Reputation: 101
Sometimes I read that using Windows ML and/or DirectML can improve the behavior of NPCs in games. It seems that both APIs are suitable, but which one fits better? Or is parallel use for different tasks the best way? If one of them is better for Machine Learning based NPCs, why is it better for this than the other API?
Upvotes: 0
Views: 136
Reputation: 80
I don't think this question can be answered for all cases since it largely depends on what your goals are and what your machine learning model is doing. I'd suggest reading more docs like Extending the Reach of Windows ML and DirectML and Is DirectML appropriate for my project?. Also, note that Windows ML uses DirectML. They work together, but you can choose which one to use based on how much control of the pipeline you need.
Upvotes: 2