Unity to cut another 600 jobs

Better than entirely undocumented internal tools/assets which are common in industries that do not have the convenience of the asset store xP

One should show some discipline though: When using content assets from the store, import them to a separate project and only copy over what you need. Otherwise your project explodes in size rather quickly (not the build size of course but the amount of stuff in your editor).

 

Yes, and we do use assets from the store, but you can only go so far with asset packs and even with them you still need artists if you want things to look right. Our main artist isn't just working on the environments either. He's also optimizing everything to increase performance on our target devices.

Other members of the team (eg programmers) can pick up these skills but that's time spent away from the tasks that they were hired for, and now their tasks take longer to be completed in addition to taking longer to complete thanks to not being the right person for the job which may require hiring more of them to make up for that.

 

Unity, however, is involved in your game's ability to actually run at all, as that's what a game engine does: runs your game. This contrasts with something like Blender or Photoshop, where those softwares are not shipped with or required to view the content you make with them.

Royalties on game engines in general have never bugged me considering part of the shipped game is the engine. So, when you ship your game and generate revenue off your game, you're shipping and generating revenue off something that's partly not your work.

You might not think it's fair for the engine developers to get a percentage of your game's revenue since nobody actually "sees" or "hears" the engine, but when you take into account the countless things the engine is actively doing in the background to present your work and allow it to be interacted with, then yes, people do, in a way, "see" & "hear" the engine, especially when your game uses a unique or cutting-edge feature of that engine that is not possible in other engines.

So, if the engine developer is asking for a small percentage of the revenue (ex. 5% for Unreal), they're essentially just asking for a percentage of revenue from the portion of your game that is their engine. And with a royalty model, the more successful your game is, the more successful their engine is, because, once again, part of your game is their engine.

I'm not saying whether or not Unity should pursue a royalty model, just saying that, in general, it's fair for engine developers to ask for a royalty on something that contains part of their work.

 

Like was mentioned earlier though, and pretty much any time royalties come up, almost no one is actually going to pay royalties and even when they do it won't be for very long. Solo developers will only rarely hit the million dollar per game threshold, and the few that come close will just negotiate before they hit it.

Most studios will start off with investments and be able to afford it out of the gate. A studio with a handful of developers will be about the only ones that would have to pay royalties, but once they're off the ground with a successful title they're not going to stick to it either.

Unreal's development isn't actually subsidized by royalties. It's subsidized by enterprise licenses and Fortnite.

 

I didnt check lately what Unity sub is offering other than increasing revenue threashold, but wonder if Unity decided to go xbox game pass route. Like You pay a sub, but You have access to all the services Unity provides (like speedtree modeler access etc), because as of now I have a feeling from all the aquisitions Unity made, the ppl who use Unity didnt benefit too much.

I am not saying to give the free access to them, but bundle them inside the sub, so there is more incentive for ppl like me to buy the sub and enjoy the services for a reasonable price for Indie.

 

Can you divorce the general problems of game design completion versus the technical problems of finishing in Unity?

 

seeing in twitter, unity also issued forced Return To Office,
and then this
https://twitter.com/MirandaDue/status/1655598913653489668
(and few hours later she got fired)

huh..

 

We touched on that a page or two ago. Short version is that we don't really know anything aside from a statement from her that she was fired for breaking the code of conduct.

 

At Disney/LF, we had one come by about once a month, usually for a half day. And as you said it was useful for solving problems, and direct action on bug reports. Certainly for that purpose it was good. But, really it was just that, insights in to the engine, solving specific problems. But it was never a discussion about architecture, or best practices or anything like that. It was a bug reporting session really. (again, very useful). But if Gigaya was meant to meant to gain insights in to actual game development, it couldn't possibly. If there were Unity folks out there who's goal was to improve the engine based on real world use (not just engine bug reports), we never encountered that, it was just customer service (which, again ... good). It was never about what we needed in terms of the future of the engine or how we worked. Maybe what you described does happen, but in the last decade+ across 3 large studios, we have never encountered that. Great if it is happening, it should be expanded.

 

That is what the license fee is for. If however, they were actively developing features and fixing bugs specifically for our game, that would be different. But they are not, they aren't a partner, they are a vendor. A very useful one, for sure, and worth the cost for sure, but not a royalty on my/our work. They aren't taking a risk, we are. Or as others pointed out, being involved in and/or inspecting our financials.

Also, just to be super clear here, I am a game developer, not a financial or business management type, so I have no real strong (or valid, really) opinions on how Unity should run its business. It is a great tool, I am happy using it, and I am not worried in the slightest about it being in "trouble". It sucks when people get cut, and I wish them the best. I am sure many studios would be more than happy to have folks from Unity.

 

Interesting! Whenever we visited studios in Europe we would be constantly asking for feedback and what Unity could focus on more, and then loop that feedback internally to whoever needed to hear it.

In reality right now, what a game studio wants and needs is not in direct alignment with what Unity wants and needs because they build different things; and the problem is prioritisation. Was any studio asking Unity to spend $1 billion on Weta? Specific example: game studios in Europe and Middle East were asking constantly to have correct out-of-the-box RtL support for game text as they had to launch their games in countries that require this for adoption. I lost count of how many times we raised this internally. But the answer was always "yeah, we are working on it for some time later". But nearly a decade of this request and Unity still hasn't implemented it and has even moved resources off of UGUI/TMP. Studios have to spend time making their own system or use something from A$ or 3rd party (which they tell us has incorrect rendering so is not useful). There is no internal fire being lit under butts to get this implemented; despite studios begging for this workflow.

As many people have said in this thread and countless other threads; there are thousands and thousands of pages of learned experience from studios and developers with feedback for Unity. If Unity is going to decide not to dogfood and instead get closer to users to understand their needs; Unity needs to do better in handling and processing that info.

On Gigaya, I disagree with some things you raised, but I don't think its worth going in circles about it; especially as its not happening anyway. Agree to disagree.

 

And this, precisely, is why I didn't see a lot of value in Gigaya. The problem is not and hasn't been a lack of information, it's the fact that Unity chooses not to use it.

I was happy to figuratively eat my hat when it looked as if I could have been wrong on that, but now we'll never find out.

So, what is it that you think Unity wants and needs?

To me your example is a great one of Unity's prioritisation being nonsensical. A lack of good RTL text support is surely low hanging fruit in at least three different ways:

• Since plenty of software has it, I imagine that it's quite a doable thing, so it's surely a good opportunity to help your customers and do that whole "democratise" thing.
  
• Every time someone who needs it has to work around it, it's an opportunity for discontent.
• Given both of the above, I'm sure it's a great opportunity for competitors to do it (if they haven't already), and then just casually let everyone know "yeah, we've got solid RTL text out of the box".

Obviously none of that will get a payday or lose customers tomorrow, but you've said it's been over a decade, and referred to studios in multiple areas.

 

Honestly, what Unity needs is to develop features and engine improvements faster. Real-time lighting is a great example. Or just full DX12 integration. Last I saw that's still on the roadmap, and its stable release was two years ago. Or multi-panel display (you know, the necessary feature for LED walls for virtual production), that's also still on the roadmap and was first seen in real production in 2019.

I honestly don't understand why Unity's development is so slow; it's not like they have fewer engine devs than other people, but it's exactly the reason we aren't using Unity, as much as I'd like to. I frequently check back in to see how engine development is coming and every time I do I have the same reaction.

"Oh, they're still trying to implement that? Hasn't that been out in other places for a while now?"

 

Adding another example to Unity needing to prioritise practical uses and user feedback over MVP implementation.
https://forum.unity.com/threads/how-to-know-when-a-timeline-is-complete.493058/

 

Unity's most powerful "tool" is community. But looks like they are not aware that thing.
Countless tutorials and documentation comes from community.
Asset store that big because of community. Also github full of free tools.
They get massive amount of alpha ,beta testers and bug hunters for free from the community.
They supporting and teaching each other on the forum even for the advanced experimental topics like dots for free.

So why not using such a powerful "tool" for everyone benefits, with open sourcing engine.
Unity wont fix their engine even problems stated internally countless times, then let that frustrated people fix it for you.
Unity wants employees that cost less? Well open source contributors works for free

Lot of people sick of stopping by some unity black boxs when trying to roll own their solutions.
So why wont Unity open sourcing the engine?
What kind of revenue coming from the source code access that overweight quality and extensibility of the engine?
Im talking about Unity engine Source Cod, not services, not Ziva or Weta. Also come on, this is 2023.

And why do we need speculate that much abut direction of whole company.
I think we need a "Management Blitz Day" soon as possible. Or accountability and questionability is not exist for higher positions?

 

They said that licences of third party stuff used in the engine is majorly in the way of that.

 

I believe this. So why is Unity developing so slowly? Why are they still struggling to implement years-old technology? Why is Unity so far behind?

And, to bring it back to the topic at hand, how does laying off 600 people fix that fundamental problem?

To be clear, these aren't loaded questions; I genuinely want to know. Because I want to use Unity, I love the workflow and (luckily) I get to make the decisions about what engine we use. But I can't justify using Unity when it's years behind the general development curve, and doesn't appear to be getting better.

 

Ngl idea sounds like a lot of fun

 

In my opinion: lack of tactical management of handling of resources (people, skills, knowledge), plenty of silo teams who don't have structure to communicate/collaborate effectively and fairly consistent pivoting of focus to whatever is new and shiny; meaning the thing that was originally being worked on or maintained is down prioritized instead or never improved upon (despite it being used by millions of people)

 

That was a nice excuse when unity was a small company. I wonder which license blocking them now. PhysX already open source so mono and CoreCLR. Umbra sucks and will be replaced anyway. Of course we not know about underlying systems, but I don't believe that Unity can not negotiate them.

 

Like clockwork... oh well. I voiced my concerns. Apologies if they were unnecessarily offensive to anyone, suppose I can be a bit of an ass sometimes. There was some hyperbole in my post.

 

Dude. That is one of the most valuable community members. You kinda should check his plaque, it is there for a reason.

Honestly, if you hate unity so much, you should just leave. The door's wide open. You can pick other tools or write your own. With blackjack and other features.

 

One thing you should keep in mind is that unity ignore list feature hides responses silently. Meaning you won't know sometimes that there are additional replies. That might make some threads confusing, until you spot "Show Ignored Content" link.

 

Saying devs dont care is a big strech. I can imagine devs care a lot, but they are constrained by management, budget and deadlines.

Maybe Unity is trying to do too many things at once? Maybe more focused approach would be better? Hell knows , time will tell.

 

Brook's Law? "Adding human resources to a late software project makes it later."

Edit: Note that this is just the common, simplified statement of the so-called "law". The fellow who wrote about it went into far more detail, gives exceptions, etc. The real point is that more people does not always mean more output or more quality.

 

And glacially slow internal processes. It's mostly on management and perhaps on the size of the company - the bigger they are, the slower they move due to teams being scattered all around the globe in different time zones. This is why I view office reductions in a positive light, a more consolidated company structure should gain some of that development velocity back.

 

I WANT to believe you, but without any empirical results it is getting harder to believe anything gets done over at Unity.
Blame anyone you will for the lack of progress, but if you are a dev and your manager is holding you back, you go to his manager, or another layer up. You escalate until the frecking CEO comes to lay down the law. I find it hard to just blame "mismanagement". Each dev is there in the mud and knows what is needed to make his/her/their work shine. If you can't get that through corporate, you might as well just quit, because you are basically there to make a paycheck and not for any passion or will to improve.

You have been on the inside, so i trust you have more insight into things, but from the outside looking in it looks very bad, and i dont see that changing anytime soon.

 

That's not really applicable here.

From the article you linked:

The people who were let go weren't new hires, nor is Unity a late project.

The people who were let go, in a lot of the cases we've seen, were specialists and extremely skilled programmers.

This one I've largely highlighted because we've learned in the past that a lot of teams in Unity in general have very small head counts already. I don't remember the specifics but I seem to recall the URP team and the HDRP team having a pretty substantial gulf in the amount of people working on them, and we've heard here that the ECS team has been hit as well. This feels pretty concerning when it comes to Unity's ongoing issues with systems integration already.

Brook's Law not only doesn't apply here, but the very reasoning of the law points to other, bigger issues, especially to the end user. It's also dangerous to assume that the inverse of the law is true.

 

He probably is very passionate and disappointed at the same time.
While his wording might be a bit harsh and overblown, some of his arguments has merit, at least to me.

 

Unfortunately, thats not really how it works in most larger corporation; the higher up you go the less in-the-trenches technical people generally are there and less understanding of some challenges that need to be overcome. Im now imagining an engine developer scheduling a 1 on 1 with Riccitello just to rant about the Addressables window being tempermental and they want to fix it.

 

Please don't say this! Anyone is welcome to discuss and debate with anyone and I don't think 'status' should limit what another person says; as long as its balanced and not insulting.

 

Words of frustration stem from the love we all have from our beloved game engine. Though of course it's important how the message is delivered.

A wise man once said "Unity's most powerful tool is community" and I'm confident Unity Team is very capable of navigating these storms and they have our full support.

 

I imagine him having a 1 on 1 to adress the fact that his work is being hampered by layers of slowdowns, and that it has impact on the roadmap of the engine, you know, things the CEO may actually care about.

But corporate in my country (NL) is much different from US (or anywhere in the world really) as we generally dont do much hierarchy, and definitely dont care much for authority

 

"If I can fix Addressables to not be frustrating in the editor, the stock price will go up by $12!"

"I approve 10 headcount"

 

i think you get what i was going for, but choose to make it look silly. I get that..
But what i get from this is that Unity upper management really doesnt care about their people, or about the product they are trying to produce.
Shame that all that supposed genius is going to waste then..

 

I was messing about as a joke (sorry); I understand what you mean and agree.

 

Status doesn't limit, sure, but people deserve common courtesy and politeness. That particular member was lacking in this department.

 

It's very hard to get an accurate picture of Unity right now. Data doesn't show financial troubles just yet. They might not be the most profitable tech company in the world, but they're also not going under. But I also see development moving at a snail's pace. Packages can't get fundamental features and optimization now going on for what... 7 or 8 years? The basics should've been covered years ago.

And there seems to be a lack of vision for Unity both on macro and micro levels - the roadmap videos in the past few years largely recount what has already been done, not what is coming up. And most of the bigger changes that are upcoming stem from decisions made years ago.

Personally, I don't see where Unity is going. All I know is that Unity has been stuck in a very long tech transition period that never seems to end and even tiny engines like Godot get automatic LODs, HLODs and real-time GI before Unity, the most used engine in the world. And no, abandoned GitHub repos don't count. I'm talking about proper engine integrations. Has Unity peaked? Is this the best it'll get? I guess time will tell.

 

Just looks different to me. So much good stuff in the past couple years. DOTS/ECS F***ing rocks and it's making great strides forward. Biggest upset in the Uni-verse (ok, lame) in years. Mono used to be stuck for years. Now C#/.NET is making good steps forward all the time (not to mention the amazing .NET CLR stuff that's not too far off. Asset DB was stuck in v1 land for so long but now v2 is starting to reap real benefits. Could name other things but to me, Unity is the best it's ever been and keeps getting better. Sure, doesn't move at breakneck speed but it's a massive platform. Don't even want it to move at breakneck speed.

 

ECS stuff is great, but that's like...the only major advancement they've made to the platform in years. When I say they're slow, I mean...as much as I don't want to, compare it to UE, which introduced nanite, lumen, procedural content generation, procedural destruction, metahuman, substrate...oh and also their own version of ECS. In the last couple of years.

Like, I'm all for taking wins when they come, but toxic positivity is just as toxic as negativity. And trying to spin it as a good thing that you count Unity's platform growth in years-per-feature, rather than features-per-year, isn't doing anyone any favors.

 

And can we count actual negatives, too?

600 less people to work on getting things ready and/or fixed and/or performant and/or done.

Post processing in various tortured states of limbo for nearly a half decade on URP/LWRP.

Calling LWRP "Production Ready" when it was nothing of the sort.

Claiming the same of The New Input System, VFX Graph, Shader Graph and Visual Scripting, long before it was close to being true.

Lack of a new, high performance, optimised/Burst/Jobified/ECS based animation system.

New audio system is stillborn, and inbuilt Fmod is ancient and buggy.

Project Tiny abandoned.

The fiasco with lightmapping/Enlighten

Shader compilation times blowing out in ways that are making Unreal compile times look tolerable.

Editor entering Play Mode iterations getting slower and slower, each year, and now requiring the turning off of the reloads of the things that we'd want to iteratively change and test in order to get fast Play Mode entry?

General performance degradations of the Editor, year on year, since 2019.1

Package Manager package version updating and then matching becoming the kind of workload that only System Administrators can enjoy.

Increasingly dated and ever more sparseness of documentation for just about everything.

Asset Store becoming a bit of a backwater as many devs have given up trying to meet the SRP update requirements.

Whatever is going on with versioning and collaboration.

Unity Networking???

 

Is it making great strides forward, though? It's been what... 6 years since announcement? And we're only now going to reach 1.0 for a hybrid version that was supposed to be pure ECS. It's not even production ready yet, and there's no ECS Animation or sound in sight. Sure, it's enabling a certain kind of game that wasn't possible with Unity before but for most of us it's not really worth the added complexity. And thought leaders of the DOTS movement have left Unity or stepped away from the company. ECS is here now, but there won't be some mass migration as some people envision and its future is unclear.

That's just Unity mostly catching up to the rest of the industry and most improvements are external to Unity in a sense that Unity didn't contribute to those improvements. Asset DB v2 is nice though.

 

This is a great breakdown of all the things that have fallen to the wayside. The only thing I can think of offhand you might have missed is the terrain system that was rebuilt from the ground up recently to be only slightly less terrible than it was before.

To be fair I do want to say that the build teams have made great strides allowing unity to be built on many platforms without any headache, at least from my experience. And I'm especially impressed by how web builds are currently running on mobile devices, despite a couple major bugs, but progress does seem to be happening in this area.

I'm also VERY MUCH enjoying the built in repository tools that are just robust enough to get the job done without much headache through plastic at a reasonable price.

It's been a little while, but the new prefab stuff has also been such a nice addition, as well as the constant little UI improvements that have ben happening here and there.

The input system and the state it is in is especially confusing, because you'd think a system so simple and integral to a game engine would be figured out by now. Unity went so far as developed the newer system on the side, to avoid breaking the original input system. And now there's these 2 half formed systems just kinda there, neither feeling fully integrated or easy to implement. Which just seems so weird to be talking about, because this should be easy to figure out for a company with so many resources. It's like the more core, easiest, most basic thing a game does: You press button, thing happens. I don't have the mental energy to remember why it's so much of a pain in the ass right now, I just remember it being so.

But how do they go about doing this with their non destructive protocols? Do they make a THIRD input system, and implement that one when it's half finished as well? Ugh...

 

Was reluctantly optimistic about dots and ECS , because although the in editor performance was never an issue, we knew that the story out of Unity was, "After we figure this stuff out, THEN we will build the perfect engine." So speed gains didn't excite me, but WAS looking forward to new editor stuff.

My theory is they never figured out the new hotness of ECS and DOTS in a way that it was easy for their teams to build their foundations on, and these new sweeping complexities just absolutely decimated the various projects and work practices many developers were gearing up for. Unity was so proud of these new performance improvements they stubbornly shoved the noses of all developers into adopting the new practices, confident they were going to "change the game" of gamedev.

So how would management react to taking a huge chance on some big pie in the sky ideal for the perfect code paradigm, and the perfect engine that was going to be super usable but also super performant being a massive DUD? They'd probably cut off all coms as the internals went absolutely nuclear and nearly all forward progress didn't just halt, but decimated motivation and faith in the platform from most all key talent as tempers raged from "Told you so" to "Well if we had only done it this way...".

Then you have a situation where the various teams start pointing fingers and internal energy is spent on blame games over picking up the pieces and moving forward. Combine that with the insanity of the past few years and you have a powder keg where everyone is just upset at one another and nothing gets done and everyone is resentful and all the core talent that knows the idiosyncrasies of the engine leave in anger, eager to see the platform fail and you have us users in the dark sitting here wondering wtf is going on, and current employees trying to piece together some piece of alien technology that they have no frame of reference for how the darned things fits together as management tries to figure out the best way of salvaging a horrible situation.

You have a company in shell shock, you have employees have actual and totally understandable nervous breakdowns as their worlds come crashing down around them. And you get this new culture we see today, no ones fault, but also detrimental to the future. Management grasps at random notions of what went wrong and what went right in a murky grey sea, and now plans and protocols become increasingly controlled and nonsensical. No big plans, no hype, just figure out how the core things work again for the foreseeable future. Lights off in the building.

My guess is, the grand irony of Unity's great failure wasn't in a lack of vision, ambition and risk, which they are all suffering accusations of now, it was over investing into a great dream, into feature creep, and allowing only a few voices to override concerns about future dev issues that would stem from global code overhauls and ballooning code requirements from the ECS and DOTS revolution.

Edit: No specific individuals in mind, just thoughts on how the current situation may have gone down given the usual company structure. I want to make a point that I'm not trying to paint a picture that it's any individuals fault for having big ambitions, for wanting to do something amazing, for investing into an awesome dream. Not being accusatory here, just trying to piece this together as more information comes to light. I think we can all sympathize with the trials and tribulations of gamedev. I could very well be wrong about damn well all of this, but I've also been in some pretty dire situations on various projects and have seen this sort of thing go down before, it's not exactly unique or complex.

 

What is specifically good about v2? One thing I noticed is multicore imports. What else does it improve?

 

I wouldn't assign motivations and attitudes to prior and current Unity members, as far as I know no ex-Unity member wants the company to go down or anything. Nor do we know the internal state and team relations. The rest of the tale sounds plausible, but we can spin a dozen different versions of events and it'll sound just as plausible. We just don't know what is going on, and these discussions about what could've happened and what is happening now is just pure speculation. It doesn't help that Unity just don't communicate.

Scales better, is faster and more reliable: https://blog.unity.com/technology/t...olid-foundation-for-speeding-up-asset-imports

 

This. The royalties hit a narrow slice of developers that both:

1) Make a game that generates over a million dollars in revenue.
2) Make it all the way to launch without raising enough cash for a custom license.

So the royalties tier exists mostly to cover the gap between free users and custom licensees. Epic increasing the threshold from $50k to $1m wasn't just a market stunt or an act of goodwill: collecting those royalties actually costs some money, which can be saved by focusing only on the big earners.

 

Unity has a lot of talented passionate developers. I my view the inefficiencies are to a large degree driven by mismanagement.

I am not on the inside so I cannot know what is going on. I do read the Twitter feed etc. It is full of content relating to stuff that I am not interested in and is not part of Unity’s core business. Since it is so deeply rooted it probably is part of Unity’s corporate culture and goes beyond trying to appeal to ESG investors. If woke culture has penetrated management to that degree, it makes me wonder. It is probably part of the California culture.

Maybe Elon Musk will buyout Unity

 

Please no. That's like the worst case scenario.

 

Microsoft would be the ideal candidate in my mind because they don't suck at documentation.