Animation Status Update (Q1 2024)

Unity Engine
,
1

Blimey, quite a lot has happened at Unity since my last post .

Animation Today
I will start with animation, as you know, it’s the title of this forum.
Since my last update, sorry to say there really is not much more to report as of today, in terms of what I can share and when we might see it. Just that we are still hard at work on the new system, but sadly did lose a dev as part of the recent layoffs. So, what do all the recent changes mean for the new animation system? Before I dig into that, I want to give a little insight for context.

Organization Changes
Our long time CEO John Riccitiello retired in October. Since joining Unity, he oversaw huge changes and growth at the company, especially over the last few years since covid hit. However, after incredible growth, came the massive downturn in the global economy, which meant many of our recent expansions got hit hard with the inevitable restructuring. During his time, he oversaw a very large portfolio, but always had games as his first love. That love not only continues with our new temporary CEO, Jim Whitehurst, but since his joining it has become a renewed focus for the whole company.

Layoffs
Jim stepped into a demanding situation, but together with the board, took on the challenge to make us very well prepared to deliver in the future. The games industry has seen maybe 20,000 layoffs since the beginning of last year, so it has been tough for a number of reasons, and Unity was not immune to this either. It is very hard to see any friends and colleagues go, and whether we like it or not, it has an indirect impact on productivity for those who remain, as we are human after all.

Our Core Technology
Now, analysts do “predict” the industry will bounce back next year, so as a company we absolutely need to be prepared if that transpires. This means we should focus on what we do well, which is real-time 3D. And while we know a bulk of it is games, RT3D is everywhere and growing, including automotive, manufacturing, architecture, and even retail. So with the renewed focus I mentioned earlier, we are investing heavily to make sure more people and companies are successful making games and real-time experiences.

GDC 2024
♫It’s the most wonderful time…of the year♫
Yay GDC is next week, where folks get to see the latest and greatest, and meet other like-minded, passionate and creative people.
We will roll out the annual roadmap. But…as I said, don’t expect us to talk about the animation system just yet. But AI will probably be everywhere, as it is evolving at such an incredible pace. Now Unity has not been standing still here, and has been investing for some time in AI tools for developers. One particularly interesting GDC AI talk which caught my eye is “Unity Muse: Accelerating Prototyping in the Unity Editor with AI”

Investment in Animation
So coming back to the purpose of this update. It has been a rather bumpy last few months, however the outcome is a laser focus on Games and RT3D where animation will play a big part, and will continue to do so for many years to come.

Chinny
Product Manager - Animation

9 Likes
2

There goes another well regarded and reputed info source. Sad to see the rate of deterioration at Unity. One small paragraph about how there is no new news and then a page of empty nonsense about “great CEOs”, how it is inevitable to screw up people’s lives and how great the AI to screw even more people’s lives.
I was looking forward to these updates, but this is just sad (no matter if you just allowed the PR department to write your update or even more sad if you really think this is an okay message).

13 Likes
3

Gee, it must be quite tough for guys still working on this big part, since not many of them got the chances seeing it through, and no user can see it delivered yet for at least half decade.

I believe it’s stressful and downhearted, but objective and unreserved, let us more clear about Animation’s going and next, thanks.

4

I’ve been prickly in the previous animation status updates, but it was mostly because of the multi-year history of Unity neglecting animation. The posts themselves appeared honest.

Not having news is “fine”, I mean it’s not quite fine, people are anxious to see something new on that front, but not having news is a thing that can happen.

What I do not appreciate is having AI dangled in front of me like keys to a toddler.

Because on top of being annoyed of what I perceive to be as a misdirection attempt by throwing some buzzwords and some services at us, there is also this: Unity’s AI stuff seems laughably bad. Like Unity Collaborate bad. Like early Enlighten integration bad.

So if you think your AI stuff is great, I don’t know what to think of the future of Unity’s animation tools.

2 Likes
5

Mind if I take a shot at de-PR-ifying this post? Feel free to correct anything I get wrong. I’m assuming best of intentions here with this.

Begin De-PR-ification

Translation: We had a small team to start with that was moving slow but steady. Then due to the low economy, we lost a team member and the rest of the team has had a hard time being productive with all the stress and drama, so we are moving extra slow.

Translation: Don’t expect anything from GDC regarding animation news. There is none, and we don’t have anything we are ready to show for animation yet. However, be prepared for lots of AI stuff. That’s not because that’s our only focus nowadays, but because that’s just what we have ready to show at this moment. Try to enjoy it for what it is, and don’t waste your time looking for animation news.

Translation: This won’t last forever, and we aren’t giving up. If the economy picks up next year like we expect, we’ll pick up steam, and maybe have something ready around then. But that’s a loose target.
End De-PR-ification

I think the main takeaway here is that while this guide is a little aged, it still remains the thing people should turn to if they want DOTS animation in 2024: https://discussions.unity.com/t/894948

6 Likes
6

Yeah, sorry my post did rather come across a bit PR-y. Thanks DreamingImLatios for your translation.
I was really trying to say:

"Look, sorry I have no new news to give all of you, even with all the recent changes at Unity, animation is still being worked on, and is a big part of our future. I will of course let you know more when I can"

I just thought something along those lines would not be well received either, so I tried to give you extra context and other info, but it came out a bit PR-y

Very interested to know what you would prefer me to do in the future…beyond “must try harder”

5 Likes
7

why your team take months for nothing meanwhile some solo dev could give us in few months ?

1 Like
8

Very few if any of us understand the rules as to what you are and aren’t allowed to share. But whatever the rules are, most of us hate them.

Assuming you don’t have any timeline updates or previews or anything like that, what we would really like to see is in big bold letters a disclaimer that nothing is final, everything might change, and ideas that sound good may suck in practice, ect.

Then, dive into some technical details about some of the design goals of this new solution, some of the problems with the old system you are intending to solve with the new solution, some of the workflow ideas you currently have that are different from what already exists, some of the cool new optimizations you are bringing, perhaps with some benchmark comparisons, some of the hurdles you are facing, some of the things you tried but didn’t pan out as well as you’d have liked, what ideas you are borrowing from the third-party alternatives, things like that.

If you did that every quarter, and the things within each discussion are different, we’ll be able to see progress.

And I guess while I have your attention, I’ve offered this in the ECS forums several times, but you are welcome to borrow from any of the tech I have created in my framework if it helps you move faster. And if you’d like, I would be happy to answer any technical questions you may have about it and explain how the various pieces work. Just PM me if you are interested!

4 Likes
Animation Status Update (Q2 2024)
9

I liked that your updates were honest and well aimed, free of the corpo fluff. We don’t need to hear how John was passionate. He was here to steer the ship, he found the iceberg in the goldmine, so we’re here. And we don’t need to hear how great the temp CEO is, he is not here for long and he’s making things worse. You don’t need to say he’s great, we won’t believe it, you couldn’t say otherwise even if you thought that. And we aren’t here for that, we’re here for animation news, which is severely lacking in the recent years.

2 Likes
10

What percentage of the system is being done roughly? How does layoffs of the critical devs that worked on the new animation system is aligned with the “focus on the RT3D”? Maybe Jim Whitehurst should ask this question to himself.

11

This smells like the animation system is on halt “till we find a new animation engineer”. Yeah, at this point, just pickup DreamingLations system and go from there…

1 Like
12

Rather than beating around the bush, just tell us that animation is dead, open-source what you can(if you really have anything), and let the community take over. It’s definitely better than telling people who have been waiting for more than six years to ‘just wait a bit longer.’

3 Likes
13

The animation system is not halted, and far from dead.

As I emphasized here we are undertaking a significant amount of development to bring you a new animation system. Not just one that satisfies today’s needs, but one which will serve you extremely well and scale for many, many years to come. We won’t ever have to go through this long gestation period ever again - well at least not for a very long time!

We have a good sized team working very hard, and who are just as eager to see this out there and getting feedback. Regrettably we have been unable to deliver it in the time frame we originally expected.

As I also mentioned how we (Unity) have made a number of false starts over the years, in a number of areas, and have frequently over-promised or under-delivered. We have heard loud and clear that we should stop this practice, so we are very much trying to adhere to that.

Even when I have little to say, I have always tried to be as clear as I can, give you some context and outline the challenges we face building this new system. The feedback on this forum has given me a very good idea how frustrating the situation is for you all, and I certainly would like it if we were further along, and could deliver it into your hands now.

Thank you for this feedback, I think it is a good idea. I will attempt to do something along these lines when I can, though I can’t promise how far we can pull back the curtain.

The first big challenge we faced was recognizing we could not have a DOTS only solution separate from one for GameObjects. DOTS and ECS is great, but not for everything. This was our first big misstep with the original experimental package.

We need a solution which works with GameObjects and ECS, and be intelligent about it. Something I can’t overstate, is that we must have an animation architecture which can grow over time to handle the scale and demands of the next-gen games and platforms, but also work with the rest of Unity - this last part is also something I can’t overstate. This is very easy to say, but far FAR harder to implement.

You might think, well why don’t you just make Mecanim better? Or maybe acquire an existing system, or something else…anything else, than spend years on this.

I also posted in my last update that the architecture of Mecanim simply won’t be able to accomplish these future demands.

I don’t know how many of you have ever tried to bolt on an external piece of software, it might quickly patch a need, but invariably it is has a far shorter lifespan, is a headache to maintain as it almost always works differently, and well, there aren’t that many options for animation.

Speaking of bolt, we acquired Bolt a few years ago and this is a good example of how integrating a 3rd party solution presents its own challenges. Visually we can make it look like Unity, but under the hood we could spend a lot of time, trying to make it completely fit with the Unity architecture. That requires resources, which some would argue could be better spent elsewhere.

How many tools are rewritten between updates to a game title? Now imagine how hard it is to have tools (Unity) which have to be ready for not one game update, but every game. Okay that is a little hyperbole, but the demands on a game engine are only growing, and we are confident that the new animation system will meet those demands, handsomely. It just takes time…well quite a lot of time it seems for us.

13 Likes
14

Thanks for reply. Currently I just want to know will official release new animation system experimental / preview version soon or later this year?

15

I am curious if Unity will be open source in the coming years, like UE. The power of the community is very strong, and there will be many people to help Unity make improvements.

1 Like
Animation Status Update (Q2 2024)
16

I think it will be source available like other package (if it’s a package). But I really doubt how beneficial on ‘the power of community’ take is. Unity users like to make games not engine dev. They prefer to fix their locally fork package instead of putting extra effor to make a PR into the main system.
How many public PR have you seen in things like SRP, InputSystem received?
I’ve been tracking those two repo on github for a long time. What I have noticed is that while people like to fork them, public PR is almost non existence.

17

https://github.com/Unity-Technologies/Graphics/pulls
81 open, 7965 closed.
https://github.com/Unity-Technologies/InputSystem/pulls
32 open, 1227 closed.

We have probably different definition of “non-existent”.

1 Like
18

I think this is because Unity is not an open-source engine, so users haven’t developed the habit of submitting PRs.
I took a look at UE’s PRs, and there are many of them.
9711829--1387738--upload_2024-3-20_11-24-18.png

1 Like
19

These was almost all made by people at Unity. Srp for example the lastest really public pr I remember was from Aras(!) *months ago

20

Even for reporting bugs, it’s easier to point out where crashes or bugs are coming from, than just blind guessing what produces the bug. In UE, we use the editor in debug mode, when there’s an exception we know exactly where it came from. In Unity there’s no way to know.