• Developing a complete product to a clients specifications.
• Collaborating within a group with artists, a designer, other programmers and producers.
• Understanding the needs of the client, and effectively communicating with them.

• I developed the terrain manipulation and spline path tools.
• To enable the terrain manipulation, I also programmed a function that generates a procerdural mesh of any number of subdivisions.
• I developed a method for saving a created tile as a .csv file, as well as a method for importing a tile by reading the data within a given .csv.
• I developed methods that allowed the terrain mesh and objects placed in the tile to interact.
• I worked on integrating many of the UI elements with their associated functions, as well finalising the main menu.

• This was my largest project to date using Unreal engine, and I have definitely improved with the softwares tools, and C++.
• Working with a team was a valuable experience regarding collaboration with other roles and utilisation of tools, such as Jira and Perforce.
• Understanding a clients requirements and adapting to feedback allowed me develop my client interaction skills.

The Hexton Hills Tile Generator was developed as a group client project for my final year of university. I was personally developing procedural tools for creating and manipulating tiles. It was made using Unreal with C++.

The client was Graven Guild. They requested us to develop a program that allowed a user to create a hexagonal tile, decorate it in a variety of ways, and export either the tile data for sharing, or the tile mesh for 3D printing.

• The terrain tools I developed were:
  Brush tool: Raises or lowers a group of vertices concurrently.
  Vertex tool: Raises and lowers a single vertex.
  Noise tool: Randomly raises and lowers a group of vertices concurrently.
  Flatten tool: Averages and sets the height of a group of vertices.
  
• I also programmed clamps that stop vertices from going any higher or lower than their current height when applied.
• Developing this further, I would have liked to have optimised the tools that deal with a group of vertices, as performance can suffer when working with high complexity mesh.

• The path tools I developed were:
  Add point: Creates a new path at a point, or adds onto a pre-existing path.
  Select point: Grabs a nearby major spline point and moves it, as well as the connected edges.
  Delete point: Deletes a nearby major spline point and connected edge if it is on an end of the path.
  Delete path: Deletes the entire spline path associated with a nearby major spline point.
• Additionally, the spline will automatically segment itself into a number of points, so that it can bend with the terrain with varying terrain.
• The spline also has a number of mesh components across it, connecting each spline point to create a solid model across the entire length.
• Developing this further, I would have liked to have developed procedural functionality that updated the spline mesh based on external conditions, such as an intersecting path, or nearby object.

• In its current state, this tool can procedurally generate a hexagonal mesh of any size and "complexity".
• It can also take a given padding value, and create of list of mesh vertices that are on the tile edge that cannot be manipulated
• Developing this further, the code could definitely be optimised. During development I didn't have much time for optimisation, thus I had to accept its state as long as it worked. I would have also wished to develop a function that allows the user to change complexity during runtime.

• In its current state, placed objects can move with the terrain below them, and be toggled to rotate to always be flush with the floor or not.
• Buildings placed in a specific way also create plateau's beneath them, shaping the terrain.
• Developing this further, It would make sense to procedurally determine the size of the created plateau based on the size of a place object.

• In its current state, the user can save all essential data about a tile to a .csv, and on a name and location for local storage.
• These .csv files can also be imported to recreate the tile based on the data that has been recorded.
• Developing this prototype further, it would be interesting to find more efficient ways of abstracting the save data for recording.

• Near the end of the project, I was working on a new terrain tool that would take image data from a heightmap, and apply it onto the tile mesh to create naturalistic terrain.
• It could select a random image from a collection of .png files (from a selected directory), access it's pixel data, and apply it to the tile to create various terrain patterns.
• This feature was unfinished, as I could not complete the work needed to translate laying a rectangular image over the hexagonal tile.
• Regardless, working on this feature was a good opportunity to practice file manipulation and utilising image data.
• Below are images from the development of this tool. It shows image data being applied to the terrain, but in a disorganised fashion.