This release contains the following changes
- Rebuild of ML training solution libraries.
- Full retraining of AI for scenario 1 (standard only - not with random VPs or Reinforcement rule)
- ML AI is now used as standard where relevant for scenario 1 - menu option removed.
- Vsync menu option added to settings
- Counter size menu option added to settings. Default is "Auto" - where a tile contains 10 counters or fewer then you see the large counter, otherwise you get the small counter. Options for always large and always small also provided.
- Unity version upgrade
- Title page visual changes
With regards to the roadmap, now that the ML training envirinment is a lot more stable, the plan is to do more training for scenario 1, and also develop the environment so that it can do training for all other scenarios.