Naughty Dog’s Animator Jonathan Cooper posted a long explanation on his Twitter account of how animation works in AAA games, providing a lot of insight into what might have happened with Bioware’s Mass Effect: Andromeda. We’ve gathered it all here for easier reading.

Folks have been asking so here are my thoughts on Mass Effect Andromeda’s animation. Hopefully people will better understand the process.

Animating an RPG is a really, really big undertaking – completely different from a game like Uncharted so comparisons are unfair. Every encounter in Uncharted is unique & highly controlled because we create highly-authored ‘wide’ linear stories with bespoke animations.

Conversely, RPGs offer a magnitude more volume of content and importantly, player/story choice. It’s simply a quantity vs quality tradeoff. In Mass Effect 1 we had over 8 hrs of facial performance. In Horizon Zero Dawn they had around 15. Player expectations have only grown.

As such, designers (not animators) sequence pre-created animations together – like DJs with samples and tracks. Here is the Frostbite cinematic conversation tool circa Dragon Age Inquisition; here’s the cinematic conversation tool for the Witcher 3. Both tools make it fast to assemble from a pool of animations.

Because time denotes not every scene is equally possible, dialogues are separated into tiered quality levels based on importance/likelihood. The lowest quality scenes may not even be touched by hand. To cover this, an algorithm is used to generate a baseline quality sequence.

Mass Effect 1-3 populated default body ‘talking’ movement, lip-sync and head movement based on the dialogue text. The Witcher 3 added to this with randomly selected body gestures that could be regenerated to get better results.

Andromeda seems to have lowered the quality of its base algorithm, resulting in the ‘My face is tired’ meme featuring nothing but lip-sync. This, presumably, was because they planned to hit every line by hand. But a 5-year dev cycle shows they underestimated this task (all this is exacerbated by us living in an era of share buttons and youtube, getting the lowest quality out to the widest audience.)

Were I to design a conversation system now, I’d push for a workflow based on fast and accessible face & body capture rather than algorithms. While it hasn’t 100% proved this method, Horizon Zero Dawn’s better scenes succeed due to a use of facial mocap.

The one positive to come out of all this is that AAA story-heavy games can’t skimp on the animation quality with a systemic approach alone. The audience has grown more discerning, which makes our job more difficult but furthers animation quality (and animators) as a requirement.

http://wccftech.com/naughty-dog-dev-explains-andromeda-animations/?utm_source=wccftech&utm_medium=related

Interesting to hear from someone who knows the business but isn't connected to this project.

So in Mass Effect 1-3 they had defaulted lip sync and facial animations based on the dialogue and in Andromeda the quality of this dialogue to default lip sync and facial animations were lowered? Interesting.

It's kinda sad when Ryo in Shenmue from 1999 has better lipsync. Not even joking. Ryo's is pretty good. First game I even saw with lipsync.

Still, didn't a red light go off during development meetings/playtesting? This could have been totally avoided. No excuses really. Especially from a development house this large and wealthy.

Well, they lowered it because, in his opinion, they would fix it later by hand. What we should ask is why they didn't do it (did they ran out of time or was it because of budget reasons?), and if there is a way to fix it once the game has been launched.

In any case, like he mentions at the end, this will only serve to increase the quality of animations from now on.

Naughty Dog be like:

"They used a bad methode, and underestimated how much work it would be to do it the way they wanted, so in the end they where forced to do a lazy quick fix. Bad planning. However dont attack them guys, makeing games is hard."

Not exactly.  What he is saying is in the old games, they had base lipsyncing and then an algorithm went in there and assigned various expressions and gesture animations based on the dialogue (I assume the dialogue system had a sort of tag system where you could mark a line as "sarcastic" or "shy" and the algorith searched for said tag).  They then went back and added further nuance by hand only in the most important instances.

Andromeda seems to have dropped the algorithm step almost entirely with the assumed intention of going back and hand-nuancing all the lines.  But part way through they realized all too late what a terrible mistake that was.  The facial animations are thus, truly and literally unfinished.

Still, this doesn't explain the anatomical errors in facial structures, weird proportions for certain characters, and really stiff walking, running, and idle animations.

Hes basically saying the reason it is like it is, was because they where forced into it.

They didnt have the time & money to do it the way they wanted, so they had to do a "hotfix" to the mess, so there was any animation to the voice acting at all.

Its bad planning. You can bet your arse it showed up in meetings & playtestings, but at that point people where like "what you gonna do? This is as good as it gets with the deadline we have folks".

"It’s simply a quantity vs quality tradeoff"
What a cop out, rather have both.
Unfortunately, Bioware fails on both.