MissDeviling
Member
Seriously loving all the wildlife and even mech enemies we've seen so far. They fit so naturally in the world, just like the monsters in XII do.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Final Fantasy XV SIGGRAPH 2016 Tech Talk and Presentations
- Thread starter benzy
- Start date
Demon Lizardman
Banned
Does anyone have a list of all the enemies that have been confirmed and seen so far?
The Last One
Member
Me too. I miss these exaggerated magic effects. I'm currently playing through FF XII and my god the magic effects are so boring. Firaga, Blizzaga and Thundaga look like early spells from ther other FFs.
Blackleg-sanji1
Banned
And to think this is just the beginning we still haven't seen royal magic nor gotten a good look at thunder other than the platinum demo, we only got a glimpse of the magic mixing. Can't wait
The Last One
Member
I can't wait to see Meteor, Flare, Ultima and Holy!
Garrett Hawke
Member
Is there a rehost of the "rendering techniques" vid? Seems to have been deleted
?? Still works for me. The video is unlisted so unless somebody here reported something, it sholdn't have trouble staying up.
Garrett Hawke
Member
Weird, trying to watch it on mobile and it just says "unavailable".
Well, if the youtube link doesn't work then you can watch it here too:
https://www.pathlms.com/siggraph/events/609/video_presentations/31752
CrimsonSquall
Member
Great catch
Maybe it's due to a copyright claim on the video. Like if they showed FFXV trailer footage they might have gotten auto copyright claimed by SE.
The original official Siggraph upload was hit with a copyright claim for Germany viewers. I re-uploaded it unlisted in the OP so I'm not sure why Garrett isn't able to view it. It should tell him if there was a copyright claim but he mentions it just says it's unavailable.
I have the same problem, can't view it on mobile, but it works on desktop
I can view it on mobile in US. I just checked the settings and it appears to have been hit with a copyright claim though. I'll update the OP with the siggraph page link.
NoctisVsStar
Member
Guys... lol
https://mobile.twitter.com/Uxian_/status/758384425289777152/video/1
Bros pushing the car
https://twitter.com/mooglekupocake/status/758387387730948096
https://mobile.twitter.com/Uxian_/status/758384425289777152/video/1
Bros pushing the car
https://twitter.com/mooglekupocake/status/758387387730948096
Aw, that's a cute lil guy. <3
Haha, so cute. Is this from today's talk? EDIT: Ah, no from yesterday when the stream stopped for a few minutes. Now I know why I missed the moving wall thing
That's terrifying, lol.
Lol, looks like Noctis is tired of some little kid following him.
That's terrifying, lol.
Naah, I'd keep it as a pet . . . Whatever that thing is, lol.
Go_Ly_Dow
Member
That's terrifying, lol.
It is. I've always hated these fucks.
That's terrifying, lol.
No, that's not terrifying, that's sooo adorable. Ambitious little Malboro following Noctis around trying to eat him.. awww
That's terrifying, lol.
Its creeping.
That's terrifying, lol.
It's Stella!
NoctisVsStar
Member
Ignis gotta be fired.
https://twitter.com/mooglekupocake/status/758387387730948096
https://twitter.com/mooglekupocake/status/758387387730948096
YankeeDonB
Member
That's terrifying, lol.
Ha ha, looks like Noctis made a new friend!
Seems like it isn't streamed.
We're missing the character building talk from yesterday too.
http://s2016.siggraph.org/talks/sessions/building-character
That's disappointing. Apart from the rendering and physics one from Yesterday it was the one I was most anticipating.
Oh well at least we have the paper that summarizes it.
If we're lucky SE will upload the slides and videos here: http://www.jp.square-enix.com/tech/publications.html
They do that sometimes.
WordsintheWater
Member
So excited for this and it looks great, I just hope and pray the performance holds up, please let it run well
Oh, right. Frogot about that.
Overview of AI/Animation Systems
From a workflow perspective, the skillsets involved in making character behaviors are: animation, game design, and programming. Animators should spend time and effort on the visual quality. Game designers have their strong intention of the gameplay directly leading into the user experience. Programmers are focused on the implementation in way that also allows for scalability and extensibility. From the beginning, we planned to build data-driven systems for the animators and the designers.
Node-based graph systems in the animation area are now widely used in many game engines [Epic Games, Inc. ; NaturalMotion]. However, these graphs grow larger and larger as development progresses and gameplay-specific features stream into the graph. To avoid this common situation, we add another data-driven graph system in between AI and animation encapsulating the complexity of AnimGraphs. The abstract states of actions are managed in this level, which we call Body Graph. We designed it as a beneficial system for programmers to organize characters’ behaviors and movements.
The beauty of this three layer architecture is that we can design tools and specifications on a user-centric basis because it is obvious to define the expected skills and knowledge of the users and tastes of usability for each system.
AI Graph Editor
The users of AI Graph Editor can make graphs composed with hierarchical finite state machines (HFSMs) [Millington and Funge 2009] and behavior trees [Isla 2005]. You can seamlessly assemble those two systems with this novel hybrid architecture to exploit the advantages of each system according to the situations. AI Graph supports the unique concepts of parallel node and interrupt node. Sometimes, natural behavior and thought can be observed as doing multiple things at once: for example, waving a hand during walking, and thinking about the next target while attacking another enemy. We can intuitively implement these behaviors with the nodes.
AnimGraph Editor
The animation system provides a lot of advanced features which contribute to motions adaptive to the environment and help avoid the users feeling a sense of Deja Vu. Various inverse kinematics techniques, physics based animation, secondary bone systems, and parts blending are implemented in the game engine.
Generally speaking, it is strenuous for animators to understand the full functionality of those systems. Moreover, there is the risk of introducing complexity into AnimGraphs due to the complexity of controlling the status and the parameters of the multiple systems concurrently. We added a feature to deal with these issues.
Body Graphs
Body Graphs are usually created with HFSMs. Some of the states in the graphs are driven with procedural systems which require composite character controls such as swimming on the ocean, sliding on a steep slope, mounting onto an animal for transportation, etc. Programmers can implement these in a modular and reusable manner coordinated with programs.
Use Case
In this section, we present a use case of a monster in a battle. First, to increase the reality of monsters, the visual ability of AI is simulated. The target of the enemy is selected from a weighted list of targets based on the visual information. How we select the target is parameterized according to each monster type. This target selection system is registered by placing a node onto the AI Graph of the monster. This target is generally used to decide next attack target. After the decision is done, the AI finds a way to the target. Using a point query system [Jack 2013] and a navigation system [Buckland 2004], the AI determines the optimal path leading to the situation.
Calculated speed and turning ratio are written into the blackboard variables as desired speed and rotation. The AnimGraph in turn uses those values as inputs for motion blending. The three graph systems have a templating system, so the same graph can be used as a module of the character. Specific logic of the template can be overridden with a similar system of inheritance as often seen in programming languages.
Conclusion
The characteristics of the various characters that appear in Final Fantasy XV are designed with these AI and animation systems. The effective toolchain and concept of three data-driven systems; AI Graph, Body Graph, and AnimGraph help to realize the fascinating user experience.
http://dl.acm.org/citation.cfm?id=2927449
From a workflow perspective, the skillsets involved in making character behaviors are: animation, game design, and programming. Animators should spend time and effort on the visual quality. Game designers have their strong intention of the gameplay directly leading into the user experience. Programmers are focused on the implementation in way that also allows for scalability and extensibility. From the beginning, we planned to build data-driven systems for the animators and the designers.
Node-based graph systems in the animation area are now widely used in many game engines [Epic Games, Inc. ; NaturalMotion]. However, these graphs grow larger and larger as development progresses and gameplay-specific features stream into the graph. To avoid this common situation, we add another data-driven graph system in between AI and animation encapsulating the complexity of AnimGraphs. The abstract states of actions are managed in this level, which we call Body Graph. We designed it as a beneficial system for programmers to organize characters’ behaviors and movements.
The beauty of this three layer architecture is that we can design tools and specifications on a user-centric basis because it is obvious to define the expected skills and knowledge of the users and tastes of usability for each system.
AI Graph Editor
The users of AI Graph Editor can make graphs composed with hierarchical finite state machines (HFSMs) [Millington and Funge 2009] and behavior trees [Isla 2005]. You can seamlessly assemble those two systems with this novel hybrid architecture to exploit the advantages of each system according to the situations. AI Graph supports the unique concepts of parallel node and interrupt node. Sometimes, natural behavior and thought can be observed as doing multiple things at once: for example, waving a hand during walking, and thinking about the next target while attacking another enemy. We can intuitively implement these behaviors with the nodes.
AnimGraph Editor
The animation system provides a lot of advanced features which contribute to motions adaptive to the environment and help avoid the users feeling a sense of Deja Vu. Various inverse kinematics techniques, physics based animation, secondary bone systems, and parts blending are implemented in the game engine.
Generally speaking, it is strenuous for animators to understand the full functionality of those systems. Moreover, there is the risk of introducing complexity into AnimGraphs due to the complexity of controlling the status and the parameters of the multiple systems concurrently. We added a feature to deal with these issues.
Body Graphs
Body Graphs are usually created with HFSMs. Some of the states in the graphs are driven with procedural systems which require composite character controls such as swimming on the ocean, sliding on a steep slope, mounting onto an animal for transportation, etc. Programmers can implement these in a modular and reusable manner coordinated with programs.
Use Case
In this section, we present a use case of a monster in a battle. First, to increase the reality of monsters, the visual ability of AI is simulated. The target of the enemy is selected from a weighted list of targets based on the visual information. How we select the target is parameterized according to each monster type. This target selection system is registered by placing a node onto the AI Graph of the monster. This target is generally used to decide next attack target. After the decision is done, the AI finds a way to the target. Using a point query system [Jack 2013] and a navigation system [Buckland 2004], the AI determines the optimal path leading to the situation.
Calculated speed and turning ratio are written into the blackboard variables as desired speed and rotation. The AnimGraph in turn uses those values as inputs for motion blending. The three graph systems have a templating system, so the same graph can be used as a module of the character. Specific logic of the template can be overridden with a similar system of inheritance as often seen in programming languages.
Conclusion
The characteristics of the various characters that appear in Final Fantasy XV are designed with these AI and animation systems. The effective toolchain and concept of three data-driven systems; AI Graph, Body Graph, and AnimGraph help to realize the fascinating user experience.
http://dl.acm.org/citation.cfm?id=2927449
Blackleg-sanji1
Banned
Lmao wth prompto
Lmao wth prompto
As ridiculous as his battle style is, it's an insane technical achievement that they're able to blend all those animations so well. In most action games it looks like a bunch of characters playing animations at each other. But in this completely dynamic shot, it looks almost like someone purposefully animated this scene to work this way.
Like, if Prompto was in a fight scene in a cutscene or CG movie, I'd expect the fight to be choreographed like this(more polished of course, but this is damn close).
When it works well, it's really impressive, yeah. But of course there will be moments of jank and jitter when too many characters are colliding and bumping into each other during combat etc. Nonetheless, extremely cool things they're doing with dynamic interactions.
The two devs yesterday even said Prompto was especially hard to implement because of his constant and hectic movement.
Lmao wth prompto
Those soldiers must be annoyed to just keep on missing him like that.
Seriously, compare Ellie running around in The Last Of Us compared to Prompto, who intelligently moves around the environment and manages to dodge stuff like he actually should. And that game had one AI character for a very large majority of the game.
Obviously the combat scenarios are different and TLOU is more stealth based. It would likely have been a different approach altogether, but you can see the effort here and the difference that it makes.
YankeeDonB
Member
His technique here is like a mix of capoeira and three stooges. And yeah, animation blending is darn impressive.
Blackleg-sanji1
Banned
Like they nailed blending his animations with his character. We know prompto is alil inexperienced compared to the others so it's hilarious seeing him try his hardest
Lmao wth prompto
Insane animations. They look great.
Lmao wth prompto
Going for that groin critical
Lmao wth prompto
At the beginning of the gif, that one robot dodged the attack from another robot. Really impressive.