[ame]
https://www.youtube.com/watch?v=YaBbiXEUwIU[/ame]
Interesting technology. The ''node graph'' and scripting behind this seems pretty well worked to me.
Even though, there are still alot of improvements to do on the blending and some movements are completely off, it's nice to see anim tech like these progressing the development of in-game animations in video game.
Personally, I think responsive and good animations are way more important than the actual visual art in terms of gameplay (yeh, I admit being a little biased saying that
).
It's always nice to see a responsive character first, but then to see him move smoothly with beautiful motions and not with cut and broken manners.
I used to hate on Mo-Cap but after seeing this
[ame]
https://www.youtube.com/watch?v=t6q80hYy7sk[/ame]
I not only have to applaud RAD studios but thanking them for making me appreciate how Mo-Cap can be amazing for achieving grounded animations ( it has to be well done of course). I'm amazed here at how they managed to gauge and play and cope up with new methods to really make the character motion convincing realistically.
Most importantly, it's overall pretty smooth, even transitions feels like the character has weight and is balanced.
Quick output and action responsiveness is another conflict though and that may irritate some player.
I would be interested to know, what do you think will be the next big step in terms of in-game animation?
Do you prefer Mo-Cap or Hand-Keyed?
I'm sincerely looking forward to work in the game industry to see all the ongoing constraints animators have because I really feel like it's a domain that needs much more exploration...
Replies
Search Bungie and Ubisoft Run Time Rig online
Mocap? yeah, hate it or love it, can't live without it. Just be flexible.
real-time lipsync still has a long, long way to go though... the FF XV lipsync looks only a little improved from fallout 3.
I prefer keyed animation just because it's controlled and clean, no need to deal with errors or cleanup, though the only Mocap I've worked with is Kinect ( it wasn't terrible) I'm sure it's probably similar to the 3D scanning vs Sculpting from scratch
But like most things, every technique has it's place.
A good step would be proper world interaction and mixing physics with animations (player gets shot, they feel the impact and react). GTA 4 did this really well but I haven't really seen it done that much more recently. Though Battlefront apparently does this really well as well.
I agree about animation, I think it gets left to the way side a little bit when it comes to focus and improvements despite it having the strongest impact on game feel and gameplay than any other part of the 3D pipeline. I think sound is in the same boat as well.
[ame]https://youtube.com/watch?v=Xsq-HxUdhKQ[/ame]
[ame]https://youtube.com/watch?v=gK0xWq9qMgU[/ame]
Off-topic, an animation detail that's stuck in my head is from Conker's Bad Fur Day for the N64. When you're orbiting the camera around and you get to the front of Conker, his eyes start following you around.
It's a texture effect, sliding the texture by changing the UV coordinates of the eyes based on the view direction:
[ame]https://youtube.com/watch?v=gUxlKO-Ie0c[/ame]
I'm sure a real animator could take a deep dive into the pros and cons of this idea.
Also cost is a huge factor. You see it in a lot of indie and narrative games (Gone Home, Dear Esther, Everybody's Gone to the Rapture.). Character and GREAT animation that is needed is expensive. Some games like, The Vanishing of Ethan Carter tried it, but imho, failed to reach the bar set by AAA games in terms of character performance. At the end of the day, whatever gets the best performance is what is key.