Custom Map Makers Wiki - User contributions [en-gb]

Getting Your Map Played

2012-08-08T09:57:03Z

Rylius: Created page with "'''Originally written by BludShoT on the urbanterror.info forums.''' * [http://www.urbanterror.info/forums/topic/28011-is-map-making-craft-dying/page__view__findpost__p__331780 P..."

'''Originally written by BludShoT on the urbanterror.info forums.'''
* [http://www.urbanterror.info/forums/topic/28011-is-map-making-craft-dying/page__view__findpost__p__331780 Part 1]
* [http://www.urbanterror.info/forums/topic/28011-is-map-making-craft-dying/page__view__findpost__p__332302 Part 2]
The text has been shortened at parts while retaining the original intentions.

== Mapping for Urban Terror ==
The barriers to entry in modding and mapping for a 13 year old game are huge compared to new games. The tools and support and communities for games like L4D2 and Portal2 are just SO MUCH EASIER to get into. Plus the user base is much larger, so that attracts creative people who want to make things that a lot of people play.

It is true that there is a disconnect between league communities and mapping communities. But whose fault is that really? Now and then, I have tried on and off for a decade to bridge that gap, with occasional success. The truth is, that almost all the mappers who made competition-worthiness a high priority in their mapping ended up with maps that people played at least to some extent.

I submit that mappers can't expect gamers to come to them to help out of nowhere. Why not go to the gamers? Going straight to the competition gamers via league forums is not for beginners though. You need to be further along to do that and have it be useful. Otherwise people will just trash your map and frustrate you.

Instead, '''you need to start smaller'''. Meet some people, find a nice clan (not a nub clan though) who will help you with initial map testing. Get your map to a point where people are telling you it is becoming a good competition map before approaching a wider audience for feedback.

The other big thing to think about is what are your goals for your map? Do you want to just make something pretty for yourself, make something for pubs, or make something core like Casa/Turnpike/Abbey. It's very often that mappers are making something for themselves, and stubbornly refuse to alter it or listen to feedback. If that's what you really want to do and that makes you happy, then you should do that. But don't make something that doesn't really fit this game and then get upset when people don't play it.

'''Designing a successful Urban Terror map (for competition) means following preset rules of what works and what doesn't work.''' It's just like trying to build a car or a rocket or something. You can't just decide to make your rocket have giant wings or something because you think it looks cooler. It won't fly.

The reasons for the constraints on what makes a good Urban Terror map are *not* just because gamers don't want to give your map a chance. There are real quantifiable constraints. Again, if you don't care about this, and just want to make your giant desert map with open spaces and humvees, or your detailed city or weird castle box map, then go ahead and do that, but don't get mad if people don't play your map.

Study the popular maps in UrT. Think about what they have in common. Scale, connectivity, defined areas and not too few or too many connections. Not too much verticality. Try to understand the constraints of the Q3 engine. The reason that Casa's areas are split up and not just buildings plunked down in a huge open area like BF2's Karkand, is not arbitrary! Your map needs to be split up or else FPS will suffer. As a bonus, when you build your map in a way that is good for FPS performance, it will also play better, because splitting up the areas results in better game play.

In my opinion, mappers need to focus more on game play and layout earlier in their mapping process. Then you could get people to play your map and tell you the gameplay is no good *before* you spend a million hours making a bunch of detail.

== The playing styles of Urban Terror ==
The thing is that Urban Terror has a very specific game play style. New players don't generally even understand what that style is. They see the game and see the guns etc and figure it's roughly the same as Call of Duty, Battlefield, or whatever. But it really isn't. "What is fun" is very game specific. You could have a really fun map in Crysis but the map sucks if you used it in Urban Terror. You could have a fun Urban Terror map but it's boring if you played it in BF3.

== Great and great maps ==
Firstly, mostly what I was talking about in this thread was "popular" maps, not great maps. (Because we were talking about how to get your map played by a bigger audience, so I said to emulate the most popular maps because they obviously contain the ingredients for being popular.) Also, "great" is very subjective. There are maps that are visually superior, but fall flat in game play. Those maps are still great in a way. So its very hard to define greatness, it's more important to define what the goals of a map are and then see if it meets them. In some ways, ut4_superman is a "great" map. It does something pretty cool that some people really enjoy. But it's completely useless garbage for other uses like serious standard urt competition. (Don't get me wrong, I find game play experiment maps like that to be very cool.) I don't consider that an insult because it was never trying to be a competition map. Turnpike makes a terrible jumps map for example, because it's not a jumps map.

== Gameplay at the core ==
Urban Terror is obviously not a class based game. A core idea of UrT is to be anti-class, anti-leveling (cs buy system etc), so all players can start each round with whatever guns they want. There are roles to be played such as snipers, autos, medics, but it's a small subtle part of the game. I'm forced to only guess what you mean by "tactic", but I feel like I bet I know what you mean. I bet you are talking about some sort of tactical gameplay that you might see in a slower game such as True Combat Elite. Urban Terror simply isn't that kind of game. A group of friends can sit there and purposely play it that way, but, that's not how it gets played at higher levels.

This is an issue of game theory (or something fancy sounding like that lol I'm not sure). Think of it this way, games get boiled down to their essentials the higher up you go on the skill chain. Players learn to squeeze every last bit of advantage and methods out of the game. This happens in all games.

So for example, let's say you are playing NHL 97 on Sega Genesis. At first, it seems like this fun realistic hockey game. You have fun and play with the full array of things you can do on there, simulating playing a real hockey game. But then users play the game more and more, and realize that if you take a slapshot from the right spot, you ALWAYS score. Now when they play, they ignore all those simulation and fun things to do, and just cheaply keep going to the easy spot to slapshot score over and over again.

Luckily Urban Terror is not so one-dimensional like that, but I feel like the example still applies (especially since I couldn't think of a better one right now). So a new user to UrT might see the imi negev and think "Cool, I'm going to lay down suppressive fire!". Sounds like a cool idea right? And it will work if you're playing vs other new players. Then you might think, ok I'm going to make a map that makes use of that feature, like put some sandbags by a door and we can have a suppressive fire negev there. But, higher skilled players know that the game doesn't work that way. They can sprint jump off the walls and slide by the negev and shoot him in the head with ease. That's the difference between how the new player kinda "wished" the game was, and how the game actually is.

The class concept was specifically never wanted by the devs. *Variety*, yes. But even then, the weapon balance that the devs chose to go with over the years (which was also somewhat of a function of 'popular opinion' of the player base I guess)(and most of those decisions were made a long long time ago) doesn't truly lean towards more variety imo. UrT has no dominating close range shotgun, which would add another element to this game. The shotgun it does have, the PSG-1 and the HK69 are virtually written out of the game because of their weapon balance. The AK and the G36 do provide a sometimes needed niche, and it would suck if they weren't available. But generally urt is all about LR's and SR8s. And the long rangeness of the SR8 role is barely long range if you considered it in the context of other games out there.

== Why Casa is a 'great' map ==
I'm personally pretty bored with Casa TS because for some reason people camp on it more (even though I don't think they need to). Casa CTF I still enjoy. I believe the reasons why it's considered great are because:
* it has good FPS performance thanks to well thought out VIS occlusion (separation of areas)
* it mixes just enough long range fighting with lots of short range fighting
* it follows my 'rule of 3', which is any place you stand on the map, there are generally about 3 angles on you. Not always 3, but, you want a few 2's, lots of 3's and the occasional 4 or 5. So, in Market, there's 3 ways in. In Patio, there's 4 ways in. In Fountain there's roughly 5. Most other spots on the map are 3s. It's about how much the player has to deal with. If it's too few, like a bunch of 2's and 1s, then it's boring. If it's too many the gameplay feels random and out of control.
* Urt uses almost all hitscan weapons (bullet guns) and it's not a simulation game. Therefore, too much verticality in a map pretty much sucks for urban terror. The experiences when having vertical firefights aren't fun. So Casa's '2 level's of height fits well.
* the map is smooth. You aren't fighting against the map and bumping into things, so you're free to focus on fighting your opponents.
* it offers good flank routes
* For CTF the paths offer the opportunity for really fun scenarios. The fact that you can cut into the middle from the sides gives you that.

== Details and wide areas ==
More skilled gamers will conceptually reduce your map down to the basics. They will figure out how to circle jump across your map as if half of it wasn't there. If while trying to do these skilled things they find they are always bumping into cars and details, it will frustrate them and make them not want to play the map.

Wide open vistas will give an edge to players who have amazing aim. A player who is more about surprise flanking and movement will be dominated by the good aimer. This makes your map less interesting and gives less 'class' type roles for people to play, because they will boil it down to "oh ok all I have to do is camp on this balcony and shoot the whole wide open area and dominate it". Also, I can't say this for certain because I didn't play it yet, but I bet your latest released map would be ruined in pubs by heavy nade spam. Since it's a box map, everyone on each team can just constantly saturate the middle of the map with nades.

Also, your "open forms" heavily reduce your map's chance of having decent performance. Urban terror is free and old, so compared to BF3 or Far Cry 2, the system specs of the audience are on average much worse. If your goal for a map was to have as many people as possible play it, then ignoring slower pc's goes against your goal. If you don't care about that, then obviously it's not a problem. I don't know how much you know about the vis process in quake3 (urban terror), but, if you have two rooms completely separated from each other, then the game will only render the player and gun models (including dead players, guns on the ground etc) in the room your are in. But then, let's say you cut a doorway to connect the 2 rooms. Now all of the models in both rooms will be rendered even if you can't see them, even if they are behind a wall. So you see in Casa, it has Z type doors where it's not possible to see from one area into the next. That way the engine won't render those things in the next room.

Modern games like Far Cry or BF3 work completely different, by using Level Of Detail across distance. So if you are far away from something, the engine will display a much lower detail version of it - because you are far away you can't tell the difference. And by doing these LoD's, the scene can be rendered faster and the FPS is still ok. But quake3 doesn't do that. It will render everything to the horizon in the same amount of detail as if you were up close. So if you try to re-create a scene from a new game like that in Urban Terror, it will be slow as molasses.

== Limiting paths and set rules ==
Designing limited paths to go by in Multiplayer is also good for some games, like Quake 3 or Urban Terror. For games like BF3 or Far Cry, open is better, because the game play is extremely different.

Think of it this way. Chess is a skilled and fun game, where players match wits and try to attack each other from various angles and plans. The chess men have strict narrow rules of what they can do. The pawn can only move forward, the bishop can only move diagonal on the same color, the knight can only move a certain way, etc. *Because* of these tight paths and limited options, chess is fun and interesting. Casa (or Turnpike, good urt maps), is like chess in that way.

Now, what if you changed the rules of chess, so that all the men can play in an open field and go wherever they want? It would make the game stupid and pointless. The skilled chess player would no longer have any advantage over his opponent. The new player would never have to learn how to actually play chess. The interesting limited set of plays and counter plays would be replaced by a big mess, and peices would be grossly overpowered. If you played chess like that with a little kid I bet they would have a TON of fun though, hehe.

That is what an open map in Urban Terror is like.

== Conclusion ==
But I say all of this in the interest of sharing ideas. I do not want you for 1 second to feel insulted or to feel discouraged about Urban Terror mapping. Even though I am saying that a map needs to follow certain guidelines to become hugely popular across all urban terror players and play types, that does not mean that every map has to be like this. Different players want different things out of the game. For example, an big open city map might be totally awesome on a Zombie mod pub server.

I don't think every mapper should be trying to make the next Turnpike or Casa. I only say these things on the topic of a mapper who *wants* to make the next turnpike, and is looking for advice on how.

A very important thing when making a map is to decide what your goal and target is. But in terms of the "Craft" of map making, those who are serious about the craft will definitely shake their head or shed a tear when they see someone make a box map.

On some level, mappers must be making their maps to be played. Because if they were only making them to make something good looking, regardless of if it got played, then they might as well be working in Autocad or 3DS Max, or Google sketchup or something.

Play Quake Live and Warsow, and then play True Combat Elite (or is it TC:CQB now?) and CounterStrike (or one of the many F2P shooters these days). And realize that Urban Terror is actually something like halfway between those 2 kinds of games, even though visually it appears like it's going to just be a counterstrike type game.

2011-12-04T09:34:23Z

Rylius:

If you don't have the Gimp normal map plugin, download it here: [http://code.google.com/p/gimp-normalmap code.google.com/p/gimp-normalmap]

We'll start with this seamless texture: [http://www.cgtextures.com/texview.php?id=34805 Seamless texture]

The lowest resolution will do fine.

If you don't have a cgtextures account, create one now.

First things first, scale it down to 512x512 pixels so it is a power of two and doesn't mess around with some texture settings.
Save it, that will be the base texture we need.

Next we will create the normalmap.

You'll find a simple filter under Filters - Map - Normal map.

First thing you want to change is the filter, 4 sample just doesn't do bumps. 3x3 doesn't do much either. (The higher the numbers, the less bumps you get, so don't even bother with 5x5 etc)

Sobel gives weird results, I never use it.

Prewitt 3x3 is the way to go. Check the "wrap" box on the right, this will give you a seamless normal map.

You can ignore most of the other options, just hit okay and wait for the map to be rendered.

You'll end up with some mostly blue image. If you want you can go test it ingame right now, but I can tell that it won't give you much bumps

To get some more bumps, press Ctrl Z to revert the image, then hit Ctrl Shift F to reopen the normal map generator.

Set the scale to -5. Scale defines the depth of the bumps. It must be negative because otherwise the mortar would stick out of the bricks which looks weird.

Generally the brighter parts will be nearer to the screen than the darker ones. You can also invert the colors instead of using a negative scale, but it really doesn't matter.

Hit okay and wait for the map to be rendered again.

You can see there's a lot more red and green in the normal map. Way too much actually. The bumps for the mortar are decent though. If you look at this ingame, you will see tons of distortions on the brick and it will look like shit.

One could paint over the bricks with the default normal map blue, but that's a serious amount of work and since we are lazy we don't want to do that.

So hit Ctrl Z again and go to Filters - Blur - Selective Gaussian Blur. This tool is goddamn awesome.

It blurs parts of the image based on contrast.

Max. delta defines the contrast threshold needed to blur something, the higher the more will get blurred.

50 is fine for this texture, though you may want to lower it if the texture has low contrast.

A blur radius of 30 is usually fine too. If you got a slow CPU and don't want to wait too long, try 20 or 15, but it shouldn't be less.

What you can see now is that only the bricks are blurred.

Now reapply the normal map filter.

You'll notice there are a lot less distortions, though there are still a few ones we don't like.

Hit Ctrl Z once again and then switch to the contrast tool.

Increase the contrast by 80 and decrease the brightness by 80, too.

Then fire up the selective gaussian blur again, set Max. Delta to 65 and reapply it.

The image is now very washed out and looks really bad. We don't care!

Reapply the normal map filter and you'll get very little distortions.

If you want even less distortions, try the 3x3 filter instead of Prewitt 3x3. You may want to increase the scale then though. (Or decrease if it's negative )

White parts on some of the bricks will create huge distortions, just paint over them with the blue color.

If you take a look at this ingame now you'll notice some serious bumps.

If they aren't enough you can still increase the scale, but keep in mind this will create more distortions and will require you to put more work in the normal map.

Hope this helps some people, just discovered the Selective Gaussian Blur tool not an hour ago myself

[[File:Cg bricks n.png|300px]]

This has distortions though. I was lazy

[[Category:Textures]]

How to make a brick texture

2011-12-04T09:34:08Z

Rylius:

Brick textures are one of most used textured in general. But very rarely i find one that matches my needs perfectly.

Either the color is wrong, or i need to make it seamless, or the brick texture isn't exactly what i want.

In this tutorial i will show you how to create a custom brick texture out of any texture you find.

==Brick Pattern==

In order to create your brick texture you need a basic brick pattern template. This pattern will make sure our texture is seamless as it should be.

[[File:Bricktemplate.png|300px]]

Now we have to find textures that will represent grouts and bricks.

In this tutorial i will use concrete1.jpeg as my grouts texture and webtreats_stone_1.jpeg as my brick texture.

[[File:Textexamp.png|300px]]

ow that we have our 2 textures we can move to blending the 3 together.

Next steps are quite simple.

First align the 3 images in this order:

*Top: Brick texture
*Middle: Brick Pattern
*Bottom: Grout texture

[[File:Step3fe.png|300px]]

Now that everything is aligned all we have to do is set the brick texture mode to multiply.

And hit ALT|+left click in the space between the brick texture on top and the brick pattern in the middle.

This will put the brick texture in a container shaped as the layer bellow it.

The reason we set the blend mode to multiply is to bring out the brick pattern bellow the brick texture.

[[File:Step4w.png|300px]]

All that is left now is to test the texture in the game.

[[File:Shot0078.jpg|300px]]

[[Category:Textures]]

Tutorials

2011-12-04T09:32:49Z

Rylius: