Inverse kinematics is used but, again, that does not solve the problem of the trade off between response time and realistic movement. It simply "snaps" the feet to the ground.

A lot of the problems occur during movement, when responsiveness doesn't even matter. I still have yet to see a game where it's impossible to get randomly stuck at a fence or rock that instantly stops one's movement for example. Making the character dynamically avoid such minor obstacles would make the game more responsive, not less.

And games with fancy animations are often not just less responsive, but also less controllable in general: in Witcher 3 the character cannot walk backwards or turn in place, so you can literally fall off a cliff in front of you by trying to walk away from it. In terms of movement the only difference between Geralt and his horse is the turning radius.

I don't play any video games but it seems like you might be discussing a separate issue. A developer must choose between responsiveness and fluidity.

Responsiveness doesn't play a role in what I'm talking about. And if Boston Dynamics can integrate it into real-life robots to some degree then it can also be done in games.

Conflicting requirements stemming from the psychological illusion of movement spontaniety.

People don't want the realistic, slow response. Normally people aren't aware of how early their body starts a movement, the conscious brain has the illusion of just having decided it but it's actually started much earlier.

> People don't want the realistic, slow response.

That's it, there's the tradeoff; if they want legs to move naturally, they have to fight against what the user inputs. There's a setting in the Witcher 3 so that the character can turn on a dime / instantly, less realistic but more fun / responsive: https://www.escapistmagazine.com/difference-between-standard...

Why doesn't the same apply to a joystick though?