With all the talk of FMV around Project Fedora, I was wondering how the team would present in-world characters. In UAKM and Pandora they were essentially animated cut-outs which worked well in the 90s era alongside Doom etc, however nowadays there's a risk it could look out of place.
I can see a few options...
UAKM/Pandora - flat FMV cutouts. Maintains sense of consistency between gameplay and FMV but with issues stated above
Overseer - no characters except in conversations/cutscenes. Removes the issue but makes the world feel a little "lonely"
3D-modelled characters - meshes with photographic texture work and motion capture animations to make them as similar as possible to the FMV actors.
This is all assuming an explorable 3D environment of course, but who knows what could happen?
I can see a few options...
UAKM/Pandora - flat FMV cutouts. Maintains sense of consistency between gameplay and FMV but with issues stated above
Overseer - no characters except in conversations/cutscenes. Removes the issue but makes the world feel a little "lonely"
3D-modelled characters - meshes with photographic texture work and motion capture animations to make them as similar as possible to the FMV actors.
This is all assuming an explorable 3D environment of course, but who knows what could happen?
Autodesk Photofly http://www.youtube.com/watch?v=9cb9Suu8w3Y
+ Rigging by hand in Maya (takes a bit of time)
+ Microsoft Kinect with FaceMoCap software (something like http://www.youtube.com/watch?v=iS8A9oXvV4I)
+ 3d props in Real World Environment = Autodesk Match Mover
= 3d Actors/LA Noire
At least that's how I'd do it. Its a much cheaper solution than doing it the professional way.
+ Rigging by hand in Maya (takes a bit of time)
+ Microsoft Kinect with FaceMoCap software (something like http://www.youtube.com/watch?v=iS8A9oXvV4I)
+ 3d props in Real World Environment = Autodesk Match Mover
= 3d Actors/LA Noire
At least that's how I'd do it. Its a much cheaper solution than doing it the professional way.
we would try to design the game so everything stays "in engine", meaning that instead of shooting characters on blue screen, we would try to computer-generate the characters and map them with the actors' faces and bodies. If we could pull this off effectively, it would open new possibilities we couldn't dream of before (one thing would be the ability to "populate" our VR environments--up until now we've had to keep Tex's physical world fairly empty)
I thought the desolate almost-empty to empty areas were kind of appropriate considering the time period the games are set in. I wouldn't want to see a busy Chandler Avenue considering the themes surrounding the area. Other times you're mostly investigating areas you shouldn't be or at night so there wasn't much need for people to be around. I never felt Tex's world felt lonely, just desolate. But I felt that enhanced the post-apocalyptic setting and how Tex often had to tread where no-one else would. Some places need more people though true, but I think some extras such as for the Flamingo bar in past games worked OK and with the Kickstarter some may bid for a tier to be an extra. I think green screen tech probably has come further too.
I'm wary of 3D models with real faces mapped to them... Makes me think of parody series where they map real mouths and eyes to objects or put real heads to digital bodies. But its true things have come a long way with the right texturing and such. But it's still tricky to get 3D people looking truly lifelike I feel even today, so it may depend on the budget. Heavy Rain and some other games do look awesome and, though still not fully "real", if it comes down to it being 3D models then if they can do stuff on that level it could work.
As 'voice just' posted though the 3D people route is something they're heavily considering though, so this post is more just my personal opinion. More info on how they think they'll pull it off might be something for the Kickstarter page maybe.
On the topic of 3D models, I wonder how far they'll be able to push the 3D environments now. I have less concerns about that aspect. Stuff like Crysis, though I haven't played it myself, and what the engine can do looks awesome:
http://forum.i3d.net/attachments/main-c ... t_set3.jpg
And it seems like the engine (CryEngine) is available for others to use since I see people on Deviantart are playing with it for their portfolios etc.:
http://virtualartsca.deviantart.com/art ... gine&qo=10
I'm wary of 3D models with real faces mapped to them... Makes me think of parody series where they map real mouths and eyes to objects or put real heads to digital bodies. But its true things have come a long way with the right texturing and such. But it's still tricky to get 3D people looking truly lifelike I feel even today, so it may depend on the budget. Heavy Rain and some other games do look awesome and, though still not fully "real", if it comes down to it being 3D models then if they can do stuff on that level it could work.
As 'voice just' posted though the 3D people route is something they're heavily considering though, so this post is more just my personal opinion. More info on how they think they'll pull it off might be something for the Kickstarter page maybe.
On the topic of 3D models, I wonder how far they'll be able to push the 3D environments now. I have less concerns about that aspect. Stuff like Crysis, though I haven't played it myself, and what the engine can do looks awesome:
http://forum.i3d.net/attachments/main-c ... t_set3.jpg
And it seems like the engine (CryEngine) is available for others to use since I see people on Deviantart are playing with it for their portfolios etc.:
http://virtualartsca.deviantart.com/art ... gine&qo=10
(Ruri_Ayanami from the old Tex Murphy ezboard).
"I don't believe in intuition, don't know why... just a feeling." - Tex Murphy
"I don't believe in intuition, don't know why... just a feeling." - Tex Murphy
Me and a friend were discussing this.
you might want to check this out. It's called light staging where they get a real actor, and light them in thousand of different increments, which are manipulatable in post, so if you were to place the real actors in 3C environments they can be lit according to their CGI surroundings.
Might be exactly what the guys are looking for:
Capture of real characters for dynamic FMV: http://gl.ict.usc.edu/Research/LS5/
And capture of walking motion for dynamic FMV integration into a CGI world: http://gl.ict.usc.edu/Research/RHL/
-Cub. =o)
you might want to check this out. It's called light staging where they get a real actor, and light them in thousand of different increments, which are manipulatable in post, so if you were to place the real actors in 3C environments they can be lit according to their CGI surroundings.
Might be exactly what the guys are looking for:
Capture of real characters for dynamic FMV: http://gl.ict.usc.edu/Research/LS5/
And capture of walking motion for dynamic FMV integration into a CGI world: http://gl.ict.usc.edu/Research/RHL/
-Cub. =o)