Question:

Your css maps 'NightFever' and 'Rush' are the best maps ever [...] I have been trying so hard to make a map like yours with a mate and so far have the most simple boring maps ever that are still not completed as I lost interest(Wasn't how I wanted it) and when looking at yours I was just soo amazed. I would like to know if you could possibly help me and my mate ever so slightly by writing a quick step to step list of how you do your maps.

Answer:

First you need a fresh idea. This can be anything. Wouldnt it be cool to play in a pacman labyrinth. Wouldnt it be cool if the map plays on a beach and anchoring ships? Wouldnt it be cool if the whole map plays vertically? The idea can be a single building in the map or a whole theme. The point is, this idea defines your map but evolves to something more exciting throughout the creation process you cant even imagine urself at this time.

Here at d00ds.com ideas are constantly flying around. Some get noted for later realization, some get scraped after a few minutes. You need to evaluate the ideas for two aspects: overall gameplay and amount of work. A football stadium may be a neat place, but I can barely imagine any interesting gameplay. It will be sniper dominated unless you change the theme so hard it my turn out like something you never really wanted. A Walmart shop on the other hand may contain interesting gameplay but keep in mind all the custom content you would have to create. All the stuff they sell has to be in there. You cant sell cat litter and ketchup only. These steps may only take a few minutes but properly executed it can save you weeks of unnecessary work.

The next step is to turn your idea into a theme. Start up flickr and google, start searching for reference photos. Imagine some cool spots like the tree house at T spawn. Find some details to spice it up. If you find it hard to come up with ideas at this point you should consider to change the theme. It will get worse as soon as you start looking at a bland wall and got no reference for support. It's better to gather too many references than to get stuck at 70% of the map cause your ideas run dry. At these moments your motivation begins to shiver and the death of your project is near.

You may realize we didn't start the editor yet. But careful planning in the beginning saves you headaches in the end. So the next step is to create an overall layout. It's very simple so you dont get distracted by details. We want to achieve longevity of gameplay at this point. As for the symbols I use rectangles for the "hot spots", circles for the "target spots" and triangles for the spawns. Scatter it on the paper and connect it with lines. The lines stand for a single route, no shape yet, no distance yet. Aim for some good interconnection but stick to a simple layout. Count the possibities given to the player. Count the routes leading to a hot spot, the number of entries to a target spot should be 2-4. Create different layouts, pick the best or combine two layouts to a new one.

Now we can pimp the layout. This is were you'll start your photoshop. Bring in the cool spots that stick in your mind. Start implementing height changes, stairs, ramps, cliffs. Turn the lines into corridors, hallways, halls. Dont make the mistake to turn every line into a corridor. This will look artificial. Extent the lines to the sides, so the running distance will be the same, but it doesn't look like a corridor. Decide which parts of the map are indoor, outdoor, underground. Check the timing so most of the fights occur at the hot spots. Check for sniper spots covering a huge area. Keep performance optimization in mind. This step will probably take some time. Thats fine. The success of your map depends on your layout. If you got it all sorted out take a break, tomorrow you will check your layout again and start hammer the first time.

So it's hammer time. Take some dev textures and put your map together with some simple brushes. This serves two purposes. First you got to check the scaling of your map. Does the player get lost? Will the windows be double the height of the player? Stop the time you need from one particular spot to another and check it against maps you like. On the other hand get some friends or bots to playtest the map. Does someone feel lost? Is someone forced to use the same tactics over and over again? Do some spots dominate the map? Are some spots out of action? And finally does the map "feel" right? Fix the problems now, it won't take long at this stage.

The next step will take most of your time. It's about time to create the map in hammer. I think it's best to start with the part you'll enjoy the most. And work forward from it to other parts. Don't build single places and connect it later. This will look artificial. Follow the flow of your map. Save up spots you don't like for the end. We want the map grow as fast as possible. Don't spend 20 minutes to find the perfect litter model at this stage. It will have almost no impact but you get the feeling of missing progress. Some spots need creativity, some spots need assembly-line work. Create these spots when you are in the corresponding mood. Always keep an eye open for layout changes that will turn out better than the original plans.

Do some playtests with your friends every now and then and start adding eye candy. Thats a slow process so do this whenever the opportunity arises. Write down all eye candy ideas. It won't hurt and maybe you can use them later somewhere else.

Before you release your first beta test be sure to tweak the performance. Maybe you will have to do it more than once, but it definitely pays off.

Listen to the feedback of your beta testers! Thats what the beta test is for. Evaluate every comment. You don't have to do it the suggested way, but be sure you address the problem somehow.

The final is final. So add everything you ever wanted into the map. Got an idea still lying around? Add it! Add the overview. Add the mission description. Add the handtuned nav-file. Add the readme.

Don't forget to promote your final release. Thats the last step on your voyage to create a map. Don't give up if the success is somewhat low. Think positive. You've learned a lot in the creation process. Focus your energies and start a new project. This time you will feel more comfortable. Maybe you will make different mistakes you did right in the first map. But think positive again, you discovered now how to do it right.

Stard00d
d00ds.com

Question:

quoting:
"Making a small bit of the map in high detail makes it impossible to finish the rest with enthusiasm. Making a quick over-view of the level and its layout makes it incredibly hard to make it more detailed afterwards."

I wouldn't have been able to sum it up this well myself. I love your "making offs", Dave, but what would really interest me is how _exactly_ you manage these two problems. At which point do you switch from rough to detailed? Have you ever finished a map where you started working on a real small detail and added the rest later? I so often come to the point where either of the two problems are so intimidating that I cannot see the map ever working out the way I wanted - which makes it painful to work on it. I'd love to get a more general workflow for this exact problem.

I know it never works quite the way you planned it but even a therotical plan (if I hear it from someone who used it with success) could give me tons of motiviation I often need and cannot find for finishing a map. Do you have anything to say about this?

Answer:

My advise so far while looking back at 3 finished maps and a fourth half done is as follows.

You need a combination of both. First you have to do the layout, but already have the theme in mind. Dont just draw lines on a scrap of paper and think how cool this plays. Imagine a rough theme for every part of the map. If you think a giant cliff would be cool in the map find a spot for it in the layout and now the cliff writes the layout.

Secondly dont think of your layout as the finalized version. Dont add too much detail to it. Layout can always change to provide more cover, shortcuts or an alternate route. It just gives you a rough idea where the routes are but not where the walls are.

Now when you are looking at your brushed layout take out the part you will most enjoy, the part you really cant wait to create. Design this part to full extent and use the motivation from looking at it and running through it to propel you forward designing the next part. You probably will show screenshots of it to close friends for your motivation.

In de_losttemple I designed the water part first. While finding a good set of textures for it the decision was made to switch from a desert theme to a jungle theme. So I made this part swampy, it worked well and I used this excitement to create T spawn using the new discovered theme. After all the displacement work I felt it was time for some brushwork. So I started working on B halls.

Now you can see these 3 parts connect to each other. Its harder to build in the middle of nowhere and connect all themes with small hallways afterwards. This will probably break the flow of your map.

You can also see how I followed the creation of parts of the map I enjoyed most at that specific time. Lets say after finishing T spawn I still enjoy displacement work. That way the underpass would probably consist of a natural cave, not a human built tunnel. Would this show an impact on the gameplay? Nothing at all! So dont feel too restricted with your layout.

de_cpl_fire was a very hard one on this topic, cause the layout was definately carved into stone. What did we do? We changed the layout on top of the play zone. A level designer needs some degree of freedom for his creative work. Otherwise he gets stuck looking at a wall thinking of ... a wall. So find spots in the map you can break out with your ideas without affecting gameplay too much.

Always think bigger. Even if the idea turns out impossible. It was an idea worth thinking of taking away 5 minutes of your time. Maybe you can use this idea later on in smaller scale. Plus you can come up with an idea everywhere, riding the bus, eating lunch or waiting for your GF ;) But picking the best idea out of 10 in 1 hour is better than starting up hammer, mapping the one idea you came up with, taking 2-4 hours of your time and noticing in the end it really doesnt work. Now the worst you can do is trying to fix this problem instead of deleting and starting a completely new idea.

Thinking is faster than creating. Use this to your advantage.

In de_nightfever we couldnt come up with a good idea for CT spawn. We had one idea, but that wasnt very interesting. So what did we do? We left it out of creation and totally ignored it while working in hammer. It was just a giant grey room without defined entrances. So the map around it evolved and one day the idea popped up by itself. In the end I think of CT spawn as one of the most atmospheric parts of the map. You just need to give the ideas its time.

So finally I would say making a small bit of the map in detail is the way to go. Just dont get lost in too many details. The custom dirt decal can wait. The crack in the wall can wait. Its already in your mind (maybe write it down) and nothing follows this detail. Dont do the hard and slow work in the beginning. Its all about motivation. You should already get enough of it from creating the nice architecture. If its not enough get some motivation from friends showing them screenshots. Now follow this track and impress them with different architecture somewhere else in the map.

At one point you look at already designed parts of the map without any feelings to it. Now its time to add this crack in the wall. Put in 2 more hours of work and come up with something new your eye can focus on. You can even show your friends screenshots again for your motivation. It will be new for them aswell.

In de_nightfever theres one pylon in the garage half ripped out from a crash with a car and skidmarks in front of it. This small detail cracked me up for 2 whole days. So I used the garage more than once to raise my motivation.

The lift was part of this tactic aswell. It was not planned in the initial layout. But one day I came up with this idea and sensed a very strong motivation to actually implement it in the map. So I did. And afterwards the lift provided me with 1 week of motivation while created in 1 day (lots of coding there ;).

Stard00d
d00ds.com

Question:

I love ur maps, I am a mapper myself and I have 1 problem no matter how much I optimize my map for better fps etc I have to cut down my detail to have a good frame rate map, but urs are very nice lookin and good fps, how do u do that m8?

Answer:

Actually our maps are optimized for 2 weeks. It's a long process (especially with all the full compiles). you have to know everything about optimization and use every trick thats possible.
this includes:

1. minor changes to the layout to block line of sight
2. hint brushes to support 1.
3. models with LOD (level of detail)
4. fade out distance for models
5. if everything else fails, occluders for models
6. no draw brushes (blocks LOS) behind displacements
7. skybox brushes (blocks LOS) in the air
8. take in account, that func_detail doesnt block LOS
9. func_detail for complicated brush work (stairs etc.)
10. create models instead of over complicated brushwork
11. avoid some special effects, try to use cheaper versions (e.g. point_spotlight is no-no)
12. avoid too many physics objects
13. nodraw on everything unseen (flat roofs)
14. use playerclips to prevent the player climbing too high and to support 13.
15. use areaportals like 4. but for brushwork too
16. use clean brushwork, no useless cuts
17. avoid large open-air areas with lots of models

some more detailed hints:

2. create a horizontal hint brush a few units above player head for every floor plus one hint brush a few units above jumping players head. sometimes its even worth to use a small diagonal hint brush to cut off 2-3 high detail models for one small specific viewing angle, if the fps are really bad at this point. you can see this in fire, bomb B, near the stairs looking to mid exit. the ivy on the right side is cut off as long as possible.
1. if hint brushes arent enough dont be afraid to change the layout. long viewing distances are always bad for fps. so if you can see another room far away just through a small gap close this gap with world brushes.
5. use occluders with care, they eat up performance aswell. so double check if they are worth their usage.
14. this is used alot on fire. try to climb up at fences and walls (low gravity just to test) und you will see, you get blocked. this prevents lame camper/sniper spots AND noone cries about their bad fps doing this.
15. this is used on fire in the red house at T spawn with the crane hook inside. you can see this house from the spiral stairs @ A. normally you can look inside trough the windows on top. but what would you see? nothing. so theres no reason to allow the player to see inside. all he can see is a fake texture, the room itself is blocked out. go into spectator mode and fly towards this house and you can see the switch. however the player cant see this switch, cause he is inside or cant see the house, when this switch happens.
8. sometimes you really need the LOS blocking from a door frame. in this case dont make it func_detail.
4. if you can see models around corners and this is the longest distance you can see these models in the whole map, fade out distances are the best solution. in fire from T spawn take the right upper route until you get to the "casey was here" decal. turn around. turn on wireframes. now slowly walk the way back and watch the pipes at T spawn fade in.

I hope this helps. always find your worst fps spots, turn on wireframes and find the things that the player cant see but are drawn. now come up with an idea to blend these things out with all the techniques you know.

Stard00d
d00ds.com