01/06 Concept Begins

This blog will act as the dev diary for our new mobile game. I [Jazz] will be keeping track with development by posting updates here, through pre-production to release, with input and feedback from the rest of the team. 

Chris Leyshon has joined the party! [As Lead Programmer]

 

Concept

Based on the market research I blogged about earlier, I came up with a small concept for a F2P phone game. The player-character would enter an endless labyrinth, where they would have to survive enemy attacks.
The breakdown for this game would be as follows:

• The game will be divided into ‘levels’ that would be separate mini-mazes.
• Number of enemies that spawn will increase as the player progresses
• Killing all enemies in a maze will give the player access to the next level
• Procedurally generated top-down 2D mazes. Edit: After talking with Chris, we have decided that creating modular rooms and paths would be a better way to move forward.
• At the base-level the game flow would be:

1.     Enter game

2.     Fight monsters

3.     Continue to the next level

4.     Survive as long as possible

5.     Repeat

Monetisation

To minimise removing the player from the experience, the game will be monetised through 'buffs' that can be accessed from the overlay that appears before games/between levels/after death. The player will be able to spend a single F2P currency to add the buff to their character. These buffs will be persistent through a single playthrough.

The main thing I will have to consider next is player retention

Level Design

Examples of grid sections. 

At the moment we are not sure on maze/level size/shape; first thoughts are that the full maze wouldn't have to be a perfect grid. The shape of the individual levels will not necessarily be important, but the numbers of rooms and overall size may increment difficult, therefore we will need to be careful with how we generate mazes. This will likely be easier to decide on once some of the base mechanics are in place.

 Procedurally generated dungeons from 'Dark cloud'

The maze itself will be randomly generated from modular pieces we create. Of course, all grid pieces would need to be connected directly with the starting grid piece and be passable, the square areas would be considered 'rooms'. And all grid sections could be any rotation [0, 90, 180, 270].

Art

We do not yet have an artist on our team, though we would like the game to be played with a classic Zelda-style top-down pixel art if possible. 

Dungeons of Dredmor

A great, modern example of the view and dungeon-style game we are looking for is Dungeons of Dredmor. It is an extensive game in comparison to the type of game I have in mind, though it has turn-based monster attacks which could be easier than fast-paced fights for mobile.

05/06 Retention

Retention

One of the main things our game lacks so far is the draw for the player to return to play and the aspiration goal – the thing the player would like to achieve overall etc. We have discussed having loot in the game, like a lot of dungeon/rogue-types. Loot and collectibles are an excellent incentive, but I have been pondering over what the loot would be used for. There must be another way to keep the player coming back to play/spend loot etc though. On a high level it should be loot-survive-escape-spend loot-adventure further-repeat

One idea I had was that loot could be spent to accumulate multiple characters with different traits you could 'buy' or upgrade and use, and perhaps take a little team back into the labyrinth eventually. Though that makes the scope much bigger than I had intended,

A simpler method for now would be character-development. the player can take the loot they have earned in the dungeon and use it in the main menu to spend on developing their character stats before returning into the dungeon to travel and fight further than before in a newly created menu. 

Editor Tools

Chris has been working on the visuals of a piece collision editor. Once he has done that, he plans to work on the actual maze/dungeon generator and then navgrid generation and pathfinding.

 Creating a path through the maze

It's a modular piece editor which allows you to decide which directions you can enter and exit each tile from. We were going to have a few set paths and rooms and then they could be programmatically rotated to suit the map. Chris had just figured out the rotation for the navgrid, when we discovered that this maze creation method was not going to work visually using the rotation.

1. We discussed at this point if the view would be fully top-down similar to the example on the left...

 with this we would be able to rotate each room/path and most of the objects in the environment would be seen top-down

2. ...or tilted up slighting, similar to the example on the right...

this view gives a better perspective of the environment and allows for interesting visual barriers/blockages that the player will need to overcome and it is also a more attractive angle to look at the assets and animations...

a downside to this is that we will need bespoke wall tiles for the top and lower wall sections as well as the side walls.

Even though Chris has finished the navgrid at this point, we both decided to go for the second POV option. It does mean that we will need to make bespoke prefab sections for each path rotation, but I think it will pay off. We want to avoid rotating the pixel art since it will look odd. 

Tile division 

Looking ahead

Chris and I spoke at length over F2P/P2P prospects. We both prefer P2P, but agreed that F2P is the way to go for our mobile title. We would like to aim for P2P for a potential steam release much later on though.

Chris and I spoke at length over F2P/P2P prospects. We both prefer P2P, but agreed that F2P is the way to go for our mobile title. We would like to aim for P2P for a potential steam release much later on though.