Unity is laying off staffers

Yikes.

What are your thoughts on this?

 

My thoughts are that it being described as "a S*** show" lines up with what I've been saying for ages.

These are all things I have said have been entirely obvious from the perspective of somebody using the engine but also paying attention to the company.

 

No thoughts.

It is simply something that is not my problem.

 

I'm not at all surprised. Unity is having a bad year and we're heading into a recession.

 

Not my problem. Hope Unity spends their resources on DOTS, Weta and HDRP.

Just a shout out here from me to the laid off staff: Hope you are all well and continue from strength to strength. Good luck to you all.

 

A big yikes but not *super* surprising. It wasn't obvious, but it wasn't exactly hidden either that Unity has been in a rut lately and hasn't really been making any headway with getting out of it.

• They've got many features 'in development' that never made it to production-ready, and those that did still had bugs.
• The features that did get through got deprecated and had to be recreated by the community(Enlighten, networking, etc).
• Many acquisitions have been made, but I've seen nothing at all come from them. Weta, Speedtree, etc.
• The engine's speed has decreased. How many times are you recompiling scripts or waiting for the engine to finish loading nowadays? Compare that to 2018 Unity (however, admittedly, 2018 did have a lot less features).
• THEY HAD A DECISION (THEN REVERSED IT) TO RETIRE UNITY ANSWERS? (CAPS INTENDED) This was the most baffling thing and honestly, really showed that Unity was out of touch with it's primary base. I'm really curious about the thought process/data that was shown for *anyone* to make that decision.

And these are all off the top of my head over the past few years. If it was this bad on the *outside*, I can only imagine what it was like on the inside.

Having said that, the engine still has it's uses and unless Unity shoots itself in the head and forces all their users to move to unreal with visual scripting, it probably won't make too much of a difference. But the only way I see it doing that is if it decides to retire Unity Answers and change what users have to code in. C#+cross platform+well documented make it a dirty diamond in the rough to make and release a game.

 

Global stocks diving down.
Streets small shops and services closing down and leaving empty premisses.
Prices go up.
Other economical critical variations.
Nothing special here. Don't try to blame Unity here.

The article is typical screamer, without any concrete merit.
For example, which department are affect.

 

Everyone is downsizing. It's a recession, the one we've been warned about for years but chose to ignore.

Anyway, whatever Unity has to do to stay Unity and not sell out to China or Saudi Arabia or some conglomerate is fine with me.

 

Every big company is laying off people, downsizing, etc.

Hell maybe downsizing will help them focus a bit.

 

Ex EA CEO...nothing more to say about it ))

S*** wont change, they cut costs due to economical issues and any dreams about "downsizing will be good" is just this, dreams.

You guys should know how corpo works

 

Reading this, all the worst suspicions from years ago come true..

 

Yeah yeah yeah. This is just the same procedure repeating across any industry. A bad financial situation and most commonly a recession, and the stocks start to plummet, which makes the board of a publicly traded company and/or their stakeholders and/or their investors nervous so they do what‘s most effective: cut the expenses and that means staff since those are the biggest expenses. Nothing special here. It‘s almost as if the article read „Rumor: presidential elections to be held AGAIN in 3 years“. Come on, post something newsworthy!

As to the story: with any layoffs of this kind you (sorry, I mean: media) will easily find (because they are actively looking for) that one outspoken angry person who won‘t hesitate to trashtalk the situation who, more likely than not, was going to be laid off anyway just for being a badmouthing prick who had already mentally quit in the first place.

And regarding „AI and engineering departments“ affected, you all know that Unity is a tech company with probably around 75% of its workforce being engineers, of course those are the ones most affected.

Good luck to those who got laid off. You‘re in a bad situation but also a good position, since you had a job with Unity that‘ll serve you well in the rest of the employment market.

FWIW I worked for EA under Ricitiello and got (voluntarily) laid off. No hard feelings.

 

Haha, I read this as „Hell is downsizing“. Now THAT would be news!

 

Indeed. One person is just one side of a story. Speaking of which articles like this one are very much the norm at Kotaku and should always be taken with a grain of salt.

 

not looking good

 

Meanwhile, in late-stage capitalism...

https://www.benzinga.com/sec/insider-trades/u/john-riccitiello

 

Engine in 2023.1 really needs big reorganization but I feel sad for layoffs. On the bright side tech industry needs workforce and I hope they will find their new job fast and with ease.

For engine reorganization I hope HDRP and script compilation get the priority and more resources to become standard for Unity and ecosystem.

 

It would be nice if, for example, 'the renderer' or 'the UI system' were development priorities.

But we've now got 3 renderers (with separate postprocessing stacks), that exist in isolation, competing for attention/resources. When new high-end features come along (e.g. DLSS), only HDRP is likely to get them - but many people are trying to do fairly high-end stuff in URP or the 'classic render pipeline' for performance.

And it's not just rendering, there's 3 UI systems, multiple physics engines, multiple input systems, Monobehaviours vs ECS, and so on. And there's no sign of any 'legacy systems' ever actually being removed, as the new ones either don't reach feature parity or just take a completely different approach that may be undesirable to some users, so there's usually plenty of reasons for people to stick with 'legacy systems' even when starting new projects.

Not sure how they can move forward from the current situation. There'd be massive outrage at any suggestion of removing the classic render pipeline, or UGUI. Maybe they could fork off a 'Unity Next-Gen' product that allows Unity a one-off chance to strip out a whole load of 'legacy' stuff and forget about backward-compatiblity for a while, giving them the chance to make fundamental changes/major upgrades that would be impossible at present?

 

Should have gotten rid of their CEO instead of workers...

 

I wonder if there's a pink-hair to pink-slip correlation?
I think it's been obvious for a long time that Unity has become an over-bloated company, losing track of (or even ignoring) their core customer base in search of larger more diverse markets. Hopefully leaner times result in them trimming the actual useless fat in their workforce and refocus on engine problems instead of valuation problems.

 

Maybe they'll become a tiny bit more nimble for a change. Everything's always two years away and after those two years, it's still somehow two years away. The paradox of Unity.

 

Unity once upon time had 3 languages. JavaScript, Boo and C#. Now you got one.
At least Unity shares they R&D and test in the field, rather keeping hidden under the hood.

 

Thats the problem.

It's that snake that is eating itself. Maybe it has ate so much, if you cut the head off, there is no snake left anymore?

Corporations with modern leaders like this lay people off when they are doing good too. Their word means nothing. You can't trust anything.

 

Yep. In the "7 habits of highly effective people" one of them is: Assume the half-life of your profession is 2 years. Act accordingly.

 

I'd count visual scripting as a separate language. Makes 2 languages.

 

I always hear that Unity isn't really making money out of their engine or asset store ("asset store is about 2% of their revenue" is what I read on reddit). They are making most of their money from advertisements.

Maybe if they would change to a rev-share system, like Unreal, they would get more money for their engine and could focus more on this part?
Assuming, what I read on reddit is true. If someone can confirm or disprove, please do so.

 

Not with the current implementation. Unity Visual Scripting automatically generates 99% of its nodes from Unity's C# scripting API. It's a C# visualizer.

 

I don't think people understand the purpose of different pipelines, it's not to make your life harder but to target different hardware. You can't just have the same rendering quality targeting a mobile phone and the latest AAA console at the same time. If you want true AAA, pick HDRP but don't expect it to perform the same way as URP which targets low end phone hardware, Quest, Switch and so on. Which is why I thought that LWRP was a much better name for URP, since it is indeed lightweight. This obviously comes at a cost - you won't be able to have the same type of advanced lighting and effects like the High Definition version has, but it can still look good as proven by Gigaya and previously Boat Attack demo which I still like and it still looks good. I always use it to benchmark phones and to see if the new LTS version builds properly.

As for having Unity Next Gen with next gen features - I would agree that this is a better way rather than having to support so many different systems with each version. It's how Unreal did it, switching from 3 to 4 was a huge leap that made everything in the previous version obsolete. But on the other hand, that was 10 years ago, different times. Now game engine space has huge competition, almost as much as the games themselves. Unity probably makes these decisions frantically in the heat of moment (that's how it often feels like), because they have to move at a rapid pace in a fast moving industry. And they have so many versions to support - 2020 to 2022 alpha, and they probably already started working on 2023. I don't see how it's humanly possible to maintain so many versions with quality development, and then have to spam patch releases for each and every one. I've seen supposedly mature 2020 LTS versions sometimes break really basic things, usually minor things and you can usually roll back no problem, but this doesn't inspire confidence in updating. Generally looking back this switch to yearly releases was pure marketing decision and to justify the subscription model, because the pace of development clearly can't really keep up with the new version number spam.

 

I don't think people don't understand the difference. It's that people think splitting dev time between three incompatible renderers to 'target different hardware' rather than just having one renderer that can have features turned off and on to target different hardware is a stupid way to structure the engine since it makes the engine compete with itself for developer hours and introduces unnecessary compatibility issues.

 

Originally they planned to only have 2 pipelines, Built-in is still here only because people refuse to let go of it and switch to Scriptable RPs. Though in some cases I can understand that, now that they are finally somewhat stable and URP almost has feature parity with Built-in there's no reason to still use Built-in. I haven't used anything other than URP/HDRP for the past 2 years and haven't looked back to Built-in even though it's not always easy transition, I try a lot of assets and there are still some paid asset devs still didn't get the memo they should let go of Built-in. URP should just become the default, and if you want high fidelity and target PC/Console first, use HDRP. Simple. I don't get why people still complain about that.

Either way, porting between platforms, especially from a more powerful to mobile/low power, will still require a lot more work than turning features off and on, and it's not like you CAN'T switch between pipelines, they are scriptable and open after all. It will require some more work, but it is doable.

 

In the real world, you often want a game to support everything, from Switch to PC. You often need one render pipeline that's scalable by toggling features and quality settings (and maybe swapping some lower-detail assets in). Switching render pipelines in Unity can be a major ordeal (if you've got a lot of content set up), it's not something you want to even attempt doing per-platform on a multi-platform project.

URP isn't lightweight any more. It's doing a pretty good job at it's attempt at being universal. On higher-end platforms you can use deferred lighting, more lights+shadows, more complex materials, and SSAO and other heavier postprocessing effects.

In my experience, HDRP seems a real performance hog, and not all scenes can make the most of the fancier features (try to use volumetrics/fog in a space scene, for example, where there's no 'sky' and huge draw distances). If it works for your project, and the visual quality in justifies the cost, then great. But it's not necessarily the best choice even if targetting high-end platforms.

But for URP users there's some super-important stuff that's 'trapped' in HDRP-land and painfully inaccessible to URP users, TAA and camera-relative rendering being a couple that matter to me at the moment.

 

It literally doesn't support any post-processing effects aside from its own built-in ones unless you roll your own stack. You should not have to use HDRP or built-in just to have properly implemented post-processing. On top of that, the only "documentation" for this is a bunch of third party youtube tutorials that are in various states of out of date.

 

This sucks. I don't see what John Riccitiello has done to deserve a 160% pay increase since 2019.

Also, "not my problem" is a weird response.

 

What a shame. Some great developers were let off, I especially appreciate any of the guys that communicate with us here in the forums, in a general way and not just announcements.

----

Spoiler: URP rant, once again..

URP is an absolute disappointment, It's 2022, it's been in development for years and it lacks some basic features you'd expect in 2022. Even it's SSAO is 3-4x! more expensive than the one in HDRP, and lower quality.

The fact we're still talking about feature parity with built-in, in 2022 isn't good.
It should've surpassed built-in, it's features shouldn't have the same old limitations as built-in. Look at reflection probes, shadows, post processing. It's the same or worse.

It's been in development since when? 2018? I mean come on, you can't pass it as new anymore. That excuse is buried and gone.

Any PBR engine should offer TAA, maybe you enjoy specular aliasing everywhere but I got tired of that years ago.

No SSR, no TAA, no cascade blending, no LOD crossfading(seriously.. you can support but for some reason all URP shaders don't, until 2022...), SSAO is far from good, shadows about the same (compared to great shadows in HDRP), no tessellation, It has decals which is great, but enjoy applying it to all meshes - no layers for decals, nothing great about post processing compared to built-in, the same fog from a decade ago, and so on.

When will we see new progress? Imagine built-in but with active development since 2018. Seriously. Improvements and new features. But no, here we are, comparing features to a render pipeline that was basically killed over 4 years ago.

In a way you don't have to imagine it, look at HDRP.
I'd imagine the graphical improvements would be: better soft shadows, TAA (which honestly was a disappointment, but some improvements are coming in 2022), tessellation support, better fog +volumetric fog, the list goes on and on.)

Actually built-in already supports TAA.

Even now you can achieve graphics in built-in that you just can't in URP due to some big assets not supporting URP.
Next-gen shadows is one example.

And I disagree about not being able to support a bunch of platforms with a single render pipeline.
Look at all the engines out there, public and private -- do they have some magical powers unity doesn't?
I mean, look at built-in, it's done that for a long time. And with some assets you could push the visuals quite a bit.

All you have to do is give options. That simple. You enable the right feature for the right platform.
Advanced, high quality SSAO? Enabled on PC/consoles, disabled on mobile and replaced with baked AO.

There's nothing impossible about it.

 

Does urp have the shader replacement rendering stuff? I'm abusing that and last time it wasn't in urp.

 

Maybe we need a 'community branch' of URP, a 'Project HD-URP' even, where we try to port specific HDRP features to URP?

(I'm currently doing some work on a project that'd very much benefit from camera-relative rendering, and have wondered how feasible it'd be to add support to URP myself, using the HDRP code/shaders as reference)

 

What we need is better SRP documentation and resources, beyond what Unity offers right now, but that's really all a discussion for the graphics forums I think.

 

First Quarter 2022 Financial Highlights

Revenue was $320.1 million, an increase of 36% from the first quarter of 2021.
Create Solutions revenue was $116.4 million, an increase of 65%; Operate Solutions revenue was $184.0 million, an increase of 26%; Strategic Partnerships and Other revenue was $19.7 million, an increase of 11%, each as compared to the first quarter of 2021.
Loss from operations was $171.2 million, or 53% of revenue, compared to loss from operations of $110.9 million, or 47% of revenue, in the first quarter of 2021. These results were impacted by an increase in stock-based compensation expenses.

 

We have three but only two of them receive developer resources. Built-in is in maintenance mode receiving almost no attention and the hybrid renderer isn't a true renderer only processing the entities before passing them to HDRP or URP.

 

tutorials, projects etc. like old standard assets. github doesn't cut it, ie. it's good for continuous updates of development progress, cutting edge, but would be nice to have packages, official releases in the package manager. gigaya was a move to the right direction. this should be for particles, audio, physics, terrain, shaders etc

 

Yeah like for putting the company in a state where they need to lay off ppls CEO should go away instead, heck his paycheck is enough to fund all those ppls jobs. His bad management his head should fall.

 

What does this even mean

 

Shadowgun is dated now but at the time it ran on 512mb phones and they custom made every shader and it was amazing for the time, using standard renderer. That was the secret to great graphics and performance in Unity, non default shaders.

Lots of people/companies had shader collections built over multiple years for every problem they could face which they had to give up or recreate alternatives to move to a new render pipeline. Does hdrp support multiple cameras yet? Looks like they finally added VR support, for a long time if you were developing VR it simply wasn't an option. There are a lot of AR apps that utilize multiple cameras for layering.

Unity does this S*** all the time going back to raknet being the built in networking, and it worked great, it was super simple, so they scrapped it to come out with uNet, broke everyone's old apps, and then they scrapped that to come out with whatever they have now. Whatever networking they come out with now is dead to me, I don't trust them to keep it running more than a year or two and it's a problem that has been solved by plenty of products and even companies dedicated to networking.

Same with the UI, back in Unity 4 we had GUITexture and GUIText elements, some people asked for css-like ability to layout elements and everyone shot that down saying it was a terrible way to build UIs. So then nGUI gets popular on the store, textmesh pro was another paid asset, Unity acquires both, comes out with uGUI and official textmeshpro. But now 2022 rolls around and guess what we're going back to, UIelements and a css styling. Loooong ass detour in between 2014 and 2022.

Lightmapping...same deal.

 

This does more harm than not responding at all, IMHO.

 

Interested in this as well.

 

hahaha that was too easy

 

wait wait i do not get it, is this a sarcastic joke? or did they really let you go?

 

Weta alone had like 275+ engineers and weren't there like a few other companies they bought in last couple of years? The guys who draw the organizational charts must be very busy.