Status Update for The Colon Case - Searching for its Visual Style
Back at the beginning of 2021, I started developing a noir-inspired detective game called The Colon Case. I was able to get basic character movement going, and made progress on being able to display blocks of text in SFML. Development on the game then stalled as other pursuits took precedence, but even so it always remained in the back of my mind.
At the time, I was planning to exclusively use the somewhat-dated MovieStorm as the basis for its visual style, exporting animations of the characters into sprite sheets. The overall process was proving to be a bit more tedious than I hoped, and MovieStorm doesn't have as much variety as I was looking for in terms of sets and set dressing.
I was also having trouble envisioning where the fun of the game would come from. I really liked the story setup, and the cast of characters we'd come up with felt like a major strong point. But at the end of the day, from the perspective of the player, most dialog-driven games are just a dialog tree where you play through it once to see what everyone says and you're done. I wanted something more... something special and unique that you don't typically see in a game.
In the recent weeks, I've been brainstorming more and more about The Colon Case's visual style and its dialog system.
I still like the idea of using MovieStorm to create sprite sheets, but perhaps only for the "portrait-view" that comes up when the player enters dialog mode with an NPC. It has very expressive facial animations that run the gamut on every possible emotion. Meanwhile, a new contender may have come up for the "map-view", where the player walks around the room to look for evidence or make their way to the next NPC. Released in 2021 on Steam, Clip Maker appears to have a level of variety that didn't exist in MovieStorm, so I'll be looking more into whether its suitable for exporting character animations. The "map-view" is top-down, so I would just need the characters to have similar enough attire to match their "portrait-view". Even if character animations don't work out, it might be a good fit for exporting sprites of props and set dressing.
The dialog system is where I've put some real thought, and I think I have some compelling ideas. I want to use a custom procedural language generation algorithm, where NPC dialog is influenced both by personality traits and a shifting emotional state. To deal with the combinatorial explosion that can happen in narrative-driven games, dialog wouldn't be expressed in a massive branching dialog tree, but instead as a response to topic and intent. So dialog authoring will essentially be a flat list of potential responses with increasing specificity on the NPC's current emotional state. This is where I think the game can derive its fun... The player isn't just asking pre-determined questions and getting pre-determined responses. The player is trying to drive NPC's into revealing new evidence by choosing both the topic and the questioning approach. So it becomes a psychological cat-and-mouse game, where the player has to figure out the personality traits for each NPC in order to get them to reveal evidence or knowledge they may have been trying to hide. Some NPCs might be more prone to accidentally revealing themselves through fits of rage, while others will shut down completely when angered. Anger, fear, joy, jealousy, relief, admiration, and so on -- they'll all be in the player's toolset for solving the game's story.
The idea behind this has me excited, so I plan to explore how I can represent all of this in the game's state, and how I can go about authoring the dialog for each NPC.
Over the next few months, I'll be primarily working on my web-based puzzle game, Peg City Parks Jigsaw. But I plan to intersperse some Colon Case for a few hours each week into my schedule! 😊
