Because it’s hard to animate. There are 650 muscles in the human body and 187 joints. Each muscle and joint they attempt to animate adds complexity to the model and increases the likelihood they’ll get something wrong if they’re animating by hand.
Most use motion capture these days but again they can only choose so many points on the body to track, they don’t capture the entire human form. They use a suit with special dots or balls on it that special cameras can pick up and track in space and then translate that into an animation program. The earlier you go back the fewer points of motion cameras could track at one time.
Put simply, technology got better.
Our ability to motion track movement performance is much higher resolution than in the past, we have the storage capacity to actually contain all that data and we have the graphical horsepower to render that and everything else in sufficient quality to make it lifelike,
Its also partly an arms race kind of deal as well. In addition to the technology the budgets for high-profile releases (think Red Dead, Last of Us II primarily) have also skyrocketed, so developers can put more time and resources into body rigs, animation and the systems that transition between all various movements that they do.
EDIT: [This video is a pretty good overview of the animation systems that go into video games](https://www.youtube.com/watch?v=NmLPpcVQFJM). Its framed in a perspective of explaining what might have went wrong with Mass Effect Andromeda’s animations at release, but it does cover a lot of the ground your question is based on.
Well… It is unreal. There is a limit of what you can do.
The basic concept is the following: To make movement, you need to give it a way to happens. So for example, if you want to allow your character to move forward, you need to map a key to go forward. Wanna run? another key. As a result, the number of movement you can put is limited. Even worse, the amount of effort a player is willing to spend learning is even more limited. So you are already limited in the number of actions you can prepare.
So you need to make a set of movement and players will be restricted by it. So you have a set of walking, a set of running, a set of turning, etc. Animations will never look smooth because you’ll still only have 1 set. Which mean that it’ll always repeat.
The second reason is that you have to sacrifice a few things for game. You don’t want your character to take 2 meters to slow down after stopping your movement. You don’t want your character to refuse moving because it start an idle where it comb its hairs. Gameplay has to take priority over realism. So a lot of details get sacrificed for the game.
Latest Answers