Slashdot Mirror
Trying Your Hand at Level Design?
Utawoutau asks: "As a student nearing graduation with high interest yet no game industry experience I have been taking a serious look at the position of Level Designer. In order to apply for such a position of course, I would need an impressive portfolio. I am aware that a number of games, Neverwinter Nights for example, come packaged with level development tools and that a number of other games have tools (official or not) that are readily available on the Internet. I am interested in hearing opinions from others that have experimented with the level design tools for a number of games as to what they found the easiest, the most fun, the most in depth, and the most impressive to work with. In particular, I am interested in a game whose tools strike a good balance between all four of the above criteria."
But the friends of mine who are do not find the industry all that they expected it to be. The fun and games that you would expect from a game company is actually politics and stress in reality.
You end up working long hours on a game that will be released when you know it's only half done, only to be laid off the week after the game ships.
Do yourself a favor and buy a Vanagon and go on a long road trip instead.
I have been pwned because my
That's the Half-Life editor. It's been tried and tested. Lots of tutorials to be found on the net. Easy to use and learn.
However, being an excellent architect is never easy:)
Just play a ton of mods (Half-Life and the forementioned NWN for example). You end up with a pretty good idea what works and what does not.
Whatever you decide to design start with pen and paper and a good idea first. Pointless mods that merely throw a bunch of monsters at you feel pointless.
Check out some classic mods for Half Life like They Hunger for HL.
I'd try bouncing a couple of levels off the fan community first. As far as games to develop go, I'd develop primarily for FPSs, and maybe a few strategy/adventure games. I don't play many of the latter, so here are the FPSs: Q3 and anything that uses it's engine (i.e. has Radiant editing tools) -- very easy to use, allows for lots of creativity. (Other Q3 engine games: JK2, RC Wolfenstein, Call of Duty?) The unreal games also have a nice bundled set of editing tools that would aid your portfolio. Best of luck!
as to what they found the easiest
:)
The Cube Engine allows you to edit maps right in the game on the fly. There is also a cooperative edit mode, try and beat that
Candle burns its brightest in the dark
Remeber if you work as a level designer in a company making a game. I can not expect to work with tools that are full fledged.
The tools you are going to work with will constantly evolve, your tools availble will be added and removed as the game moves on.
To create an impressive portfolio that will pop eyes where you apply. Design very good levels for a wide area of games. And last but not least, they have to. Absolutly HAVE to be well balansed. Especially with a multi-player game. I have seen some very pretty levels done for counter-strike but that was very poor balanced. So they where never played.
Another big thing when you design a level. Make sure you make your own textures. If you have solid texturework you will have a far better shot at the job.
-L
- To understand recursion, we must first understand recursion -
If you've never designed a level before, how did you hit on the idea of designing them for a living?
(I don't mean this as a flame; I'm just curious.)
First of all you have to have lots of experience in how the gameswork, and an artisticside. Good levels are very hard to do and they are done with lots of sweat from their makers, you have to weight in all the diffrent things in the game so that the level is a fun one, and beautiful.
This is not something you will get from getting a couple of tools that are easy to use, but from your own mind. That mind must also be able to adjust itself to the new tools, i.e. going from the worldcraft to new Radiant, and that will only be done if you have the time to do it.
The two biggest are Q3A and UT, check out the dev tools for them, they are equally good, just choose the one that suits you. If you want to dev for linux then you can get a very good start with GtkRadiant (the Q3A tool), but UnrealEd is very good as well alas only with windows support.
I recommend that you'd also take your time to learn a little about:
Programming
Scripting
Photoshopping (Sorry! I said GIMP'ing, of course)
3d modelling
Sound editing
I have done level design for a few engines and a number of different games. I do it as a hobby, so I am no professional. The most fun I have had designing maps, though, has been with the Torque Engine.
I say this not because it is the most advanced engine out there, but because I have actually done work on games and not mods. A few of my maps will actually be released in the wrapped version of the independent games I have worked on. That is a cool feeling.
By working with the torque engine on an indie game, I also get to work with the engine developers to add needed features.
As far as the tools used to create the maps, QuArk is used to create buildings (that same site has more information) and an in-game map editor.
--Nycto
1: steal decent levels from web sites
2: say there your own
3: get job
4: Stab manager in back, get his job
5: golf with the producers backers
6: back stab producer
7: get backer for YOU game
8: SHip it when it's half done, retire.
I figure about 8 month worth of work.
The Kruger Dunning explains most post on
UED (Unreal Ed) is only really meets two of his "criteria".
:( ). ;)), but eventually some of it starts to stick. Maybe one day you'll even be half decent, if you keep at it. Funny how a lot of things in life work that way, no? :)
In depth? Extremely.
Impressive? Well, maybe Intimidating is a more appropriate word, but sure.
Fun? Depends on how much of a sadist you are. It can be fun, but it can also be a lot of friggin work.
Easy? HAHAHAHAHAHAHAHAAAAAHAAAaaaaaaaahhhh... good one!
Ok seriously now though. UED is a pretty damned fine level editor if I may say so. Powerful as all hell, but its not exactly idiot proof. Its not impossible either. I managed to learn the ins and outs of UED (and to a lesser extent, the Unreal engine) just by reading tutorials, dissecting other peoples maps, and screwwing around... but it took a while. Months really. Even after three years of occasionally booting up UED, i'm still learning new things. Though to be fair, a lot of it is stuff that changed from UT99 to UT03 (haven't had the urge to map as much for the new game
But hey, if people really want to learn, there isn't much stopping them. Most people just dive right in. You'll probably be frustrated and attempt to quit (repeatedly
I'd like to add that it depends on what game you like to play what level design tools you should use. I like NWN, so I use the Aurora Engine, even if the Quake Mod tools are better or easier to use. Good level design means you have to understand what makes the game fun. I could never design a good Quake mod because I don't like Quake.
So, for level design my advice is: pick a couple of games you like and see how the tools for those games are. If they are not too daunting, jump in. It'll take a couple of weeks to get familiar with ANY tool, but there are usually good forums that'll help you along.
You may be barking up the wrong tree. Level design, particularly for modern 3D engines, is more of an art than a science. There are many mappers who have been working at it for years, for various games. In particular, you might look at some of the maps released for Quake III and even the original Quake, along with more recent games like Unreal Tournament 2003. The problem with assuming that you can just "jump in" and become immediately skilled is that it completely ignores the skillset which is unique to that position. Quite honestly, almost no programming experience will transfer to the field. If you are still interested--and it is truly a hobby/profession which requires interest--Half-Life is actually a good place to start. Try not to get overwhelmed with concern about newer features at first--learn about brush layout, visibility occlusion, et cetera. These will help you with any engine. Also, I think that these skills are more transferable than something like Neverwinter Nights or other, simpler level editors. Making a level in a 3D game like the FPSes I listed above is very similar to 3D modelling, so you can transfer skills between the two. Func_Msgboard is somewhat dead these days, but you might find things of interest there--and there are some truly great mappers who hang out there. :) Good luck.
First, read anything and everything by Chris Crawford. I hate most new media theorists, but he's an exception. He's a physicist by education and programmer/game designer by experience. I'd especially take a look at "Chris Crawford on Game Design".
You should also take a look at some of his old game design articles in Next Gen magazine. He had one article on level design in Doom that was quite unique.
Secondly, from a tricks/tools perspective, gamedev.net or xgames3d.com are your best bet.
What do you mean my sig is repetitive? What do you mean my sig is repetitive? What do you mean....
having recently tried a number of Q3A mods that really did nothing to balance the game.
Most of the really good things that define a highly playable mod from a 'cool, lets try something else' one are in the basic mechanics of the game.
Where are the weapons and are they appropriate for the playspace? Do they work well for a few players? Lots of players?
What about the playspace itself? Does one side have a clear advantage? How about a particular path? Many of the classic levels avoid dead end ways making it tough for players to just sit and wait for others to show.
For games like capture the flag, where is the balance of power? Is it possible for a team to cover all their bases without being forced to venture off for new weapons/ammo?
Visuals have never been as important as play mechanics are, but they do play an important part. A dark spot or interesting texture placed just so, might allow a player to hide for a bit, or clash horribly with the target, making for either a sneaky element of challenge and tension, or a frustrating experience...
Spawn points should be in areas where players have a fighting chance at actually spawning without being camped too often. Best ones are those where the player forms near the action while campers always are watching their back.
All of this is specific to Q3A --hey, I still play the game because it is well balanced and interactive, but other games have similar issues. The parent is right, you gotta play.
If it were me, fun first, beauty later, but that's just me.
Having played Q3A a lot, I have been thinking about this too. Be sure to check out other mods and play them. You will get a great sense of what you want to do. --Then do it!
Blogging because I can...
1: Get a job at some faceless corporation with a name like 'Enron'.
2: Design kick-ass games after hours.
3: Get your work stolen by your sneaky co-worker.
4: Beat the Master Control Program for domination of your codez.
5: Become the newest exec (you even get your own helicopter!)
"Greetings Programs!"
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
The most important things are a quality filter so we can leave you alone and have you produce good work, ability to come up with a working layout (this is fundamental and oft underestimated), a decent technical understanding and in our case, a love for multiplayer gaming.
Having experience shipping finished maps implies very good things about your ability to manage your own time, and your ability to finish what you start. That's a pretty rare trait. Extra bonus points if youur work gets played. (Probably means it's part of a pack or a mod.)
It also falls into the "nice work if you can get it" category, as mapping careers are relatively tough to find. I don't know of any other companies in Canada who are asking for the same technology experience that Threewave is. Digital Extremes comes close.
A thing that struck me when reading through the other posts is that a lot of people seem to say "use the newest and most advanced tool available". That is definitely something you will have to do in order to get a good portfolio.
But for learning the basics about the different aspects of level design you might want to try your hand at something simpler. Try finding "old" games like Doom or Warcraft2 (I think there was an editor for that?) and try your skills at designing levels that are only two dimensional. Believe me, there's a lot that can go wrong even without the extra options available in the 3D editors.
I know it may sound like a waste of time, but it will give you a few good experiences, I think. Due to the simplicity of e.g. Doom's level structure you can spend a lot of time working on the small details like how to align two platforms so a player just barely can (or, if you're wicked, can't) move from one to the other. You can experiment with how to let the player use different objects/walls/etc as cover when shooting large monsters and so on.
One of the other posts mentioned that you would undoubtedly be forced to learn to adjust to using new/different tools that what you're used to so another benefit of "working your way up" from the old games is that you will eventually have to move on to new tools. Make sure you focus on learning the fundamental parts of what level editing is rather than memorizing the exact functionalities of a single level editor.
Oh, and after having made levels for 2D games you'll definitely appreciate 3D editors and the freedom they give you.
A last piece of advice is to try and come up with something original instead of "yet another multiplayer map". Some years ago me and a friend designed a series of "Quake Hinderbahn" levels. It's basically an idea we got for a LAN party where we wanted to host a different kind of competition - and we made an obstacle course and had people record demos of their fastest runs.
I learned a LOT about how Quake works from that.
Good luck with it.
It's 19:11:42. Do You Know Where Your Meat Body Is?
and "The Art of Computer Game Design" here
and the related Game Design Wiki
Good luck! -Joel
Where to start...
Well, I guess the easiest, most obvious level editing suite out there (in the realm of FPS games anyway) is Valve's Hammer. It's quite scary just how easy it is to make levels with that util. Upon starting it I was able to figure out pretty much all of the basic features just by looking at the buttons. That's quite an accomplishment, if you ask me. The whole way the editor is layed out, and the process you use to design a level in the editor are both painless and relatively easy. If you're looking for a place to start, and games like TFC, CS, and NS are your forte, might as well start here. I think the only problems you may run into are in the setup options, and true to its oldschool roots, getting down and compiling a map, then tweaking that, can get pretty gritty.
The only other editor I've logged any sort of time on was the unrealed that shipped with UT2003. I missed the whole UT generation, so I can't vouch for the older unrealed, I'd assume it's basically the same. However, after coming from Hammer, learning unrealed is a real pain. The interface is kind of counter-intuitive and the whole logic behind building levels is completely the opposite. It's quite weird. Once you get into the unreal mindset though, creating a level isn't hard at all, it's just that initial hill that you need to climb over. Well that, and unrealed is insanely buggy. I cannot tell you how many times I've lost work due to unexpected quits or fatal errors that seem to make no sense. I guess as it is with every program, save early and save often. This is the editor I eventually found most technically impressive. After learning something new about the editor, or pouring after technical docs and taking a stab at it myself, I am still wowed when I get a new effect (be it graphical or gameplay) working.
Now note I have not logged any significant hours with these editors (read: I have not produced any well known or well thought of maps) as to know the individual quirks and the nuances of each editor. However I can offer the complete newbie's look into it, which I guess is better than nothing. If anything, I say that the Half-Life community has more tutorials and help geared toward my audience than the unreal community, as finding good and easily digestible information is hard. Epic seems to be trying to remedy this with their opening of Unreal Uni, or whatever it is they are doing, which offers video tutorials and forums for developers. (A good thing, but I'm fearing the mod potential is going to waste in these days counting closer to HL2, with potential developers looking for the next big thing rather than weighing the assets of what's out there- which is hard to do for something not quite released)
As for most things tech, I guess slogging through it and exploring is probably the best way to learn either of these.
I guess on the RTS front, nothing is easier than Blizzard's warcraft 3 map editor. The thing is incredibly simple to use, though not quite as intuitive to the newbie's eye as hammer. I was able to get up and building levels (after finding I was unable to figure the tools out myself I went to the documentation) in about 10 minutes after reading the rather friendly documentation that comes with the toolset. I guess the thing is that the tools take 10 seconds to learn, but to master them and build a balanced and fun map, will probably take you a life time. (not to mention a keen understanding of the game.) The tools are fun to use, and going from the editor to a game to a multiplayer match you're testing with friends is quite easy and gratifying.
Neverwinter Nights, I found, was rather easy. Scripting seemed to get kind of nightmarish quick, but I quickly lost interest with building with those grossly simplified tools. I guess the real challenge was figuring out a way to make and import your own tilesets, but in the beginning when I had just spent 50 and tax on the game, it was a disappointment for it to have such poor mod support right off the bat.
I have heard good th
Hey, it's my OPINION that dogs have eight legs and make a sound like a car horn every time they take a piss.
THe most experience I'd have with game editors is the Aurora Toolset for NWN. I think it's one of the more flexable yet easy to use game editors out there.
Back around Feb 2003 I started an online module called Mandrake that went well over 90+ areas when I decided to move on. I had a good base of players too, not near as much as some places but a good base. Since then I've worked on about 4 other module designs of my own including an arena one for the PVP section.
I'd say the scripting language and the flexability to created hundreds of possible paths for a module are it's greatest power over all the other editors out there. It's also relativly easy to use, the only thing is you don't get a nice print bound manual for the scripting though the script editor in there has a nice help reference side bar. It's also really nice to be able to script in counter measures for cheaters.
I'm not sure I'd ever want to do this professionally though. Seems it would make a better hobby then a job but thats my opinion.
~~ Behold the flying cow with a rail gun! ~~
People are answering your question as if the goal is to design the best looking levels, or the easiest way to build them, or other construction related questions. But those are not the right questions you should be asking.
Just as an architect needs to find out what the goal of their building is, a level designer needs to have a goal for a level. An amazing house that doesn't have the rooms a buyer wants is still a failure. An amazing level that doesn't advance a game, or isn't any fun, is also a failure.
Before you start building a level, write it out. Put together a story of how the player will move through the level. Sketch storyboards of interesting challenges that will occur. Start with a rough layout of the rooms you plan to link, and describe whats going to be happening to the player as they move through those areas. Figure out which areas are dull, and either liven them up or cut them out. And once the flow of the level makes sense and seems enjoyable, a level designer passes the documents off to an artist, whose job is to make it look good.
Far too often, the process gets reversed. The levels are designed from an artistic perspective first, without first determining why the levels are there, or where the player will be. Time gets spent fleshing out regions that the player runs through once, hunting for something to do, and never looks at. Those areas may get thrown away in playtesting, or just kept in and bore the players. Not good.
So basically, if you want to be a level designer, design levels. Don't be a CAD designer; that is the artist's job. As a former game developer, I would have appreciated it. As a current game player, I would appreciate it perhaps even more.
I have a number of friends working at well-known game companies, and from what they say, I think it's not my cup of tea (it still could be yours, of course). Their biggest complaints: working 80+ hours a week for what seems forever before the game comes out; dealing with execs; getting stuck working on a lame part of the game for a little while (esp. early levels/concetps); hearing users complain that a certain feature of the game sucks when they worked hard on it and weren't given enough time to do so; and end up wondering if it's really worth it. The biggest joy for them all, though, is when the final version goes out for shipment. There are always parties, and the next few weeks at work sound pretty spectacular (i.e. just playing video games). Like any job, you gotta take the good with the bad. Just know the bad, and you'll probably be fine, but know it.
Speaking as a player, I'd encourage you to seek out any competitive communities built around the game(s) you're interested in designing for. While Neverwinter Nights is a cool enough game, I'd also reccomend you check out getting some experience with the "Quake-related" FPS games.
One of the tools that's used to create maps for several game is Radiant (supported on both Linux and Windows) and supports a bunch of games. Mappers familiar with Quake3 were able to transfer that knowledge to RtCW when it came out, and in turn that was transferrable to ET. Plugins for other games (Half Life, Soldier of Fortune, and some others I'm blanking on) is also available.
Couple that with experience working with 3DSmax (or gmax if you're on a budget) creating models, and you should be good to go.
If you're also looking for ideas on what to make maps of, I'd suggest trolling around and seeing if you can find a competitive community for the games you're interested in designing for. Stuff like Half-Life and its mods (most notably Counter Strike), Quake3/RtCW/ET, and the MOHAA/CoD stuff are going to be hot ticket items right now. Organizations like The Team Warfare League or the Cyber Athlete League might be worth a look to get an idea of how people are using the games and what kind of maps and what features they'd benefit from.
Looking to the future, everyone's pointing at stuff like Doom3 and Half-Life2 (obviously). But it might be worth taking a look at games like Far Cry and Painkiller as possible sleeper hits coming up on the horizon.
Good luck in your efforts. :)
the main danger of that kind of job, is you'll end up being bored of it, a few months after you start. It's very repetitive, and probably not as creative as you believe it is.
____
nico
Nico-Live
In case no one's mentioned them, I really enjoyed Mark Saltzman's Game Design: Secrets of the Sages. May be a bit dated (1999), but good reading on the genre. I understand he had a new one out in 2003 called Masters of the Game - don't know what it's like though.
Remember that level design, texture design, and even texture application are often three different jobs done by three different people (or, more often, teams). Level design is architecture. Texture design is a very specialized form of 2-D drawing. And texture application is basically interior decorating.
Assuming that you want to lay out levels, then study architecture. You want to be able to make things that look real, atmospheric, and cool. A good exercise is to find a dynamic and interesting space in real life and model it as precisely as you can in a game engine. This will give you a feel for the level of detail needed to build a truly immersive level.
Also, don't limit yourself to one tool or even one game. Some of the more full-featured and late-generation editors have a huge amount of crutches that you can easily get dependent on. Make an RPG dungeon and an RPG city. Make several RTS maps (these are great for gaining a macro-level understanding of balance and flow) for single-player and multi-player games. And, of course, make a lot of FPS maps. Even a dippy little fragfest map can show off your talents if you put a lot of work into it.
Finally, be aware of industry trends. For instance, single-player FPS campaigns are starting to veer away from the strictly-linear style of the genre's forerunners (look at MOHAA for an example of why... the levels are very stale and scripted). Play, play, play. And take lots of notes.
1st, realize that making games is not the rockstar job that lots of people make it out to be. The game industry keeps itself alive on the blood and sweat of people who love making games. You will work 10-14 hour days for weeks at a time (or years, as some cases may be). You probably won't get a bonus when the game is done. You probably won't be paid very well. You'll have to work with people who think they are god's gift to the world, and they will probably be your boss.
But, you will also work with really cool, creative people. You will get to do something you (hopefully) love to do. You will get to create games. If this small paragraph sounds better than the larger, first paragraph, then by all means, pursue a career in the game industry.
So... the nutshell I can come up with at 4:45 in the morning (yes, I've been working since yesterday morning).
Get a day job first. It's a different path for everyone, but odds are you won't break into the industry anytime soon (it took me a couple years). You might be able to get in as a tester or intern, but it's almost as hard to make the jump to the dev side as it is to just break in.
Grab the whatever latest version of Unreal2KX XMP Super Mega Championship Edition is out. Epic has done a good job of marketing their engine and tools to devs, and a lot of places have picked up the Unreal engine and it's editor, UnrealEd. This can give you a slight advantage just because being familiar with the tools can be a big selling point to some companies.
Next, learn how to use it. Not just part of it, all of it. How to make and import textures. How to make and import meshes. How to make and import sounds. Even learn the basics of unreal script. You may not be actually creating art assets/code in the position, but as level designer you are where 'the tires hit the road'. Everything has to come through you at some point to go into the level so you have to understand everything that is going on behind the scenes.
Make a few multiplayer maps of whatever flavor you want. Focus on a few key areas:
1. Look and feel. No BSP holes. No meshes intersecting each other at weird angles. Everything lines up. There is a good sense of 'space'. Lighting is good and reflects the mood appropriately but isn't overboard. Textures are aligned properly.
2. Wiring. Doors open and close when they are supposed to. Switches work the 'right' way. Events happen when they are supposed to. Areas are zoned or antiportaled correctly.
3. Gameplay. For learning, I put this last at this point. These maps your learning how to use the editor and trying to make them look as good as possible. In general, in gameplay the player shouldn't get lost or stuck anywhere. The next area to explore should be obvious. Paths are clearly marked. Framerate is good at all places in the level.
Now, you need to make some single player experiences. You probably won't get a job making multiplayer maps (I've never made one professionally) so you need to be able to create good single player experiences. This is the hard part (learning the tech just enables you, this is the actual work!). Even moreso than understanding the technology, you have to understand the game you are making and understand the game design.
To learn how to create good single player experiences, don't just play other games, analyze them. Watch how they create tempo and how the flow of the level works. When is the player challenged? How often? When is the player rewarded? How often? What types of challenges are present? How difficult is the game? Why were certain game and level design decisions made for that game? How would the level design be different if the character could jump twice as far? Shoot twice as fast? Once you start playing games with these sorts of questions in mind, you'll start to have a better understanding of what it takes to create a level. It will take awhile, as long or longer than it takes to learn all the tools. I'm still learning and I've been making levels since Q2 days.
Good luck.
If you want to be a game developer, whatever else you do...DON'T GET MARRIED.
assert(birth_date<time-86400)
If you are serious about learning how to use UnrealED, take a look at what 3dbuzz.com has to offer: free video training modules (VTMs) on how to use the thing. Some of the modules are pretty hefty -- weighing in at 100+MB on some lessons. These can be found here.
On a somewhat related note, if you pre-order the special edition of UT2K4 you'll get the VTMs on a DVD.
The most brilliant developer I know works in game development, and he is frequently between jobs and looking for work. This is a guy who's got the answer to any question I've ever seen asked of him off the top of his head. I wish I could give you examples of his over-the-top brilliance, but I'll have to just say that his screen name has become a synonym for "wisdom" and leave it at that. If this guy can't hold a steady job in game development, I don't know how anyone can expect to.
--- 11 meters/second, or 24 miles per hour - the airspeed velocity of an unladen European swallow. Really.