Devlog #1: Lost Echoes – First Dev Log

Welcome to the Journey

Hello, everyone! Welcome to the first official devlog for Lost Echoes, a sci-fi psychological horror game I’ve been pouring my heart into for the past three years. If you’re new to the project, this post will give you a quick overview of what Lost Echoes is all about. For those already following along, feel free to skip to the New Features section to dive into this week’s updates.

Lost Echoes is a chilling blend of sci-fi and psychological horror with survival mechanics, drawing heavy inspiration from games like P.T. and Iron Lung. Initially developed in Unity, I made the switch to Godot after some licensing concerns to ensure a smoother development process. The game takes place aboard a compact Torus Station in the Spirit Tamer universe. Your mission? Complete a series of tasks, interact with an NPC via the Comms Console, and manage your character’s basic needs to survive the eerie atmosphere.

With that context set, let’s dive into the exciting new features I’ve been working on this week!

New Features This Week

Check out the video below to see these features in action, as shown in the original devlog recording:

.

1. Enhanced Dog Entity Behavior

Building on last week’s addition of the dog entity, I’ve implemented a set of checks to make its behavior more dynamic and immersive. The dog now reacts based on environmental conditions—specifically, whether the area is dark and if the player’s flashlight is on. This creates a tense dynamic where the dog only responds when it’s clearly visible to the player. (The video above demonstrates this effect, though it might be subtle depending on your screen!)

2. New AI State: Search Mode

This week, I introduced a new AI state where the dog actively searches for the player. To give you a sense of how this works, let me share a neat trick I picked up in Blender that makes this feature shine.

Blender Rigging Magic

Instead of manually tweaking values to make the dog’s head track the player, I used Inverse Kinematics (IK) in Blender. The dog’s head and neck are rigged to lock onto an empty node, which simplifies the process and ensures smooth, natural movement. This empty node is scripted to maintain consistent orientation within the Lost Echoes environment, adapting to the Torus Station’s unique layout. While the scripting details are a bit complex for this post, the result is a lifelike tracking system that feels intuitive in-game.

How It Works in the Game

In the game, I’ve set up predefined points in doorways that the empty node can lock onto. The system checks if a door is open, and if so, the node aligns with the doorway’s position, making it look like the dog is peering into rooms as it searches. Here’s how the AI flow ties together:

• Search State: The dog roams, checking doorways and scanning for the player.
• Spotting the Player: If the dog sees you, it pauses and observes.
• Player Proximity: If you get too close while looking at it, the dog bolts, adding to the tension.

You can see this behavior in action in the embedded video, where the dog’s search and reaction states come to life in the Torus Station’s eerie hallways.

What’s Next for the Dog AI

Moving forward, I plan to refine the animations to make the dog’s movements even more polished. I’ll also expand the finite state machine to include both normal and aggressive states, adding depth to the dog’s behavior and making encounters even more thrilling.

Wrapping Up

That’s all for this week’s devlog! I hope you enjoyed the video and this peek into Lost Echoes’ development. If you want me to dive deeper into the technical details (like the Blender scripting or Godot implementation), let me know in the comments below. Your feedback helps shape these updates, so don’t hold back!

Until next time, stay spooky and take care!

Follow The Silicon Abyss for more updates on Lost Echoes and other projects. Subscribe to our newsletter or join the conversation on X for real-time insights!