You're right. Watching the Nvidia one now. EDIT: The Nvidia one was definitely less creepy but I wonder if it can really be used to say anything. It was a mor

You're right. Watching the Nvidia one now. EDIT: The Nvidia one was definitely less creepy but I wonder if it can really be used to say anything. It was a more limited demo. But it did seem to support my hypothesis that decoupling the eyes, etc. was a mistake because that one is using whole facial expressions together. Or at least that's what I gathered.