinker!!

We shall soon forge what is to become the greatest Level Design Resource tutorial known to DBP-kind.


The cat era has begun. Shedding light onto out dark and humourless world.
jasonhtml
20
Years of Service
User Offline
Joined: 20th Mar 2004
Location: OC, California, USA
Posted: 6th Jan 2006 01:52
wow, this is very helpful... and about the game mechanics and puzzles, that will be very useful because i have a lot of puzzles in my game (and it actually saved me from a puzzle i was designing that a person wasn't tested earlier, so im making simpler, previous ones of the same type)

Deathquest (MMORPG)
Deathquest Thread: http://forum.thegamecreators.com/?m=forum_view&t=61108&b=8
DD Studios Website: http://www.geocities.com/jasonhtml/
Bizar Guy
19
Years of Service
User Offline
Joined: 20th Apr 2005
Location: Bostonland
Posted: 6th Jan 2006 04:06
This thread is great, I knew most of what was on it, but I've also learned a whole lot I didn't or never though about before.

Just one question, why the @#%! isn't this a sticky?!

Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 6th Jan 2006 14:06
@jasonhtml Glad to hear you managed to catch a puzzle that wasn't taught beforehand. I'm sure it'll make a difference to the players of your game

I'll try to do another setpiece example - any genre requests? (I was thinking 3d cartoon platformer.)
Van B
Moderator
21
Years of Service
User Offline
Joined: 8th Oct 2002
Location: Sunnyvale
Posted: 6th Jan 2006 15:24
Very interesting thread Tinkergirl, I'm inclined to agree with Bizar that this should really be sticky.

One thing that I'd like to see covered is ways to make the level seem endless while keeping the player in a relatively small area. Like when using terrain it can be difficult to get the heightmap right so that your not seeing the level edges - It seems most of my heightmaps are either islands, or inverted islands!, some tips for better methods would be appreciated.


Van-B

Put away, those fiery biscuits!
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 6th Jan 2006 16:59
Thankyou for the sticky - I'll try to do something on the 'smoke and mirrors' of restricted wildernesses, but maybe Megaton is going to touch on that in his 'hands on' stuff. Nothing wrong with us both helping though.
Bizar Guy
19
Years of Service
User Offline
Joined: 20th Apr 2005
Location: Bostonland
Posted: 7th Jan 2006 05:21
Quote: "I'll try to do another setpiece example - any genre requests? (I was thinking 3d cartoon platformer.) "

Yes! I'd love to see an example for that!!

Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 8th Jan 2006 02:50
Platformer Design Setpiece: The Volcano

I've used this before and it's fairly common in 3d platformers - you'll have seen it be used in things like Crash Bandicoot (the full 3d ones on the PS2) and Jak and Daxter.

The basic premise, is you're in a vaguely cylindrical area and you work your way upwards inside that cylinder - but there is a rising, deadly hazard. Here's an example of it:



How it plays:

Here, the player enters the 'Volcano' from the bottom of the screen at the start point. There are two ways the player can go - to 1, or 1a. At this point, the orange lava level is not rising, so the player can explore somewhat. On 1a, there is a pickup, but nothing more. 1a is a dead-end. You'll see why in a moment.

The player will then follow the platforms in the numerical order - it is always obvious which platform is next. Once they reach the large platform in the centre - platform 6 - 'something' happens and the deadly floor/lava starts to rise slowly, and the start door seals off.

The player will realise that up is their only option, and head for platforms 7 and 8. They may, at this point, be in a panic and suffering a little from vertigo (they're fairly high up, and the floor is moving). If they fail the jump from 8 to 9 - they will land on 1a. This is the REAL reason for 1a. They will have enough time to leap back to the start, and on to 1 etc.

They exit the volcano at platform 13 - usually accompanied by the lava level either stopping, or the exit sealing up behind you.

How you can change it:

Leaping up the inside of the Volcano can be made as long, or as short as you like - the example here is quite short. Depending on the size of your volcano, you could have several central columns, or just one - you could even have fights along the way!

It doesn't have to be a volcano - and it doesn't have to be lava. It could be water (if your character can't swim), it could be poison gas, it could be toxic waste, it could even be maggots! (Ew!)

Adjusting the difficulty:

Always start these kinds of areas very simple - initially each platform should be obvious as to whether it's the next platform or not. There shouldn't be any "leaps of faith" in such a dangerous environment.

However, if you've gone for a bigger 'volcano' setup, then feel free to add moving platforms (adds extra tension when the player has to decide if they want to wait for the next time it's close!), shaky-falling platforms, dangerous but quicker shortcuts, timed hazards (venting steam, lava geysers, laser beams etc) and appropriate enemies guarding platforms (fire sprites, robots, etc).

If you want to make it a little more interesting, special valves or similar could temporarily lower or stop the rising level - giving the player a brief breathing space.

You'll need to test the speed of the rising level quite carefully, and measure out the height gained per jumping second, to the increase in height of your lava. The shorter your volcano, the less this needs to be fine tuned - you may find it easier to create several connected 'volcano's with individual rising speeds, rather than try to balance one huge one.

A last note:

Try not to be tempted to put pickups in areas that the player may miss the first time through the volcano, unless your game is designed with multiple level replaying in mind. Players will get annoyed if they instinctively pick the 'wrong' path the first time and miss out on pickups and can't go back to get them.

Also, if your game has save points - make sure you sprinkle them liberally at the start, and end of the volcano - your players will not thank you if you make them do the area before the volcano several times if they fail on it - and they'll be particularly cheesed off if you make them do the volcano again if they get hit by a baddie the moment they finally complete it.

However - the volcano setup is pretty much guaranteed to get your players blood pumping, and in 3d platformers, where the pace is often defined by the player's willingness to explore, it creates urgency. The sense of achievement and relief once they walk through the top exit will be great, and they may be on a high. It's a good level finale.
Bizar Guy
19
Years of Service
User Offline
Joined: 20th Apr 2005
Location: Bostonland
Posted: 8th Jan 2006 22:41
Oooh... Very nice! More, more, more!

What about general level layouts in a 3d platformer, based on collecting multiple items around the level (ex: stars in mario64)?
I know it can be done many ways, but something as a general rule to it would be great.

Tifu
19
Years of Service
User Offline
Joined: 13th Sep 2004
Location: Scotland
Posted: 14th Jan 2006 00:43
Nice notes. Puts me in the mood to make a platformer

Oddmind
19
Years of Service
User Offline
Joined: 20th Jun 2004
Location: Atlanta, Georgia
Posted: 16th Jan 2006 23:14 Edited at: 16th Jan 2006 23:16
Ha! very nice tinker, i also noticed that when you were talking about the end of sly cooper... That was a total rip off of the last scene of metal gear solid 1 >. Those copy cats.

I love this thread tho, great stuff man!

EDIT: i forgot that i had seen something on the history channel about a battle that took place in the 1500's between france and england. France was gonna get pwned but because of the funnel shape that england had to go through (no cover) there shear numbers was their weak point and the ended up being trampled and getting stuck in the mud, literally... while the french came up and stabbed them with little daggers.

formerly KrazyJimmy
Yskonyn
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: Netherlands
Posted: 10th Feb 2006 00:57
Great resource. As Tifu said: It makes me get into the mood doing a funny platformer!

Yskonyn -
"It's better to wish down here you were up, then to wish up there you were down."
"The ONLY time you have too much fuel on board is when you are on fire."
Tinkergirl
20
Years of Service
User Offline
Joined: 1st Jul 2003
Location: United Kingdom
Posted: 11th Feb 2006 00:46 Edited at: 11th Feb 2006 00:46
Hiding World Edges
(As requested)


It can be very difficult to gate the player in a way that still gives the impression of something 'over there' - that there is a rest of the world just around the corner and that where the player is, is not actually the entire world. (Even though, it may well be.)

The image below gives one non-urban example of how you might do this - the aim being to keep the playing area small, but the sense of scale great.



How it works:

People can tell that a wall is a wall, very quickly - they dismiss it and it's instantly boring. Not good.
Many layers (of non-regular shapes) creates a parrallax effect as you move and with each step a little more of the hidden distance is revealed - it remains mildly interesting.
The concept of "what's around that corner" also keeps an environment interesting - if they don't know where the world ends, they don't know what might be there, or what might appear from there. It's unknown - and unknown is interesting.

While this environment (and I've kept the playing area purposely blank - that's not the focus here) is tiny and quite obviously limited, linking a few areas like this (through tunnels or bridges) instantly helps create a hub or web style level (see above) and generates an interesting play space.

Note also, that in this environment, the play area is 'U' shaped - you cannot see it entirely from one corner to the other. Elaborate on this in your own levels - create more bends and add blockages that must be negotiated and your play space will seem bigger than it is.

Keeping it cheap:

While this may seem very wasteful on polygons, most of the 'distance' features can be one sided, lower resolution textures (as the player will never see it up close) or even just 2d plains.

In the images, the 'backs' of cliffs have polygons, but you should aim to remove all polygons that the player will never see.

How else can it be used:

The same technique can be used in city environments to even greater effect - a couple of chain link fences leading out to 'just out of reach' busy main roads (really just a tiny segment with randomised traffic looping past) will do wonders to make your players think they're in a living breathing city, but without the hassle of having them getting run over by cars, or bumping into pedestrians.

Other 'gating' mechanisms in city areas:
-Dumpsters in alleyways
-Strategically parked vehicles (vans etc)
-Scaffolding
-Roadworks
-Raised bridges (in a London or medieval setting)

Other systems in rural areas:
-Fallen trees
-Ruined building walls
-Thorny bushes
-Dry river gullies
-Fences/Dry Stane Dykes (that's stone walls to normal people)
-Rockfalls

A few ideas:
-Have an airship flying around the cliffs in the distance, adding interest and movement.
-Let the player play the opposite side of the cliffs later in the game (with a low poly version of the area mocked up) and the feeling of being in a large area will be magnified.

Final Thought
The best way to make the player not realise how small an area they're in, is to keep them busy - always pushing forwards. It's hard to do (what's to stop them from standing still?) but with engaging baddies, a fast paced game - or the classic rolling boulder a la Crash - and they'll thank you for a small environment.

Attachments

Login to view attachments
Bizar Guy
19
Years of Service
User Offline
Joined: 20th Apr 2005
Location: Bostonland
Posted: 12th Feb 2006 02:25
Wow, this may be your best toturial yet, keep it up!!

Classic Evil
18
Years of Service
User Offline
Joined: 4th Nov 2005
Location: watching you through your window
Posted: 12th Feb 2006 02:31
yes... I... Agree...

There are two types of people in this world: people who are extremly cool and are good at everything and people who are me.
Peter H
20
Years of Service
User Offline
Joined: 20th Feb 2004
Location: Witness Protection Program
Posted: 14th Feb 2006 00:55
yeah, thanks for sharing!

"We make the worst games in the universe..."

Login to post a reply

Server time is: 2024-04-25 09:55:25
Your offset time is: 2024-04-25 09:55:25