Ugh, I could never stand LA Noire's facial animation. At best it crosses the line into uncanny valley, at worst it looks as stiff as I'd expect it to be when you stick some actors in a booth after the fact and try to get them to act like they're lying and all of that sort of nonsense. I've honestly never seen anyone actually say it worked for the gameplay frankly, and from the videos I've seen and what I've seen of people playing it I don't see how it would anyway. If anything, being able to read people would be a detriment because some of the animation and acting is so bad.DavidBowieNoReally said:Nope.
This is the best.
The entire tech and gameplay was built around it to make sure not only that it looked good, but it worked for the gameplay aspect (gameplay revolved around reading facial expression and body language to sniff out a lie). If it wasn't top notch, it would be detrimental to the gameplay.
Anyways, I have yet to see anything better.
Not to mention it's one of the most horribly inefficient ways I could imagine doing it. HL2 and all of the other source engine games have had very smooth and natural looking facial animation which has improved with each game, in part because they didn't try to focus so hard on realism that they made something that looked unnatural instead. And the best part is they didn't need to spend more than five years figuring out how to sit actors in a booth and scan in every minute detail, then scale that down to a usable model. Valve's facial animation is largely procedural and built into the engine so it can handle it automatically with a minimum of effort to setup, and a minimum of effort to tweak and perfect. And it doesn't make me feel like I'm looking at some poor imitation of a human being when I see it.