Unnamed "Tug-Of-War" Strategy Game

[lolz123]

Hello all,

I am currently working on a larger project with some friends, and I thought I should share the progress so far. Criticisms of all kinds are welcome, whether it be about the game idea itself, the graphics, or something else.

I don't have a working demo for you guys to try yet, but I have a few screenshots to share. I will post in this thread when new features are added to hopefully get some feedback on them. So, there might not be much to comment on now, but hopefully there will be soon.

The game is a strategy game, based on the Starcraft 2 mod "Colonial Line Wars". Both teams build buildings that automatically spawn units in waves. The goal is to push back the enemy all the way to their base, such that you can destroy their command center. It is sort of like a tug-of-war, where you need to build certain units to counter enemy units as well as achieve the proper formation to maximize effectiveness.

However, there is a twist. If you have seen my past forum posts, then you probably know that I love artificial intelligence. So, in this game, each wave the player rates their units. The next wave is then bred based on those ratings using an evolutionary algorithm. This way, you can indirectly micromanage your units by fine-tuning their AI.

The AI has a large number of sensors and can even speak audibly to other units. They invent a language through a combination of reinforcement learning as well as the player's selections. Right now I have some generated audio of beeps and boops that are modulated by the AI.

For those interested in what AI techniques it uses: It uses Long-Short Term Memory neural networks (LSTM) so that the AIs can form memories. These are placed into an actor-critic architecture, and are updated similarly to the networks in the Continuous Actor-Critic Learning Automaton (CACLA).

The game uses the D3D (Deferred 3D) engine I posted about a while ago, which uses SFML for audio, 2D rendering, windowing, and texture loading. The engine's parallel processing on the scene objects allows us to run hundreds of AIs in real-time.

Here are some pictures of an early test battle. Please ignore the UI, it is a placeholder.





The environment:

[Grimshaw]

Looking good! I guess by now I only expect good things from you man

[Nexus]

Looks and sounds really cool so far! You should really finish and polish this project, I could imagine it being very funny to play

I have played the Civilization Wars mod in Warcraft III, maybe you know it? The principle is similar, there are three lanes where units fight automatically against each other.

Did you also consider a variant where units are rated automatically? It may be less of a game with less user interaction, but it would definitely be an interesting experiment.

By the way, are you still developing your D3D engine?

[Peteck]

Looks really great and would love to play this. But can't help wondering if deferred rendering is abit overkill for the game as it is right now? Wouldn't simple forward rendering give better performance? Cause I don't really see any light beside the directional light. Also the shadows could use some better anti-aliasing. Maybe some linear filtering on the shadowmap?

But love the concept of this game and I will be looking forward to try it for sure!

EDIT: Also it seems like your shadows have some "peter panning" issues

[lolz123]

@Nexus: I played Warcraft III a long time ago, but I cannot remember if I played that particular mod. The game will have a VS AI mode, where the AI uses its own reward function. It would be cool though to see the battle play itself out on its own though 

The development of D3D has not halted, it gets new features as needed to make this game.

@Peteck: Deferred shading is good for post-processing effects like SSAO, which is used here. Additionally, if we want a nighttime scene, we will need lights. I also prefer the elegance of deferred shading over forward shading.

The shadows right now use percentage-closer filtering, I might switch over to variance shadow mapping eventually, since then I can use the hardware filtering.

As for the peter panning, I think I went a bit too hard on the bias

[Peteck]

Do you have any ideas for a date for when you would release a playable alpha version?

[StormWingDelta]

This should be interesting to see.

[Veritas]

I really dig the concept! Please keep us informed.

[marchred]

Thanks for sharing this project .

Make me curious about you're AI knowledge

[Peteck]

Any updates on this project?

[lolz123]

@Peteck: Nothing too interesting, just some under the hood networking stuff and menu systems.

I have mostly been working on ERL recently, since I was at the AGI-14 conference (artificial general intelligence conference). Hopefully though this project will end up using ERL!

Here is how the options menu looks like: