Tech and Tools
When it comes to animating in the industry, a lot of animators are split between implementing things in engine and creating animations/tools. Before getting into this section, it’s important to note we’re using DCC (Digital Content Creation) as a catch all term for work done outside of the engine.
As to what software DCC covers, Autodesk is still the go to software with most people using Maya (86%), Motionbuilder (39%) and 3Ds Max (18%) with open-source program Blender (9%) coming in a distant fourth place. The remaining 3% are various softwares, like Cascadeur, Akeytsu, and Modo.
For rigs, 53% are Custom Built, 27% Automated/Scripted and 14% are Out of the Box (e.g. HIK, CAT or Biped). The remaining 6% are a mix of options.
For engines being used, we have 41% Unreal. 37% Proprietary engines and 14% Unity. The other 8% use a mix of 3rd party options.
For balance of time spent in engine vs DCC, 47% spend eighty percent of time in DCC, 32% spend half their time in DCC and half in engine, 10% spend all their time in DCC with no in engine work, and the remaining 11% spend most/all of their time in engine.
Breaking down numbers a bit more to see if there’s a difference in time spent in engine for Unreal vs Proprietary, we see there isn’t significant difference between the two, but there are some slight indications of specialized proprietary knowledge pushing more people to the extremes & Unreal’s well documented workflows allowing for a “dabble in engine” approach.
Now we come to the question that kicked off the entire survey! Who is responsible for the implementation of animation?
For this question, we’ve broken down Implementation into three categories: Basic (e.g. exporting, adding to lists/sets, blend times and tagging),
Intermediate (e.g. updating/adding to existing state graphs, blend spaces, additives and layers),
Adavanced (e.g. state graph architecture, visual scripting, procedural systems, ik and ragdolls).
For Basic, it breaks down to 74% animators, 15% tech animators, 6% designers, and 4% programmers.
For Intermediate, it was 55% animators. 28% tech animators. 10% programmers, and 7% designers.
And when it came to Advanced, the percentages are 45% programmers, 38% tech animators, 11% animators, and 6% designers.
So overall, our results show that animators are doing the bulk of implementation up until the needs are more advanced at which point programmers and technical animators take over responsibility with a consistent 6% of designers involved in all levels of implementation.
This breakdown of responsibility was something I also wanted to check for Unreal vs Proprietary engines.
Surprisingly, animators and designers seem to have MORE responsibility in regards to implementation in Proprietary engines over Unreal.
Following up on all of that, we asked respondents to rate how they felt about the statement “animators should be responsible for doing as much implementation as possible” from a scale of 1 (Very Much Agree) to 5 (Very Much Disagree).
As the chart shows, the majority felt animators should have significant to major responsibility in regards to implementation.
One of the factors in that is how user friendly it is to use the implementation tools. In response to that question, we had mixed to slightly positive impressions. In the following chart, 1 means Very User Friendly and 5 means Not User Friendly At All.
Once again, I wanted to check this between Unreal and Proprietary Engines, because animators seemed to have more responsibility in proprietary workflows.
Even though animators are spending MORE time working in proprietary engines, they are reporting those engines to be less user friendly than Unreal. While there is likely a correlation to more users meaning more dissatisfied users hitting against the edges of a tool, the numbers are significant enough to assume factors like easy access to tutorials make it more user friendly than Proprietary engines.
The last bit of tech responses to cover are Machine Learning and Motion Matching. These topics come up a lot when we talk about the future of game animation, so it’s good to get a picture of what everyone’s experience with them looks like.
It turns out still a bit in the future, because out of the respondents 77% are not using motion matching and 87% are not using machine learning.
Since Unreal is the primary engine being used in studios and it doesn’t support either motion matching or machine learning out of the box, I broke it down by Unreal vs. Proprietary again since using these methods require dedicated programming support regardless.
It would be pure speculation on my part as to why these numbers show more than double the adoption rate of these technologies when using proprietary engines, but it was an interesting point of data to find.
The next stop on our survey tour is Team Size!