Overseeing Level Design

In Star Wars Rebels, I primary goal was Creative Director.  With respect to level design, there were many things I had to consider during the production and per-production of the game, including various qualitative aspects and some more functional considerations including workflow and process.  For now, I will speak to a few of things we considered.  

I mostly on worked to define the broad strokes of each level, scripted movements and their overall concepts.  I worked with 4 separate level designers to conceive, design and approve each level.  I worked with the senior level designer to craft the playable experience early in the project.  Because of the tight time constraints, We made assumptions about enemy and player abilities.   The platforming mechanics were normalized around a grid based system.  We had a senior level designer define the jump and movement parameters using some basic Developer Gym levels. 

In parallel, I worked with the Art Director to establish the visual target.  We selected to produce level-them-kits that could be permutated to create more varied scenarios, that where iconic and that worked to the strengths of the engine.  We also picked elements that were naturally reoccurring and repetitive.  This would help reduce costs to production without having the player notice. I worked closely with the brand stakeholders to sign off on the selections we made.

I oversaw the workflow for level production and worked with the client to define the overall parameters of what the team would build towards.   As Creative Director, I also defined specifications for enemy and entity progression.  Each level would be 2 minutes in length and consists of 3 unique entities.  The roster of levels were conceived at a very high level level.  Each level was described with one sentence and some reference work art.    We defined some level as WOW levels.  These levels would be more expensive to produce but contain set-piece or scripted aspects that would be used as tent-poll between simpler grinding levels. Since the player was partially motivated by experiences of the game,  we wanted properly reward them with enough new elements.  A metric was defined to help quantified the pacing and uniqueness from moment-to- moment.    This parameter was known as value-per-square-foot.  Level designers were coached to work towards this.  One aspect to consider during iteration was how much we front loaded the game.  It was important to hook the player in early and impress them with cool and epic 'WOW' moments.

We established a formal work flow to produce the level content of this game.  Blocking for levels were produced as wire-frames inside our editor before geometry and a basic lighting pass were applied.  Levels were balanced and game play was approved before they were they were passed to the art team to light and polish.  

Perspective Test

Perspective tests renders and per-visualization tests of 3D world rendered with 2D sprites.  I wanted to validate the degree of camera shifts we could achieve before perspective errors on the 2D sprites became noticeable.  A layer based level design approach was ultimately abandoned due to brand restrictions.   A non-contentious ground plane was initially chosen to reduce production budget and polygon budgets.

Level Design: SW_B10_AST_Mines_TramAscend

In the end, I felt compelled to get my hands dirty and design a single level.   I would have loved to have done more, but time was spare for me.  The basic level level, with scripted elements and basic lighting was handed off to one of my level designers and Environment Artist to finalize. The concept focused on diagonally moving elevators.   Pardon the lack of cinematic/in-world shots of the level.  The game was a side scroller so levels were blocked with assumption that the camera would look towards the -Z-axis. 

Description:  Traveling through a volatile and hazardous mining facility, the hero's attempt to return to the surface.

Enemies:  (StormTrooper Sniper, Stormtrooper Basic, Minion Droid)

Playtime: (2 minutes)

Here are are a few images of the entire map:

Adventure Style Levels

Early level design schematics of metroidvania style levels.  I wanted to go with mini levels that had multiple paths. Sections could be reused or closed off to create new levels.  This would have allowed us to reuse significant blocks of levels without even copy and pasting.  This would have fostered more adventure style exploration as well.  In the end we had to abandon this approach, in favor of a more linear based level structure, as the complexity in managing it was outside the scope of what we could achieve.    We still see notions of this construct in the final product, manifested through secret rooms, areas and dynamically content seeded by the games quest system.

Gameplay and Scenario Design

Once the core gameplay was established, time was spent designing a vocabulary of reusable scenarios. The following diagrams articulate a few examples how we communicated these scenarios internally and to stakeholders.  Players feel empowerment when they recognize and adapt to a pattern.  These scenarios implemented and then seeded into the various levels. The goal was to create a set of platformer scenarios that could be reused and permuated to create recognizable but escalating challenges.  By reusing a set of scenarios,  Additional scenarios were added as the production when on and game play mechanics naturally emerged. I fostered design session and helped drive the associated user story documentation.  I worked closely with the produce and level designers to ensure we could propagate these scenarios properly through the game. 

Another challenge was new entities would often emerge over the course of the production, such as physics based anty-gravity platforms.  These entities were unplanned and emerged organically.  They were very cool and we often for free.  We often assessed if we couldintroduce these mechanics into levels that were already assembled without impact to schedule or the integrity of the project.