In honor of recently rewatching this movie, a thread of Iron Man (2008) references that have better usability for AR than the ridiculous heads-up display:

Integration between existing computer space into world space: one of the most under-appreciated constructs right now, it showcases a really great example of how humans don't always want to work in 3D space and like existing infrastructures.

Manipulation of object-based UI, skeuomorphic design, UI platform gizmo for rotation: showcases the minor manipulations of objects and the importance of highlighting for reference (although some text would be nice)

Multi-modal input integration: showcases the ease-of-use between using a pen as a form of input to interacting with real-world scale objects then transitioning to hand input

This one is less desirable than the immediate trashcan, but gamification and aim assist is always a majorly enjoyable component to any UI!

The suit isn't all bad though: semantic environmental understanding! We love when a meshing/mapping system does the work for us.

Speaking to semantic environmental understanding: we also have examples of horizon (floor? lol) detection for understanding user distance from the ground and also contextual references to landscapes and architecture! Wonder if they got an API from Google Maps or Niantic??

All this to say: I've ragged on Hollywood before for having poor UI and usability for AR reference - but I've come to eat my words and say to you: It's not all bad! Just choose your reference wisely!

Good tips:

1. Don't put things that close to a user's head
2. Put contextual UI near the physical or digital object it references
3. Design for multiple recognizable inputs (if you can)
4. UI affordances (table or wall) for where/how digital content can be manipulated