Animation & Mass Effect Andromeda

A Very General Overview

Congratulations! You’ve survived another unbelievable week in the world of gaming.

With Horizon Zero Dawn, Breath of the Wild, and Nier Automata, March has felt like an overwhelmingly positive time for the games industry, but the last few days have seen that good sentiment tanked by outcry over the upcoming launch of Mass Effect Andromeda.

If you’ve somehow missed it here’s the brief: Mass Effect Andromeda, a new stand alone addition to the Mass Effect series, launches later this week in the US and European markets. Review copies are already out there, along with roughly ten hours of game that’s been made available for subscribers to EA Access. However, the discussion around the game has become almost exclusively centred on the quality of the character animation, both facial and full-body. It’s now easy to go online and find footage of goofy walks, derpy faces, and unnervingly twitchy eyeballs. The backlash against the studio Bioware has been quick and severe, with many gamers outraged that a AAA game can have such poor quality work, a situation exacerbated when some of the developers came out and said there were no plans to improve the animation with a launch day patch, and being vague about them ever being corrected saying: “that ship has sailed”.

Unfortunately for some it wasn’t enough to direct their anger at Bioware as a company and  they needed to find an individual. Of course the person they picked was a woman, an animator and former cosplayer, and of course vitriol and death threats followed shortly after. This has prompted Bioware to issue a general statement decrying threats against their employees and promising to support the staff affected.

Watching a lot of this unfold was particularly interesting for me. I’m fairly vocal about being a gamer but increasingly quiet about my once-life as an animation student. Seeing a lot of assumption and misinformation bouncing around I wanted to step in and try and clear up a few things – very generally as I was not involved in this particular project – but hopefully try and give a better understanding of how some of these things work and where they can go wrong.

  • A Job Title

Just a quick one first up, coming from my other former life in Human Resources. The particular employee that was singled out as responsible for the animation was singled out because she listed her title as Lead Facial Animator, leading a lot of people to believe she was the head of all animation, or that she alone was responsible for all the unintentionally creepy faces.

Job titles are bad and generally not very descriptive. From the perspective of an employer “Lead Facial Animator” can mean anything from overseeing all facial animation output from the team to being one of four or five other “Lead” facial animators. Lead might even mean nothing more than “not a junior animator”, and might simply indicate a slight difference in pay rate based on relative experience. Not their influence over the project as a whole. And sometimes employers just give you nice and important sounding titles because it sounds better when you introduce yourself to visitors. Ignore titles, ask people to describe to you what their role actually is.


  • Motion Capture

Mo-cap (or motion capture) is tricky as there have been huge advances in technology over the years – but it’s not the instant-animation solution some people think. Firstly mo-cap equipment is expensive, it takes time to record all the actions you need, and facial motion capture is still quite unpredictable. The quality of the animation data you get out of mo-cap is connected to the quality of the equipment you record with and the skill of the actors you have performing (and not just performing in general but performing for mo-cap).

Once capture is done the data is brought into animation/modelling software and is little more than a swarm of reference points. While you might have recorded someone waving their hand – and the animation data will reflect that wave – it doesn’t know what a hand is or where it is. You must assign each point of the data to the relevant part of a rigged character model and that can be extremely intensive depending on the level of detail and the amount of data. Then once everything’s connected you still might find that something doesn’t look right, that for some reason when the model moves its shoulder in that particular way it makes the hand disappear inside the body. There is often a need to go back in and manually smooth out problematic parts of animation when going from human motion capture to human character model, and the Mass Effect universe involves creatures for whom human motion-capture is not going to be enough.

All of this takes more time and more work. Unless you’re working on something with the scope of Cameron’s “Avatar” – bearing in mind even a 3 hour film doesn’t equate to the amount of content required for a 25-30 hour game – with the work and expense involved in motion capture it’s often out of the budget to use it for more than the most important characters. Mass Effect Andromeda is a galaxy of new worlds to explore and filled with people to speak with, but I can’t see there being money for giving this sort of treatment to anyone other than the player character and perhaps the five or six other characters that make up your main squad. That makes for hundreds of less significant characters who are going to need a different approach entirely.


  • Character Animation

So if we can’t afford to use motion capture for the majority of the character models we’re going to have to do it the old fashioned digital way (just like Disney used to make ‘em ;)), which is a person sat at a computer likely going through the process of blocking and inbetweening.

Blocking is the process of finding the big, most important poses in the animation sequence – the arms thrown up in the air in a cheer, the stance of the character mid-walk. Once these blocked points are considered evocative enough of the movement you want to create you can then do the more fiddly part of working out what frames you need to create inbetween to get you to them. The end result – a nice smooth animation of someone walking (hopefully in a way that conveys their personality) or throwing their arms in the air. But it is time consuming and there’s no way of skipping any part of it and keeping the quality. If you block badly what you inbetween will still be weak, if you block well and then fudge the inbetweens it will still look odd.

Until you animate something you don’t really notice all the intricacies of movement, the parts of you you don’t realise you lower when you reach others up, the ones that take your weight when you lean, it’s really fascinating but also time-consuming and easy to get wrong even with basic movements. The number of different actions that might need to be performed by npcs in a space action rpg? Hundreds, probably thousands, and not just humans but other alien races, alien monsters/animals that might also exist in the environments, creating whole new rules of movement for alien species – it’s a monumental task for any team, no matter how experienced.


So somewhere in the animation process something goes wrong. Perhaps the animation was done to a lower standard, maybe they weren’t able to clean up motion-capture data well enough for it to look right when applied to their character models, or there’s the possibility that the animation data once applied to a rigged character model just malfunctioned. It happens. You can take solid animation and a well-rigged model (a rig being the skeleton that informs the model which parts can move) and bring the two together and it goes as well as oil on water. The whole process is extremely reliant on multiple pieces of software working together and sometimes they just don’t. There are many places it can go wrong, or become distorted, even with the best team. When you factor in potential human error as well it’s not really a surprise when it does. It’s just that we normally don’t see that stuff, normally it’s caught in testing or by whoever’s overseeing the animation team. Maybe they need to just tweak something small to get it working, maybe they need to scrap that whole action and start over. We just usually don’t get to see it.

If you want my theories for what happened – and they are only theories – here’s what strikes me as likely causes.


  • An Understaffed Team

This feels a strange thing to accuse a company with the renown of Bioware, but it can happen to anyone. The more ambitious the project the more people you need working on it, and Mass Effect Andromeda is ambitious. But game development is an expensive business with a budget that needs to be spread between multiple departments. It’s certainly not unusual for a company to chose to invest more in one area to the detriment of another, so it’s entirely possible somehow the animation team ended up with less funding for the animators required to take on this volume of work. That isn’t a slight against anyone who did work on it but one person organising books will manage less in the same time as two or three. Sometimes you just can’t make up for manpower.


  • An Unrealistic Deadline

Mass Effect Andromeda was announced on June 15th 2015. A tentative release window was given as late 2016 but the release date was then announced as March 21st for the US, and March 23rd for Europe. This date has not changed. For the majority of the time between the announcement and release there wasn’t a great deal of information given about the game, and there was a slight sentiment that the lack of information might mean development wasn’t as far ahead as we were lead to believe. Others fully expected the release date to slip, as has happened to several other games recently – but it didn’t.

My worry is that the release date was always inflexible, and that a game that could benefit from another month or two to smooth out issues was simply denied them. Again, even the most talented staff can only do so much with the hours they have, and if the pressure was on to produce more work in less time then it seems obvious that it could lead to a drop in quality. As much backlash as delays get I think enough games have shown effective use of additional time in recent years, and that we’ve had some more polished games as a result. March was already a seriously overcrowded month, with multiple strong titles and a console launch, would it really have put them in such a bad position to take another month?

In the end all of this is just my attempt to try and explain a process that – when done right – looks almost effortless, but which can suffer for a multitude of reasons. No single animator can tank an entire project – and if a project of this magnitude had had only a single animator then it would still not be their fault. This is a project that requires a large and talented team, working solidly for months if not years, and any decision otherwise is no fault of the animators themselves but the people organising the project as a whole. For what it’s worth I hope we will see some fixes in later patches, or that the rest of the game will be engaging enough to help players overlook its flaws (looking at you Skyrim), but none of this is worth turning on an individual or descending into threatening or harassing behaviour.

If this incident has burned you on buying games then I’m sorry, hold on to your pennies and wait for reviews first in future. If it still really stings then I recommend you get yourself on a computer and start looking at how to fix things yourself. Study animation, learn programming. Be the change you want to see, no one knows better the ideas in your head than you and it’s a far better way to use your energy.

Go forth and game.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s