Good analysis, though I am a little put out that his ideal solution is more mocap than systematic technology. I mean, overall, it sounds great; more "real animation"! In practice, though, that adds a lot to the production (costly tech, application time, recording space increase needs, etc) and also changes the performance (actors wearing camera helmets do not act like actors projecting their whole being into a single microphone before them.) The simple fact that a voice actor can hold a script and look off it all day long while a mocap actor must memorize lines and deliver them straight on to camera (unless they have some kind of telestrator mount for mocap? but even that adds eye/muscle movements) is a night and day difference, and that's just over a piece of paper. VO is a honed skill, and performance-capture acting has made great strides but it will always be different and will always depend on the situation if it's the best solution.
So I keep hoping that technology and procedural adjustments (the randomized gestures in Witcher 3 sounds genius) and blending of animation techniques in a scene will help animators get the job done easier and better. Sounds like the pro here doesn't trust it to get there, which is unfortunate. On the other hand, if he believes mocap technology will continue to get better and easier to use, that's a win in that direction. You'll never get better than longingly handled, hand-adjusted, fully managed animation, but we're going to need tech on one side or the other in order to cope with the demands and costs of modern games.
