EA is considering implementing Dynamic Difficulty Adjustment in its games. This isn’t a first for adaptive difficulty in games, but a recent patent shows precisely what the games publisher is thinking, and it is considering using AI/ML methods to fine tune game difficulty to maximise player engagement and enjoyment.
The patent document, filed in 2020 but published late last month, identifies two main reasons it is looking closely at game difficulty settings. First, software developers wish for gamers to play their games for as long as possible, and this length of engagement is a good measure of the success of the software in being a source of entertainment. Secondly, games that are pitched at too easy or too difficult levels will mean less play time from most users.
Thus, EA is looking at “automatic granular difficulty adjustment” in games it publishes. It foresees a system where the adjustments are transparent to the user. Furthermore, it proposes being able to adjust difficulty using an ‘interactive computing system’ in the cloud, based upon:
- User progress speed in the current game vs the expected duration of gameplay segments (your skill level).
- User cluster categorisation compared to other players of the game.
- User history in a range of games that use this Dynamic Difficulty Adjustment system.
As mentioned in the intro, dynamic difficulty settings in games are not new but are relatively rare. Left 4 Dead and Resident Evil 4 have used such adjustments to maintain a challenging game environment before. However, EA’s system seems to be one of the first to use the power of cloud computing and AI/ML technology for granular difficulty adjustment.
Personally, I like games to be on the easy side, so would still hope for some kind of manual slider difficulty adjustment. Part of the lure of many video games is your character’s /vehicle’s unrealistic levels of power and special abilities.
Be the first to comment