1. Design the levels on graph paper
2. Block out the levels in 3d
3. Create 9 –12 pencil drawings of locations in each level
4. Have your Environment Artists build the final geometry over the blocked out geometry
5. Color the 12 level drawing in order to plan the colors, lighting, special effects and textures
6. Light the levels after textures are done
7. Place final cameras

GD = Game Designer
AD = Art Director
LD = Level Designer
CA = Concept Artist
EA = Environmental Artists
VFXA = Visual Effects Artist
CM = Cameraman

1 Design the levels on graph paper

1. LDs- create level flow chart
2. GD, LD and AD- discuss level content in meeting
3. CA and AD create preliminary color and design sketches to inspire the LDs, EAs and to show the client
4. LDs-draw out level on graph paper
5. CM plans cameras based on 2d map
   

2. Block out the level in 3d

1. EA scan in level map
2. EA place it on 3d plane to scale
3. EA extrude level geometry to the elevations written down on map
4. EA import 3d block out into the game
5. CM place preliminary cameras
6. The LDs tests the game play in the blocked out level, place temp encounters and event triggers. The LDs give notes to EA who made the 3d block out, till the level is working for game play
   

3 Create 9 –12 pencil drawings for each level

1. AD and CA all decide on 9 –12 locations to concept out
2. CA takes 9- 12 screen shots of the blocked out level
3. CA prints those screen shots
4. CA does a tight pencil drawing of the setting over the printout of the blocked out geometry
5. CA and AD go over changes
6. AD, GD, LD, and CA go over final level drawings.

4 Have your Environment Artists build the final geometry over the block out

1. AD gives EAs all the concept art and reference materials
2. EAs divide up tasks between team members
3. One EA starts to build sets
4. Another EA starts to build the terrain geometry
5. Another EA starts to make the textures and lays in preliminary lights
6. AD oversees the work
7. EAs create concept art for any changes they want to make. AD and LD have to approve any alterations to the set plan

5 Color the level drawings in order to plan the colors, lighting, special effects and textures

1. CA takes the 9 –12 piece of concept art per level and paints them color before the textures and lighting for the levels are done
2. AD oversees this work
3. AD gives the EAs the color concept art
4. The EAs use the color work to create textures and place the lights
5. If there are effects in the level such as fire, smoke or magic portals and the like, the CA paints there into the correct position as well
6. Then the AD gives it to the VFXA so he knows what to start working on and where to place the effects, and what they should look like. If he needs more information then the CA or SBA then draws out the FX sequence. The Ad oversees
   

6 Light the levels after textures are done

1. EAs do a final pass on the level lighting after the textures are at first pass
2. AD, LD, and the GD review the lighting

7 Place final Cameras

1. CM placed final cameras based on LD, GM and AD’s notes
2. Cameras approved by LD, GM, AD
   
   

Detailed description

a. LDs- create level flow chart

We actually never did this on TTT. But we did do it on Indiana Jones and the Infernal Machine, and it was very helpful and I wish we had done it on TTT. It is basically a level flow chart done on paper to give everyone and idea of the basic progression of the level.

Here is Rohan 1 from The Two Towers. It is a good example of what a flow chart should look like.


Figure 1

This is so basic but clear that at this point a lot of good ideas can be added to it. This leaves a lot of room for creativity. A lot of people can contribute at this point, but the Level Designer, client (publisher), producer, and art director should be involved. Which leads to-

b. GD, LD and AD- discuss level content in meeting

So now that the basic game play and theme of this level is establishes the team leaders can all join in and start brain storming ideas to flesh out the level. A story board artist or concept artist should be in the room to rough out ideas so everyone gets a clear picture of some of the more complicated ideas. The ideas that are agreed upon should be written down and added to the Game Design Document. A CA and AD create preliminary color and design sketches to inspire the LDs, EAs and to show the client. The Concept Artist should then pick a few of the most dramatic areas and starts to come up with cool settings for them. The Level Designer should then incorporate these concepts when he starts to draw the final level map. They can also be used to show the publisher what the team is

d. LDs-draw out level on graph paperNow the level designer has all the input he needs and spend the next three days drawing out a detailed map of the level on graph paper that is accurate to the foot. He places on paper with pencil all the level elevations, enemies and interactive objects. He can also place roughly where he thinks the cameras should go and how they should work. This is also where he can layout where the portals and sectors too, though this will obviously change quite a bit. At least it can be roughly placed so the Environment Artists knows where they should be.


Figure 2

The red circles are enemies. The red lines are the portals. The colored lines are the sector boundaries. The top of the arrows indicates the height of an elevation. Blue objects are breakable or interactive objects. The Green T indicates treasure or power ups. The dark green shapes are trees and the light green areas are bushes. The white area is the walk able area, and the black line around them is the collision geometry. The blue areas are, of course, water.

1. Environmental Artist scan in level map
2. EA place it on 3d plane to scale
3. EA Extrude level geometry to scale written down on map

Once the client, the producer, the art director and the Game Designer have approved the detailed level map, the Environmental Artist will scan the map into the computer. The scan will then be projected on to a 3d plane, and the heights are then extruded to the correct elevations, all this according to the notes on the map (see figure 3). He also puts in temporary structures that fit exactly to the measurements the Level Designer wrote on the map. Collision geometry is also placed in. The Environment Artist then converts the geometry and map into textures that can be imported into the Play Station 2. The Lighting Artist can also place preliminary lights and the camera man can do the same with the cameras.


Figure 3

d. The LDs tests the game play in the blocked out level extrusion, place temp encounters and event triggers. The LDs give notes to EA till the level is working for game play.

The level designers now run through the level and time it out. If the engine is far enough along, the level designer could also place the gates, temporary interactive objects and the enemies to get a good idea for the flow and the timing of the level. If there are some issues with the block out, then the Level Designer can get the Environment Artist to change the blocked out and fix it. The Client, Producer, Art Director and Game Designer should also play through the blocked out areas.

1. Art Director and Concept Artist all decide on 9 –12 locations to concept out
2. CA takes 9- 12 screen shots of the blocked out level

The basic geometry is there, now it is time to plan out the details. The best way to do this is for the Art Director and his Concept artist to load the 3d map up and then take 9 –12 screen shots of the most important areas that need to be planned out. The Environmental artists will need this information so they know what build and what to texture. Also this art will help the Client see what the game will ultimately look like.


Figure 4

3. CA prints those screen shots out
4. CA does a tight pencil drawing of the set over the print out of the blocked out geometry in 3d.

The Concept Artist prints out the screen shots, and then traces over them with pencil on a piece of marker paper so that he has the perspective and scale correct.

Then he refers to the preliminary concept art, set designs and reference photos and starts to draw in the details, like sets, props and foliage, as well as any particle or special effects. See Figure 5 and 6.


Figure 5


Figure 6

Once the pencil sketches are done for those areas, the boards are then given to the Environmental Art team to use as a guide. A list of all the objects and buildings can be generated by these drawings, and estimates can be assigned to each of these models. This will help make the schedule more accurate.

For example in the scene I did we can determine that we need to build a fence, a guard tower, a bridge between the towers, a palisade wall, palisades gate, two houses, a tree, tufts of grass, and a stable building. Also the programmers now know we need to create realistic puddles and splashes when the character walks through them. The grass is going to need to sway and so will the trees. They smoke and fire are going to have to act like they are being blown in the wind.

1. Art Director gives Environment Artists all the concept art and reference materials
   This is pretty elf explanatory. The AD gathers together a book of all the concept art, storyboards, and photo reference (in this case photos from the Lord of the Rings set) the team needs to do the art.
   
   
2. Environment Art Team then dives up tasks between them
   The EA team has a lead and he and the other two members decide how to divide up the tasks. For example Andy is the EA team lead. His team has been assigned three levels to do in sequential order. He assigns Josh to build the buildings and trees, and he assigns Ed to create the textures for the landscape geometry. And he himself will do the landscape geometry. They all sit next to each other and work together. They use the level storyboards to guide them.

If they want to make a change to the level, then they make a change to the level storyboard and get that approved by the Art Director and if necessary, the game designer and level designer. If approved, then they can go ahead with the change. For example, what if the frame rate is taking a hit in this area, so one of the Environment Artists says ‘add a cliff to the back of the stable so it blocks the view outside the wall (see figure 7)’. He can take the sketch and draw in his idea. Then we can show this to whoever needs to approve it. This is simple and quick and easy to understand.


Figure 7

1. CA takes the 10 –12 piece of concept art per level and paints the color and lighting into them before the textures and lighting for the levels are done. AD oversees this work

Once all the detailed pencil drawings are done the Concept Artist works with the art director to determine the colors for the level. In this case it is late afternoon in Rohan, and set out on rocky plains. The Art Director in this case says ‘use lots of pale yellows for the grass and warm earth tones for the buildings, and use a cyan blue for the fog. The light from the sun should be an orange yellow and the ambient light should be a pale cyan blue (see figure 8).’


Figure 8

If there are effects in the level, such as fire, smoke or magic portals and the like, the Concept Artist paints them into the correct position as well. This will tell the programmers that the game needs realistic smoke and fire that blows in the wind and around objects like buildings. They can add this task to their priority list. This tells the lighting artist there is going be flickering orange lights placed at the fire locations, and it tells the particle affects artist where to place the fire, smoke and heat distortion emitters. He can add this to his task list now too. The environment artist now can see where the smoke is going to blacken the geometry and can create damaged models too.

1. A. Environment Artists do a final pass on the level lighting after the textures are at first pass.
2. Art Director, Level Designer, and the Game Designer review the lighting.

Now the lighting artist and the texture artist can see what colors they need to use and where they should be placed. The texture artist has to finish first before the lighting artist can finish up, because textures can affect the color of the lights. Again, they both look at the color story board art to see how to make the scene look. The color art represents the art director’s vision for the scene.

Once the first pass of lights are placed, the Art Director, the Level Designer and the Game Designers then check the work to se if it looks good and that it works for game play. The lighting artist does a second pass, and maybe even a third pass till it is final.

1. CM placed final cameras based on LD, GM and AD’s notes
2. Cameras approved by LD, GM, AD.

The final step is placing cameras. Now the planning for the cameras should be done roughly in the 2d map stage by marking on the map where the cameras will go (see figure 9). Special attention should be paid to the portal and sectors, because you can only show one portal and two sectors in the shot at a time. Then attention should be paid to the scene direction between the cuts, just like in Filmmaking 101, and then game play coverage is the next important thing to consider. Last on this list of concerns should be creating dramatic and aesthetically pleasing shots and camera angles.


Figure 9

After the 2D map stage of camera planning is done and approved, the cameraman can then place first pass cameras in the blocked out geometry stage. This is where the real work of camera placement can get done, because once the player is running around and temporary geometry is in, the cameraman can get a real feel for how the cameras are going to look. He can’t do the final cameras till the creatures are in, and all the game play works, and the lighting is done. But he can do the bulk of his work in the 3d block out stage with much of problem.

I really want to emphasize how contentious camera placement can be. Programmers, game designers, level designers, and the art director and environment artists will all want to have their say because the camera placement affects all that they do. For example is the camera is always facing north in a scene then the geometry from the north side of all the boulders and tress can be removed to save memory and speed up load times. But if the Environment Artist has no idea where the camera will go he will just assume the cameras will see the north side of all those objects and build them in. Another example is if the camera needs to back up in order create a long shot, but it backs up inside cliff geometry or into a tree, the world geometry is going to have to be reworked to accommodate it. But if the cameras are placed during the 3D block out stage, this problem can be detected early and the environment artist can make sure when he builds the final world geometry, there is room built in for the camera.

If the camera placement is done early on, one can avoid the ever so painful tug-o’-war that will happen between the three disciplines near the end of the project. So plan cameras from the beginning. Your future self with thank you for it.

When I first came on the TT team as the Lead Environment Artists each environment artist built, lit and textured his own level. I didn’t think this was a good idea but the Art director at the time wanted it that way because that is how they did things in the past. I wanted a more streamline, assembly line like system where artists specialized in the tasks they were naturally good at. When I took over as Art Director, with the blessing of EA, that is what I did.

EA wanted to ‘to go deep, rather then wide.” We had twelve modelers, and in the past we would have had them work on one level each. But with this new philosophy we created four ‘fire teams.’ A stupid name, I know, but that is what we called them. They consisted of one team leader and two to three subordinates. The team lead coordinated all his teams’ efforts and was responsible for all the art in his team’s level. The tasks were assigned to who could do them the best. Fort example Rob was fast at geometry so he built the landscape, while Ian was good at trees and buildings, and Josh, who was good at textures, would create all the textures or do the final pass on the textures the other Environment team members did.

Usually the team leader made the world geometry, another team member would build the sets and props, and a third would do all the final textures. This could not have been done without EXTENSIVE preplanning. Each team had detail level maps, 3d block outs, concept art, storyboards and color storyboards as well as tons of photo reference from the movie. It was the job of the Art Director, me, to make sure they followed the plan.

In order to give all the locations a consistent look we would assign the same team one or two settings. For example one team did Rohan Level 1 and Rohan Level 2. While another team did Fangorn 1 and Fangorn 2, and Balin’s tomb.

Also, the team would work on one level till it was at Alpha, and only then would they start on the next level. That meant that some levels near the end would either get cut or shortened in order to make it in to the game. So the levels we thought were the most important, the most difficult and most labor intensive were done first. This created a short game, but if EA had instituted this system early on they could have gotten in a lot more levels.

I hope all this is clear and makes sense, and more importantly is adopted for games where it is appropriate.