B.A.M. | Sprint 1 | 9/1-9/10

      B.A.M. is a sci-fi strategy action game I and my team are developing for our Advanced Production class at Chico State. I am one of three Programmers in our 14 person team. This post details my work process across our first Sprint.

Enemy Targets Building
    For this Sprint, my work consisted of the fundamental enemy AI. I have very little experience with Unreal Engine, so I've been a little slow with my work so far, but I've really been learning a lot. To start, I made a base enemy as a Character and dove into Behavior Trees for the first time. The enemy really has two main behaviors: attacking the nearest building and retaliating against the player when it's attacked. Currently, the attack and movement are both templates to be filled with the proper mechanics.

Enemy Targets Player

Base Enemy Behavior Tree

Enemy Object Pooling
    Once I got the enemy prioritization and movement done, I moved onto getting it to spawn in the world. For this, I implemented an object pooling system for sake of optimization. However, this object pool gave me a bit of a headache because of the way my enemies were functioning as characters. I ran into issues with the behavior tree using old data when an enemy was activated and how to turn off and restart that AI Controller logic. Ultimately, I did get it working in the end and learned a lot more about Characters and Controllers in Unreal. The enemy spawns within a random interval of 10-12 seconds.

In-Level Spawn Locations

Weighted Spawn Locations
    As part of the enemy spawning, we wanted in-level placeable spawn locations where enemies will spawn one-at-a-time at randomly. These locations also are weighted by how many buildings are nearby - enemies are more likely spawn at a location with more buildings nearby to attack. This weighting took me a decent amount of time figuring out how to do it. As it stands now, every location has a default weight of 1, and gets an additional +1 weight for every building within the acceptable radius nearby. Locations are then picked with a random number up to the total of all weights, ie: two spawn locations with weights of 5 and 4 means a random number up to 9 is generated, if it's 5 or below the first location is picked and if it's 9 or below (after checking the first one) the second location is picked. The default weights and acceptable radii are all instance editable, so each location can be tuned after it's put in the level. This weighting system is also planned to be expanded to take into consideration different locations being more likely to spawn one enemy type over another, or for different buildings to be more important and give nearby spawn locations more weight.

In-Editor Debug of Spawn Location Building Detection Range

Comments

Post a Comment

Popular posts from this blog

Grapple Go | Sprint 3 | 9/25-10/9

Image
     Grapple Go is the game my team and I are developing for our Mobile Game Development class at Chico State.  I hold the role of programmer and m y team consists of Justin Culver as producer, James Songchalee as designer, and Sophia Villeneuve as modeler. Our game is a 2D side-scrolling infinite runner, similar to Jetpack Joyride, where the player mainly interacts through the use of a grappling hook to dodge obstacles and enemies. The player's goal is to try and cross as far a distance as they can while racking up their score to be later used to buy upgrades, helping them in their next attempt.     This post details some of my process working on the game during our Sprint 3. Demo of Shield Powerup Challenges & Problems     I think the biggest challenge I faced this sprint was in the process of refactoring and reworking a lot of my previous code. I spent a good amount of time reworking my player controller and grapple controller, ma...
Read more

Grapple Go | Sprint 1 | 8/28-9/11

Image
Originally Posted: Sep 14, 2025      Grapple Go is the game my team and I are developing for our Mobile Game Development class at Chico State. I hold the role of programmer and my team consists of Justin Culver as producer, James Songchalee as designer, and Sophia Villeneuve as modeler. Our game is a 2D side-scrolling infinite runner, similar to Jetpack Joyride, where the player mainly interacts through the use of a grappling hook to dodge obstacles and enemies. The player's goal is to try and cross as far a distance as they can while racking up their score to be later used to buy upgrades, helping them in their next attempt.     This post details some of my process working on the game during our first two week sprint. Demonstration of the grapple Differences in the Process     The main difference with project so far, as compared to my past projects, is that the size of my group is larger. We're a group of 4 as opposed to 3. This has made com...
Read more

BG3 Goblin Camp Scene | Sprint 2 | 11/5-11/12

Image
Originally Posted: Nov 12, 2025      This project is an interactive game scene recreation inspired by Baldur's Gate 3's goblin camp  my team and I are creating for our Game Asset Production class at Chico State. I hold the role of lead texturer and my team consists of Julian Kroeger-Miller as project lead and producer, Els Fouché as lead programmer, and Destiny Smith as lead modeler.     This post details some of my process working on the project during Sprint 2 of our work phase. Finished the Template Blockout for the Central Area      I'm not very experienced with Unreal, so between this project and the last Modular project I'm learning a lot about how to use and interact with it. Seeing as I was essentially assembling the entire scene iteratively, it required constant trial and error figuring out the correct dimensions and frequently adding new pieces or variations to the asset list. Blueprint Prefabs Tusk Prop Stone Vase Prop Cu...
Read more