Unity Technologies: please make a game.

A better option for customers, not for an internal Unity team with the role of dogfooding the engine.

Remember, after a Unity release goes LTS it never receives major changes, only SDK updates and critical bug fixes. Things like new APIs, new SRP features, IL2CPP generation changes, and even optimizations happen only in the latest in-development version, to avoid bring back unpredictable bugs to the LTS.

An internal Unity team using the LTS wouldn't be able to benefit from most of the complaints they raise, because any change that isn't a "safe" bug fix will not be made on the Unity version they are using.

Using the latest in development version would allow for a much shorter feedback cycle for larger kinds of changes and fixes, specially if this team has access to the source code to do debugging and experimenting themselves.

This also helps with a persistent issue with Unity's development cycle: Unity relies too much on customers to test and provide feedback for beta/alpha/tech versions in order to reach anything resembling a stable release. But experienced customers increasingly won't touch anything other than an LTS because they cannot afford to put their projects to be guinea pigs for unstable versions with uncertain returns. This creates a negative feedback loop where the less people use pre-release versions, the less stable they are at release, causing more people to stick to the oldest LTS they can get away with, repeat. Internal development teams can help this situation by putting pre-release versions through more realistic scenarios than synthetic test projects.

 

Ok, i will try to express my idea but, in advance, sorry for my english.
That is a thing the unity guy (the one making the other questions) need to clarify (in my opinion) because:
If is a +10hs game, (maybe) i agree with you, because that can help to create a solid and more advance engine for the future, but...
Is is a >2hs game it can be like a demo to show what is the best Unity can do with the stable version everydoby is using, without custom advance weird things, that is my point and i believe that is the point with those questions before, but IDK...

 

I am a one-man paying customer. I have met many others at games conferences who are. And beyond subscriptions, many spend a lot of money on the Asset Store.

And those features are Unity's business as they have an engine used for making games, and making games on mobile includes those features, some of which Unity have features for, but not always great ones.

But basically you are saying don't focus on games as Unity are involved in other areas, and you are saying that in a reply to someone who replied to a request from a Unity employee for information about what games they should make, so your point isn't relevant. Plus, as they are still involved in games, there is value in making sure their games customers are getting the best service possible.

 

we always have these discussion,

BUT to fit the format of a forums, the complaints tend to be generic and abstact,
AND we discussed solution that tend to be generic and abstract!

How about a visualizing what's the experience that lead to those complaints is concretely, by showing specific instance of the unity experience?

Well this video is a good case study, and the person making it is humble, genuine and hold no qualm to unity:

That kinda speaks for itself, the whole thing is summarized in teh first minutes, and the rest of the video show concrete failure from the engine.

The situation with unity now is that: meme developer (like Dani) and small predictable -> valley of death -> experience veteran and college students jump starting their career with designer brand game -> establish studio with code access.

The meme developer are the new "asset flip dev" but with positive spin and impact by building community around fast dev cycle, making simple project where defect are part of the spectacle. However As shown in the video above, there is a problem with scaling to more ambitious project, so a lot of devlog just dies midway through just like unity's attempt with the open project, meme dev work around this by constantly rebooting or jumping arond project, as finished isn't the main income, it's the video spectacle for the audience.

This is a missed opportunity for unity, because the valley of death don't allow team to evolve and ecosystem to mature, ie translate into a new tier of paying customer for unity, allowing meme dev to consolidate and form bigger team.

At the same time, I know for a fact that many veteran, hold by corporate politeness (see video like unity's showcase of dev), don't go hard on blaming unity, but plot to jump ship by making their own engine or switching to another. The only thing preventing them is sunk cost fallacy, but this is waning every time a promise is broken, as other tools become more mature, unity probably won't have a leverage any longer.

So if unity were to make a game, they should go for the jugular, a fortnite clone sum up all the challenge unity falter on. A multiplayer game that work correctly on switch and mobile while displaying a large number of npc units? has persistant features, destructible environment, physics and is now going on the domaion of roblox by expending its creative mode into a proto entirely customizable metaverse? Which itself is becoming a full blown game engine like "core". Unity wouldn't even compete since it would play catch up too late.

BTW when people ask me what engine they should use, I no longer show unity, I direct them to this: https://www.youtube.com/channel/UCEwhtpXrg5MmwlH04ANpL8A/videos
You would be surprise about how portable it is in reality.

 

nice video.

Its painful to watch because what I see is somebody who is clearly more technically knowledgeable than me, but they are struggling with a task that I've solved easily (by using a different engine that solves the problem for you) and they are losing steam because of it.

This is why when people say "game dev is hard" I want to ask, is it? Or are you using the wrong tool for the job?

I see a developer liek in the video who is making a derivative game that will use same tools and techniques as millions of others before it, and yet the person is using an engine that requires you to build common camera collision solutions from the ground up...

Now the fanboys will of course say, "unity is better because it doesn't assume anything about your camera." And I completely agree - if you are a developer whose vast experience shipping games means that your team will actually benefit from writing everything from the ground up to fit your own project.

For any beginners, soloist, or basically anybody who is making a game that doesn't require a bunch of oddball solutions to common problems, I don't think unity is the right tool. It just cost more time and brain bandwidth. Hence people getting this idea that game dev "is hard." I mean it's not a breeze but I dont think it is harder than like... I dunno, any other job.

I get that some people dont want to come to thier beloved software forums and read people dogging on it, but 99% just going to leave and not say anything, and to me that is a lot more cruel. At least say something so 1. others can potentially avoid same problems you faced and 2. developers of software aren't being totally ghosted by their customers.

 

I enjoy Unity for its ease of use. I guess I just like C# and the editor compared to say C++/blueprints and Unreal. That being said, it appears Unreal is doing its best to help indie developers out. I have been messing around with Unity for several years and thought this is it for me. For me world building, level design, art, etc is the pain point. When Unreal let people have Megascans and integrated Quixel right into the development cycle it was a bit of a shock. Plus their other acquisitions. It appears they are trying to help developers, indie and AAA, be able to build worlds. Its all about speeding up the process. Lately I find myself watching more and more tutorials on Blueprints and C++. Unity bought SpeedTree and nothing changed. If Unity had similar options to say Megascans and SpeedTree in a monthly subscription I would pay.

If Unity were to make their own game, they would be able to use the endless amount of paid assets on the store. Maybe for this challenge they have to create their own assets.

Just my 2 cents.

 

Actually he is using cinemachine, which is the built in camera solution FROM unity, and it has a bug that wasn't solved for year, and he don't know how to bypass (he is not making his own camera). More painful are the many heseinbug he had by the end.

The only thing that is directly his "fault" is the shader thingy, which he built himself and have obvious limitation for anyone who is expert at shader, not really his fault that's learning his limits. Everything else looks like unity's though! Tools that break, and compromise the artistic vision, like polybrush and the lightmapper. He was quite resilient and patient on things that should be the reason you used unity to begin with...

 

It is difficult.
If you use better tool than you will attempt even more difficult project .

 

lol, that is so true.

 

That was a tough watch, engine literally falling apart around him. But I'm not sure if directing people to some guy's JavaScript experiments is a viable alternative to recommending Unity.

 

You want to make games with the engine fine. The name game engine has evolved to mean a real time interactive engine and is used by many more industries than just game studios and one man ops. I dare speculate that the majority of cash coming in the door is from non-gaming sources such engineering, entertainment, tutoring/education/edutainment, construction, product viz, military equipment simulators, combat scenario training, commercial training simulators, museum installations. These folks plop down 1500 a seat for many seats and often subscribe to the other bundles at 1300 a pop/year/seat. The reason you do not see these folks on the forums is they are probably under NDA and busy at work.

Unity is a framework. Unreal is a game engine for FPS and RPG scenarios. They are, at the core of it different domains with a different clientele base with some overlap. I would personally rather see them spend this effort on a UnityEngine and UnityEditor core MIDI class so that the tens of thousands of controllers out there can be easily used and swapped. This benefits the community as both gamers and others can leverage their creativity and external hardware to control the user experience.

 

Unity started as a game engine. Companies that ignore what originally made them successful do so at their peril.

But this is not an either/or. Unity is of a size that it can do something for the game side, and do stuff for the other uses. This suggestion is extremely unlikely to affect the chances of a MIDI class.

Also, ignoring the customer who does not contribute much directly in a financial way is a bad idea. Lone developers add to the talent pool that bigger operations can access. I have been asked for advice by several studios considering Unity (by people I'd worked with before going solo). I have given a talk on Unity to a local tech group. I have taught Unity at University. My contribution to Unity is much more than the money I have handed over in the years since I bought Unity iPhone 1.7.

 

“Unity is a framework” is a weird to say it is not good at anything.

 

Many successful Unity games on Steam and elsewhere beg to differ. Clearly Unity's good, at least at something. Otherwise, it wouldn't be so popular.

 

I have a hammer with a broken handle. I occasionally want to nail things to a wall and I use that hammer. I can’t hold it properly, so I sometimes miss, sometimes I hit my finger, sometimes it slips from my hand, but with enough perseverance and personal injury I successfully nail things to the wall.

Obviously, that is proof that the hammer is good, actually.

 

I wonder if some people here see everything as a nail when, in fact, it's a screw. Use the right tool for the job and you won't injure yourself. But it's true that many of the Unity tools work on paper but fail in real world construction work.

 

I'm very noob creating games and programing and i manage to make 2 games (working 3 or 4 hours a day) and working on something more complicated right now, so, Unity es good for making games. You can also see a list of games made with unity: https://en.wikipedia.org/wiki/List_of_Unity_games
The "perfect" software does not exist. I have problem using some of the tools from unity too, but, Unity is very veeery good for make games

 

I don't want Unity to make a game.
I just want Unity to listen to user feedback and make good on it when they do.

 

Clearly you and the moonjump character have no clue as to what a framework is. Of course you can make a game with a framework that is built for realtime interactive 3D..like duh. Clearly you do not understand markets and capital domains neither but knee jerking from a gamer POV. Unity is going to Mars as a series of medical simulators as well. Quite cool stuff..big money they cost..very much like games but are not games but simulation trainers on how to carry out intricate medical procedures in the Mars mission module with only onboard supplies.. The point is that Unity is NOT a game centric engine. It is a framework for game-like applications. A small percentile need a built in leaderboard. Let them write it and learn how to make use of the framework to save and load those stats to their users. Tell me how a leaderboard is of use to an Architecture firm with 500+ seats installed around the world. That..btw..is a real world number there I know of from a 2017 client i was contracted to.

 

Are you ever going to pipe down. It is just simply nauseous and hyper-repetitive at this point. Being disgruntled daily is no way an adult approaches life with any success. Unity is good enough at something with me at the helm that i was able to bank 20K in 2 months and knock the clients feet out from under them. Unity is good enough to go to Mars with NASA and use it onboard SpaceX flights and UI/UX for self driving and smart cars..so..your miasma is somewhat your personal problem. Please do not drag me into it anymore thanks or I will have to speak my piece again..

 

Then change your profession from carpenter to cubicle monkey or something relatively psychologically unstressful so you can enjoy life as it appears you can't use a hammer without bludgeoning yourself and messing up the construction site.. Simple solution.

 

You don't have to speculate. You can see unity's stock yearly/quarterly reports. They don't explain everything, and somethings aren't clear, but the majority of their (~60%) revenue comes from the mobile gaming market (ad monetization) and cloud solutions. Medical simulations/Architecture/Reflect etc aren't the big money makers.
For more info you have the investor webcasts.

 

Honestly starting to get fed up with unity.
I tried cryengine for a very short while and it's painful to see how amazing it's terrain is -- In comparison unity's terrain is utter S***. I won't even get started on SVOGI.

GPU lightmapper is still bad, but hey, they now finally admitted it has 'regressions' and we should fallback to CPU lightmapper for now.

Editor performance is literally slower the higher you go. 2019 is faster than 2020, 2020 is faster than 2021, 2021 is faster than 2022. And it's not some small amount of time you lose, which you might expect with growing software.
it's a massive difference. Random "hold on", play wait time is longer, and so on.
They recently spoke about the slowly dying editor performance over the last few years, and are working on small improvements in 2021 but it's basically caused by a thousand small cuts. Caused by them not caring about editor performance for years. Many small things adding tiny delays, and so fixing it is going to be a slow process.
Starting to think the community did more editor benchmarks than unity.

 

So basically Unity is broken badly according to the naysayers but makes 60% of its cash via games on mobile with this very broken engine. Amusing to say the least when any sort of rudimentary analysis is proffered in regards to the stats and prevailing negative caterwauling. I guess it ain't broke enough that those developers could not make use of it to fatten their bank accounts.

 

“It can’t be bad look how popular it is” and its variation “it can’t be bad, people are using it to make money” are both silly arguments, I hope we can agree on that. Thanks.

 

They sure are bad arguments if it's to counter that Unity doesn't have flaws, but they are good arguments against the belief some people here have that things were catastrophic etc.

Also I do wonder whether the people who got sneak peeks at the other sides "where the grass is greener", have actually dived as deep over there as in Unity.
Since you hardly notice Unity's flaws at the beginning either.

However I think games development and often software development in general is a lot about compromises and finding different solutions or straight up "fake" the visual result you desire with entirely different ways.
At least so far I haven't hit a true roadblock with Unity...

 

Umm... what? I'm not sure what you're on about here. I've never talked about Unity as a framework one way or the other. My comment was about the "not good at anything" part. Many successful games demonstrate that the engine is good, at least at something because Unity games get shipped and sell well on the daily.

EDIT:

Also, that's a pretty bold claim and basically historical revisionism in a thread where gamedevs ask a game engine developer to develop their own game for once so they can see and address the shortcomings of current workflows.

 

It really is!!! Don't know why the team hasn't added basic features which other engines already have like brush mask filters for the placement tools, slope alignment tool, GPU instanced trees, more advanced mass place options, multi prefab placement brush, good and performant terrain and streaming system for gigantic worlds etc sculpting wise it's quite good though.... The world building tools development feels extremely slow polybrush and probuilder had a lot of potential to be powerful tools but those feel abandoned now...there has been no progress of these tools for years

 

I agree I took you out of context and lumped you in with irrational bashers. Please accept my humble apologies. Unity still is a framework. Unreal could not possibly handle the project I am tech lead on currently. It is too game centric.

 

Not really.

It means it is flexible and is applicable to wider range of problems.
Being focused on games means less flexibility and being constrained to specific tasks.

 

Nope. Yer just a gripin' away. I do feel however that if it is such a nemesis to your life that you are living the myth of Sisyphus. Perhaps you should migrate your projects instead of spending the next ten years here gripin' away and rolling that boulder uphill to only have it inevitably rolling downhill to crush your efforts. Bless your heart.

 

I see the arguments are still caricaturing the opposite side.

It's either:
- unity is a catastrophic engine that can't make anything
- or unity is fine people make successful games* with it I don't see what's the issue (*has a lot of pre requisite).

Personally I focus very specifically on the valley of death, in between scale of project and team, to highlight the issue. ie the problem isn't how bad unity is, the problem is very potent pain point compromising dev and creative vision. And disallow the market to be bigger and to spread into new territory, by promoting safe concept around unity's limits.

Basically that's something you see on business culture that get "corporatized" , ie you see a short term logic of survival bias, ie something don't make money or is not used (because it's not mature, therefore nobody use it) it doesn't get priority or get sunsetted, often silently. It's a vicious cycle.

You only get "corporatized" when you have already achieve a certain success, maturity or notoriety, which mean inertia can carry you long enough through bad decision, that look good in the short term, and make good on the resume for bad manager to jump off whenever they are done with.

I mean what is the vision at unity right now? They haven't use "democratizing game dev" in a while. They use to be very forward thinking, shaderlab and surface literally made shader palatable to a whole new generation, which is why unity games are so varied, because they removed PAIN POINTS from writing shader, allowing people to focus on the core of dealing with shaders.

This was pretty much highligthed in blog post about it, it was a reflection on game dev in general, and writing shader in specific, reflecting on how tutorial tells you how to make a single light, but scaling to many wasn't obvious, hence the concept of surface and having a specific light program separated from fragment and vertex. Although it had limits, it made writing lighting much more easier, didn't remove old stuff, and still allow you to roll your alternative solution.

This is very different now, where SILVER BULLETS (right way to do things) are the direction, but they fail to meet reality, add pain points at the edge of the vision, reduce engine flexibility, reduce creative diversity, are short sighted issue without any foresight. Again the reason Nanite and Lumen are such attention grabber, is that they are designed around removing pain points, not offering silver bullets.

My argument of the valley of death is that it will increasingly make a demand to be fill, up until something come and disrupt the market in way that will look obvious in hindsight. I mean minecraft is a collection of concept from very niche games, so was battle royale games. Basically the big success were from place where money wasn't made, until it blew up and made all the money. As a creative tools you want to encourage that kind of underground thinking to reap later. See unity and mobile, unity had success by entering early in mobile, AND STAYING THERE with concrete vision that had foresight, before it was the big money maker, and now they are coasting that.

Saying unity has massive pain points issue, is different than saying it's bad. The hammer with broken handle is very apt to my eyes, especially when they advertised the hammer will also deal with screw, when it doesn't.

What people mean when they say unity should make games, it mean they need to experience the pain points to design FOR it.

 

If Unity were to make a game what type of game would be the hardest to make with Unity?

I suspect it would be:

• Multiplayer
• AAA
• Large Open World
• Procedural Generated*
• 3D Headspace Audio
• Have both DLSS and FSR screen resolution systems
• Use Direct Storage API
• Use Mesh Shaders
• Run at 250 hz on modern hardware

The idea being to push the technology envelope especially beyond the classic Unity tech demo into a playable world.

*Game development has to move to more procedural approaches to balance the workload needs of A+A+A games.

 

AAA isn't a game type, it's a budget.

 

See, people focus on the wrong part of the story of Sisyphus. Before the boulder part everyone talks about, he got to cheat death a bunch of times and once he even trapped Thanatos himself and freed humanity from Uni death.

 

Or a level of Quality.

Blacksmith, Heretic, Adam all have had very high quality levels but with limited scales.

If they can take that level of quality but push it out procedurally e.g. Mega City scales and make it playable then that could be a good game demo.

 

It is a budget, and people use it (correct or not) to also mean what appears to look similar to a high-budget blockbuster. I think we should ask for clarification if a person means AAA in presentation or AAA in budget, if it's not obvious.

It's interesting because from experience working with people, I know many programmers can't really tell the quality of the art or production values of what they're looking at. So for some people, something like Snowpiercer on netflix is AAA and for others, it's pretty cheap and dressed up.

YMMV by assuming everyone will mean or care to use it correctly as a term. I'm rambling because food is almost done. Adios.

 

While we're correcting people for focusing on the wrong part of the story...

The part where he stopped death happened purely as a side effect of saving his own behind. During this period of being "freed", as you put it, humanity suffered immensely because people couldn't die but they did still age and they did still get sick and injured. And it wasn't an isolated incident, he often deliberately schemed such that others suffered for his benefit. Being tricked into choosing* a never ending task wasn't just punishment, it also prevented future, escalating schemes.

So yeah, I fully agree. People do focus on the wrong part.

* The other choice was a normal afterlife.

 

Yes, because the actual interesting part is to make sure you try to defy Unit the gods, in every way possible, so when you finally are way past your prime after cheating death a bunch of times and living to an old age, you can finally live a mundane and monotonous life without any regrets.

AFAIK, all Greek mythology characters that went against the gods are humanist heroes, because the gods were simply whimsical petty assholes.

I’m Greek, I can do this all day, but let’s not turn this into a Greek Mythology forum.

Let’s just stay on topic about how Unity is terrible and everyone who says otherwise is just wrong.

 

@AcidArrow Am sorry to say this, but fear your way of behaving here is rather reducing than increasing the chance of Unity staff actually taking something useful out of this whole thread.
You are acting like a hater right now and haters are by default disregarded by developers.

 

That’s not my problem, that’s Unity’s problem.

But I’ll bow out anyway, I don’t like being used as an excuse for Unity ignoring user feedback, even though it’s just an excuse.

 

I would love to have one though. A forum where we could share or discuss things like this in order to give each other story/setting ideas.

 

I reccommend https://worldbuilding.stackexchange.com/ xP

 

The irony is palpable here.

 

Well, I certainly didn't expect to learn about Greek Mythology in this thread.

 

I support this idea. Look how well Frontnite has been for Unreal. Anyone can builds games but can Unity build a game and have be as successful as Frontnite ?

And if anyone is thinking what for, I would say for A lot reason. To learn what the engines needs, to learn just how hard it is for us developers, to find BUGS, to demonstrate the true power of UNITY, the list goes on and on... I think its a great idea and fully support this thread

 

Fortnite's success is not something you can replicate out of thin air. There's no ongoing new genre trends that could be jumped on the same way right now. Epic had all the assets and the backend ready to whip up a quick BR mod for their zombie co-op shooter to try to salvage a failed game and it worked. There were so many coincidences at the time for it to succeed like that, it's nuts. No one can replicate that.

 

It doesn't have to be a successful game, it just has to be a successful demonstration of what Unity can do.

 

What if Unity just had to make their tech demos playable?

We would have playable:

Unity own Weta workshop so a 3D Tech demo like this is just a few days work/jam.