After a few years of waiting, Wildfire Games have released a big upgrade for the free and open source real-time strategy game 0 A.D. with Alpha 24 now available. Still one of the most impressive open source games around, their attention to detail on it is impressive.
This latest version is named Xšayāršā, after Xerxes the Great, ruler of the Achaemenid Empire from 485 to 465 BC.
Quite a feature-filled upgrade with highlights including building snapping, improvements to the renderer, a hotkey editor, unit formation improvements, status effects, improvements to the online Lobby system, new models, new skirmish maps, unit behaviour improvements and more. Release trailer is below:
One of the features that makes me most curious about this release is the inclusion of a "Reinforcement-learning interface", which they say can use Machine Learning to teach the AI how to play. Sounds very interesting and the team is rather excited to see it included as it allows for the setup of more complex environments without having to recode everything. Looks like we might see some fun AI to fight against in future.
Quite a lot of rendering improvements were made too, along with bumping up the lowest possible OpenGL requirement to run it to 2.0 so it should still work on pretty old GPUs. With this they added in anti-aliasing support with FXAA and MSAA now possible plus there's also a Contrast Adaptive Sharpening filter now available.
Now that this long development cycle is over, which included an upgrade to some major internals, they're hoping to be a lot more regular with releases.
The full list of changes can be found in the release post.
Find out more on the official site.