Early Level Design Sketches

   I am going to be sharing work from my sketchbook. Last January my team and I began work on an ambitious project. We wanted to make a third person cover based shooter. Something along the lines of Gears of War or Uncharted. I am a big fan of Gears of War and my project lead and friend who is a big fan of the Uncharted series. We wanted to make this game inside of UDK and we wanted to make something that was drastically different from out previous creation Freeze-E Frosty's.

Note: This blog will feature early and rough sketches I did throughout the course of my time as Lead level designer.

Early in development my head was full of ideas. Here I was already planning the early parts of the game, from weapons to location and level layout.

  When the project started we gave it the code name of "Max Power". From my viewpoint we wanted something that sounded cool but didn't take itself to seriously. When I started putting ideas in my sketchbook I was brainstorming as much as I could. This early in development we were trying to figure out what we wanted to do. The UDK engine was great for someone who wanted to make a first person shooter and we knew that making a third person cover based shooter was going to be a challenge but we were willing to accept it.

  Early on we wanted to nail down the story. When we made Freeze-E Frosty's we were able to figure out just what areas the player would explore and what assets, vfx and sfx we would need.

  When It came to level design the first thing I was thinking about was the pacing. We had to start off with a good tutorial that brings the player straight into the action. I was playing a lot of games for reference. What games had a truly engaging tutorial? What games had a bad tutorial? I was looking to design something that would instantly teach the player the fundamentals of our game. The fundamentals at this time was gun play, puzzles and cover. 

This early sketch was to determine how the entrance to the temple would look. We picked a hourglass theme for the temple design.

   The second sketch I made was of this camp the player would explore. The early story had the player working as an aid worker out in the desert. The camp where he was at had stumbled across some ruins of some sort of ancient temple/civilization. The main character leaves the camp to head into town to gather some supplies and when he returns the entire camp has been destroyed. There are bullets casing all over the camp and the doors to the temple have been blown open. The player would than explore the campsite until they realize that survivors of the raid had escaped into the temple. The player must travel into the temple to find these survivors while also contending with the threat of these treasure seeking desert bandits.

  While the player explored the destroyed camp we would be teaching the player how to interact with the environment and how to walk and move throughout the world. Once we get inside the temple we would teach the player some simple puzzle solving and than we would teach them how to use a gun and take cover as they first encounter enemies. The player would be driven forward to find the survivors by a walkie talkie that keeps them in contact with the survivors. We used this mechanic in Freeze-E Frosty's as well.

Working out how cover would work in the game.

  Before I even begin making the level inside of UDK I sketch it out. Some of the first things I thought about was the flow of the game. Cover is going to be a huge part of our game. Just how much cover should we have without it affecting the flow. When I talk about the flow of the game I looked to games such as Unreal Tournament and Call of Duty. Games that allow for fast and frenetic game play. The thing is with a cover shooter is that the pacing is different. You want the player to have the need to take cover but at the same time you do not want the player to just sit back and turtle up.

 The layout to the level was something I was constantly thinking about. I wanted to give the player options. They player should be able to move around the battlefield and engage enemies at different ranges and angles. The encounter should be fairly random with enemies laying down covering fire and trying to out flank the plater. These were all things I was thinking about.

Entrance to the level inside of UDK.

Puzzles were something were going to be a big part of the game early on. We dropped them because we felt as if they slowed the pace down too much.

  One game we looked at for inspiration was the Tomb Raider series. I had played and enjoyed the puzzle and exploration aspects of Tomb Raider : Underworld. Early on we were trying to implement a very similar style of game play where the player can climb ropes, ledges and shimmy along walls.

 Some early Ideas I had was using ropes to swing across gaps and clinging to walls to avoid crumbling floors. The first puzzle room would be devoid of areas that would kill the player. One thing I thought about was making the first puzzle room really easy. The idea was to make death in this tutorial real impossible and keep the punishment level low. When the player is unable to complete a task such as crossing a gap they would fall into a pit. To keep the puzzle from punishing the player we make the fall a short one. We also make it very easy for the player to get back up to the rope and try again. If the player was to fall into a pit and die they would be a bit deterred. If they are having trouble learning the rope swing mechanic and they die over and over again they will just simply quit the game. The tutorial level is meant to maximize learning and minimize punishment.

  If there is one tutorial that sticks out in my mind as a good one it is the tutorial from Red Dead Redemption. When you are on Bonnie's ranch it teaches you everything you need to know about how to play the game while still working in the confines of the story. The tutorial teaches the player how to ride a horse, use the lasso, lock onto enemies, fire/reload a gun and use the map.

Sketches for what would be the center piece of the temple. I came up with the absurd idea of having a giant hourglass in the center of the temple that controlled everything that went on in the temple.

More sketches of the hour glass chamber.

Hourglass chamber inside of UDK.

   Initially we chose a temple because it would mostly be an indoor game. With the game taking place indoors we would save a lot of time with assets and we wouldn't have to worry about dealing with a horizon or filling space in the sky. Game play wise I was thinking about setting up the temple to be something sort of out of Legend of Zelda. The temple would have one main hub area that would connect to different parts of the temple. This would allow for maximum asset usage and minimal asset creation.

   The main hub area was going to contain a giant hour glass that would control all of the rooms in the temple. The player would need to scale the giant hour glass to activate it, revealing rooms and different paths in the process.

  Just about every puzzle room would be based around the use of hourglasses. Combat rooms would also be a part of the exploration as well. The player would need to back track to different rooms to unlock more areas inside of the main hub area.

 I based this element of the game around games like Legend of Zelda and Metroid Prime. What I loved about Legend of Zelda was that the temples were built around the item you find in that level. Once you find that item the rest of the temple is an exercise is learning how to use it. This is important because the item is than used to defeat the boss battle at the end of the temple.

The quicksand pit room and the wave room were two puzzle rooms I would of loved to see realized one day. I also drew a picture of that frog suit from Super Mario 3.

  When I was coming up with puzzles I was trying to develop something that had the player being an active part. There were two puzzles I came up with. One was the idea of a wave room. The room would be a long hallway that was full of water. The hallway would move a lot like a seesaw. This caused waves to move from either side of the hall. The player would not be able to open the door at the end of the hall until they drain the water out of the hall. The way the player gets the waves to stop is to find an opening in the ceiling of the hall. Once the player gets there they must break an hour glass that was controlling the timer for the seesaw effect. Once the hallway stops moving drains open up in the floor and water rushes out. This allows the player to escape from the other side.

  The quicksand room had the player constantly moving in an attempt not to get sucked into the sand. Sand would be pouring into the room from a hole in the ceiling. The floor would be constantly shifting making it very unstable and unsafe for the player to stand on. The puzzle here was fairly simple, all the player had to was scale a wall while avoiding parts of the wall that would continuously try to knock the player off the wall. Once the player reaches the hourglass that is controlling the puzzle and deactivate it a hidden door would appear for the player to escape the room. 

   On the technical side of the quicksand room I was experimenting with different volumes inside of UDK to get that sinking feeling quicksand would produce. I came close by mixing a water and a gravity volume together but ultimately it was decided that Juan could just program a quicksand volume. Later on in development we decided to drop the temple idea and to drop the Tomb Raider style exploration. We decided to focus on the shooter aspect of the game.

A sketch of how terrain would affect the level layout.

The above sketch inside of UDK with some placeholder assets. Art assets done by Juan Valenica.

More level layouts with terrain.

Close up of the UDK level. Placeholder art assets by Juan Valencia.

  We wanted to mix the combat up. We began to leave the idea of a temple behind us. We wanted the game to be more open, open in a sense where we are not indoors all the time. At this point we started to look for something that would allow for that. We looked at a few different locations. We wanted some place that screamed with life and energy. Juan ended up picking this place in Japan called Yakushima forest. It was a visually interesting place.

  As we stepped away from the indoor temple feel of the game we kept the hour glass aspect. I began to make a level that was something that would come after the tutorial. At this point in the game Juan had programmed a new volume into the game that allowed for the player to climb walls. I designed a map where the player would climb up a cliff, fight some enemies and then once again climb over a waterfall where they would engage a small group of enemies.

  The main idea behind this map design was to get the player ascending, to get them climbing up toward there goal. The player started at the bottom of this map but as they moved forward they followed the path they would begin to ascend the mountain until the reached the top. This was supposed to be something that was more action oriented.

  This is one of the first areas I made inside of UDK. It was here were we first started testing out enemy ai and cover placement. Very early on the enemies were very rudimentary. The layout worked and served to move the game forward but the AI had a lot of shortcomings. 

The guard tower sequence is when we started addressing the limitations of the Ai inside of UDK.

   I was trying to add a guard tower to a level. Early play testing of the guard tower weren't going so good. The enemy AI would just trace the players location which lead to the enemy continuously shooting at the player. The enemy bot would never lose sight of the player. The problem with the guard tower was that it was supposed to be something that the player could avoid. As soon as the player entered the vicinity of the tower they were instantly spotted. As soon as the player was spawned they were spotted.

  This lead to a team meeting about programing in behaviors for enemy units. I picked enemy types (Pistol guy, shotgun guy etc) and Juan did some simple programing that changed their behavior and attack patterns. At this stage the enemy types were still being picked out. We decided that focusing on the level layout was something that was more feasible. Programming the enemies was not top priority but it was something that was on the top of our list of fixes.

- Danny Q
@Dannylv100
​