R

real-time strategy

Project Information This project was created as a portfolio piece for an Unreal Engine developer.

The game is a real-time strategy focused on capturing resource points (marked on the map with a coin icon) and destroying the enemy base. The player engages in combat against opponents who both guard the paths to the base and resource points, as well as attack in waves.

Game Mechanics Resource Point Capture: Resource points on the map (marked with a coin icon) can be captured to generate periodic resource income. Control of these points can be lost if the enemy recaptures them.

Unit Production: The player can produce various types of units at their base by spending accumulated resources. Production takes time.

Army Control: The player can select and command multiple units simultaneously, issuing orders to move, attack, or capture points.

RTS AI: Enemy units automatically switch to attack mode upon spotting the player’s units, and revert to passive mode if they lose sight of them.

Minimap: The minimap displays units, resource points, buildings, and the visible area not covered by the Fog of War.

Fog of War (FOG): The game features a Fog of War effect that hides parts of the map outside the vision range of the player’s units and buildings. The player only sees the surrounding area, while the rest of the map is darkened, adding elements of uncertainty and strategic planning. The visible areas update in real-time as units move, and enemies outside visibility remain hidden.

Camera Switching: The player can change the camera position with a key press to adapt the view for different situations and improve visibility.

Game Settings: Graphics, sound, and control settings can be adjusted through the in-game menu for player convenience.

Death Screen: Upon the player’s defeat, a widget appears offering the choice to restart the mission or return to the main menu.

P.S. The main focus was on gameplay mechanics rather than visuals.

Data-Driven RTS in Unreal Engine 5

This project started as a portfolio piece for an RTS game but evolved into a performance-focused learning experience. The goal is to create a highly scalable RTS game capable of handling large numbers of units (tens of thousands) with minimal overhead using Unreal Engine 5's Mass Entity and Niagara.

Initially, I used Unreal's Actor and Component systems to develop basic units and commands. However, as the project progressed, I realized that this approach was too resource-heavy for managing large unit counts. To solve this, I switched to Unreal's Mass Entity Framework (ECS), which allows for better performance and scalability by organizing data in a more efficient, parallelizable format. This transition drastically improved performance and made handling large numbers of units feasible.

For visualizing these Mass Entities, I moved away from the default Instanced Static Meshes and opted for Niagara particle systems. This change not only provides better control over visuals but also improves performance. However, integrating Mass Entities with Niagara is a challenge, as there’s little to no documentation on how to pass data between the two systems or make them work together efficiently.

Currently, the project is in its early stages, with the core Mass and Niagara systems being integrated. While some of the basic unit systems and mechanics are in place, much of the work has been focused on developing solutions to bridge Mass and Niagara into a custom hybrid architecture.

Future Goals: Expand the RTS systems, including pathfinding and resource management.

A real-time strategy game made with LÖVE based on the original Stronghold by Firefly studios.

An isometric real-time strategy game

Real-time strategy game set during the American Civil war. https://trilarion.blogspot.com/search/label/civil