[b]Edit: meant to post in the SC general thread[/b]
wait, what happened to this studio?
[media]https://www.youtube.com/watch?v=JAOhjPtMZdE[/media]
Sounds like a complete waste of funds and development time to make an indepth facial animation system for a game like that.
The stuff I've scene with the facial animations, I've wanted for some time. Also the wing commander games all had focus on the cinematic, even privateer. I haven't played freelancer, but the wing commander games still hold up better then some modern AAA titles in cut scenes and some games have lost the touch of blending the two.
Oh a good game that used the same tech: Until Dawn
and a bad game that used the same tech: The Order: 1886
ones you didn't know: mad max, arkam knight, shadow of mordor, cod advanced, the grey goo trailer, and probably others.
[QUOTE=elixwhitetail;48890717]the MISC Reliant concept art was created by David Hobbins, who's also done some concept art for the new SW films.
[t]https://robertsspaceindustries.com/media/3hqfep3qygo85r/source/Reliant_LandedConfig_Final_Hobbins.png[/t][/QUOTE]
It'd be cool to see some Chris Foss designs in the game:
[t]http://static1.1.sqspcdn.com/static/f/1008242/16457488/1328613927730/XXX_0177_Chris_Foss_Untitled.jpg[/t][t]http://fireballtim.com/wp-content/uploads/2015/02/06-easter-island.jpg[/t]
[QUOTE=Ganerumo;48890961]Sounds like a complete waste of funds and development time to make an indepth facial animation system for a game like that.[/QUOTE]
well, in previous chris roberts space games, your interactions with other characters outside of combat were a pretty important part of the game, so I cant say I'm surprised that they are putting a lot of effort into making things like facial animations look good.
[QUOTE=dai;48888351]What kind of quality do you expect out of keyed animation
Like yeah you can put as much money into a fleet of animators and a year of tweaking things so they're not quite uncanny but at what point does that not sound like a garbage use of time all its own? I'd be amazed if I didn't hear "wow those idiots are wasting so much time when we can just mocap shit nowadays, like when the actors are recording their lines or something!"
Having them actually act instead of read a script into a mic in an enclosed room gets a lot more quality out of their interactions, it sounds pretty efficient to me[/QUOTE]
Scalability, really
Theoretically if you have technology that can do accurate-enough facial animation without needing to mocap actors you can continue adding NPC's (or add a huge number of NPC's) without relying on doing acting sessions for every single one. If this is a kind of game that will be regularly updated years after release that's a pretty good benefit to have.
Also its easier to just re-key the animation to get exactly what you want out of certain scenes when things change or need to be added instead of having to go back out and do another mocap session.
[editline]13th October 2015[/editline]
[QUOTE=KingKombat;48888486]When they break it down, it looks fucking realistic as shit, but as soon as they bring him back into that hall or whatever it's just not realistic at all to me. That one demo with the bald girl blew my away though but that was about it
I'm still waiting for one of these developers to blow my motherfucking mind and reveal that the little "interview" segments were actually computer generated with their engine in like real time or whatever. That'd be a fucking seller to me[/QUOTE]
This is my problem as well, and I don't think its a tech issue
Clearly we have the tech to pull this off well. Hell even games with basic tech can do it well. Pixar for example have been masters at this for years.
The biggest I issue I see in what makes a lot of facial animations feel uncanny/not realistic compared to highly accurate mocap (until dawn) is that people just aren't creating animations that express themselves like normal people. Things can animate realistically all you want, but things just moving realistically isn't what makes it truly believable. What makes an animation truly believable is how realistic an the character expresses the animation. This is why you can get a Mr. Potato Head to act in a way the audience believes - because it expresses itself well in a way that makes sense. Stuff like the guy making a speech in the hall do not in comparison, even though the animation technology is much more advanced than it was in Toy Story.
Little things like what the eyes tell you about a person's mood, what kind of personality type that person has, if they are reclusive, etc. It requires someone with a good idea of what kind of person the model actually is, and then be able to know how to animate their face in a way that accurately shows mannerisms that such a person would do. Basically, it requires the animator to have good acting knowledge and good artistic knowledge of anatomy to animate a face well. Or rather, if its all dynamically generated, the programmer has to be able to code this functionality in the generation. And let's face the facts, most people coding facial animation have technical backgrounds, and most people in technical backgrounds don't really have any acting & anatomy sense to pull it off a truly realistic animation.
The guys programming this system essentially have to be smart actors and smart engineers to make it look truly lifelike in motion, which just isn't likely.
Sorry, you need to Log In to post a reply to this thread.
