Much as I love open source and the Godot engine, this seems more like a wake-up call on the importance of corporations not being able to unilaterally, retroactively change contracts in ways that impose new charges and violate existing contractual commitments that others have relied on. If they can do that then open source doesn't offer much real protection.
Open source is that protection. There's no way for an open source project to change their terms that drastically. The whole hasicorp disaster is a prime example of that. Terraform is forked and continues on like normal, users are not impacted.
For programming platforms (like game engines) this is even MORE of an advantage, as with something like terraform you could conceivably rewrite your stuff in a matter of weeks if you have reasonable testing. For a game that is not possible.
Nobody can take away your rights under the MIT license, there's no legal mechanism to do that. You are protected, fully, from shit like this.
Nobody can take away your rights if the contract you sign with the engine developer doesn't allow it, as people have been pointing out is the case for other engines. IIRC Unreal's contract gives you access to a particular version of the engine in perpetuity, with source and the permission to make your own additions/changes to that source. It doesn't guarantee updates, but neither does the MIT license.
Apparently this is exactly what is happening with Unity[1], where the TOS previously said that you can continue using an old version, but now they are walking this back and trying to apply fees retroactively. Whether this is legal or not might be debatable, but unless users mount a legal challenge they are probably stuck paying or finding another engine.
You're not stuck with an old version forever, that's very hyperbolic and you purposefully ignored OP's example of Hashicorp and Terraform and instead responded with sarcasm.
That's only a problem for you if you yourself have a business model that depends on curtailing the freedom of your users. Playing the victim when a license prevents you from victimizing your own customers and users in the way that has happened in this incident is pretty rich.
> Terraform is forked and continues on like normal, users are not impacted.
That's a massive overstatement of the current state of OpenTF right now. Plus they're still dependent on Hashicorp's hosting which also changed their license terms in response to the project.
You're right that because it's OSS you can do this kind of thing where it wouldn't be possible at all in a proprietary system but "just fork" requires a community to organize around it. Without a bunch of backing orgs this wouldn't have happened.
The problem is today everything has ToS, you get a ToS update in the email and if you do not like it you have to stop using the product, and lot of software this days require an account , and has online features or requirements.
I got burned by Steam, I have a super old laptop with some old games on it, Half Life, so one day I got the laptop out, Steam updates and f** itself, the old system is no longer supported but they had to f** things up s the games won't work.
I agree, open source is not required, but we need to own our software not rent it.
> I agree, open source is not required, but we need to own our software not rent it.
That's just a start but also require freezing your OS and have lots of spare hardware or have rights to emulate it :)
And same with data - buyed songs for example :) Big industry already is killing media you can own - cd, dvd, blueray depends on outdated cpu, pendrives decay before becoming usefull backup media...
In the age of asholess open sources and resources are best. Let's bring more viral licenses then GPL !
It is a wakeup-call also to do due dilligence and risk management when entering contracts. And yes, clicking "agreed" in the installer is very often a contract that you should run by your legal advisors. Or at least think very hard about.
_Some_ kinds of Open Source might help there, since _some_ open source licenses are very easy to comply with and very business-friendly. On the other hand, all the game studios now complaining would also howl and whine when forced to be GPL-compliant and release their source code.
And even with business-friendly take-what-you-want-and-never-give-back Open Source licenses, there is always the risk that the project you are using does what Hashicorp did with Terraform and stuff: Change the license for all future releases to something you won't like, cutting you off from your necessary updates and fixes. Maybe there will be a community maintaining a fork, maybe there won't.
You could change the license for future versions and charge for said updates, for the market that godot targets that would be more a little screwing to them.
How long would it take for someone to take over the project (if ever).
And it's at exactly this moment the community would fork the project and development of a free version would continue. This is not a real risk for a community-driven project, only for corporate-driven projects where a single entity owns copyright on the on all or close to all of the codebase.
The important distinction isn't who owns copyright. It is rather that there needs to be a community opposed to the license change and able and willing to do the work.
One could even imagine scenarios like an originally MIT-licensed software splitting into a commercial company offering commercial paid licenses, plus a community (or even the company itself) offering a GPL-licensed fork. Of course one could then still maintain an additional MIT-licensed fork, but if the rest of the community is happy with GPL and all the development just happens there, your MIT fork will "starve"...
While I'm able to understand your argument, IMHO the MIT license is not displaying that well. Community is plural, and fork with MIT could be like Windows: Closed source. End of the (fork) line.
Given the project itself is still strong, this might not be a problem, but then I see no reason why it has chosen it in the first place if not for that specific option.
>clicking "agreed" in the installer is very often a contract that you should run by your legal advisors.
The law often makes use of the reasonableness standard. I'm hard pressed to believe that carefully reading the hundreds of TOS and EULA's hoisted upon us is reasonable.
If you're reading my comment you must have read the TOS/EULA for ycombinator, firefox (or chrome), your wireless or ISP, the keyboard app on your phone, odds are you have an email, perhaps a google account, remember the OS licence. Ever listened to a music streaming service, watched youtube? Messaged using an app? Banked online? Have some managed passwords? Like games, how many? For the common people, what about social media?
That's about a dozen "contracts" and that's lowballing it, multiplied by each update to the "agreement" (pray they don't alterate further) multiplied by the requirement to also read and acknowledge the privcy policy. All this for services that have become when not essential ubiquitous and constantly shift under you. You'd need 8 figures to run that by a lawyer, or a part time job to carefully consider. That's not a reasonable arrangement.
to be more charitable to the above, we're not talking about consumer software. if you're starting a company selling something there is a fair bit more reasonableness in asking you to read a contract
Exactly. This isn't about Joe Blow not reading the ToS for his eleventieth browser toolbar. This is about a business not reading the ToS for an essential, integral component that will cost you dearly to replace and might bankrupt the business.
What you are proposing is literally impossible. If I were to comb through every ToS and EULA and every other legal text related to tools I use as a developer, I would not have any hours left in the day for, you know, actual development. Even then, I wouldn't have time to read and comprehend all of the legal contracts. Hell, if I stopped sleeping entirely and spent 24 hours a day combing through legal contracts, I still wouldn't have enough time to go through all of them.
Are you implying that one needs a crystal ball to predict which companies will screw you over with ToS changes in the future, and using that crystal ball in the present time, we can then skip reading ToS for all those other companies, which will not screw you over, and only read the ToS for the one company that will screw you over?
what use are contracts if buisnesses refuse to read them. Sure you as an employee might not but if you're running a company you really ought to read it on your key software. We're not talking about an email provider, its the key engine you're using. It's like if you didn't read the contract with the manfuacturer you hired to make your product...
Well, in Finland where I live, ToS and EULA are generally not considered to be "contracts", and for good reason.
I'm confused what you mean by saying that the employees of a business don't need to read these legal texts but that the "business" should read them. The "business" is not a physical life form that has the ability to read, only the employees of a business have that ability, since they are humans (unlike the business entity itself, which is not human). Perhaps your idea was that businesses should hire a team of lawyers whose only job should be to read through the various ToS and EULA legal texts that their other employees merely click-through?
in lots of countries, ToS and EULA are generally not considered to be enforcable "contracts"... for users of free consumer software under certain circumstances
If you as a buisness sign up to amazon web services to host your entire backbone on, you better read the contract. If you don't, you are irresponsible.
Someone, somewhere at the buisness should take the time to read a simple little document before using it as the backbone of everything they do at a company, yes. It does not have to be a lawyer...
I'm extremely sympathetic to the idea that there are too many eulas for free products, things like games or whatever where it really doesn't matter for 99.9% of users. But for the 0.1% that make their livelihood from the program, they should take the 25 minutes to read through it at least once.
I'm not saying this because im a big bad lawyer that hates you (I am not a lawyer at all), im saying it because it's a tiny thing you can do to save a whole lot of heartbreak.
You just gotta decide when its important to read contracts and when you don't care. I don't care about the contract when I get new lenses for my glasses, or when I pay for netflix. But if im moving into a new house? I read the bloody contract.
If you had read the Unity contracts before this mayhem, would you have been able to predict that Unity will screw developers over? I for sure wouldn't, because I'm not a lawyer.
> The law often makes use of the reasonableness standard. I'm hard pressed to believe that carefully reading the hundreds of TOS and EULA's hoisted upon us is reasonable.
Yes, but there are two sides to "resonableness": The "what" side and the "who" side. "What" is the thing that should be resonable. But the overall resonableness just as much depends on the question: "for whom is this reasonable?". While a consumer cannot possibly be expected to really read all the ToS everywhere, a business maybe can be expected to do so. Especially for things that are very critical and integral to the business, like the license of that one framework you are building all your software upon. So I do think requiring a business to read and understand the Unity ToS is totally resonable.
They don't though, if any developer had the means to take it to court it is incredibly likely that Unity would lose.
While TOS are rarely worth the toilet paper they could be printed on, I'm curious about whether arguments could be made about whether existing subscribers from 2019 could now sue for breach of contract and costs associated with (re)development.
That's the big issue. This is going to hurt smaller developers more because they can't afford to fight it.
The big question is: to what extent to Unity games need to be able to talk to Unity's servers? If they're looking at number of installs (and apparently that includes pirated copies even?), serve ads, and probably provide other services, that sounds like the games need a connection to the server. In which case they may be able to disable your game if you don't pay. And then even if you could sue them, the real damage is already done.
Did they make a commitment in the first place? It is up to the developers using the engine to make due diligence and inspect the terms of the contracts and licenses of the software they use.
I would feel very uneasy if my product was based on a framework, where the provider makes no commitment and reserves the right to change the licensing terms at any time.
Unity made a specific committment in their terms that if those terms changed in a way that disavantaged existing developers, those developers could carry on using the existing annual release under the old terms, and also that they would notify developers of changes to their terms. They then sneakily removed that committment and almost immediately imposed this licensing fee for use of the runtime, when one of the specific advantages of Unity they promoted was not having any such fees, and retroactively applied this to new installks of all existing games. It was shameless and some company offering open source software could jsut as easily retroactively decide it wasn't open source at all.
Thanks for the info. I'm not into game dev and did not know that. That was a pretty nasty move and they should't be trusted anymore.
The last part I do not agree: a license cannot be changed retroactively if there is no provision in the original license for doing so, and no open source license have that. Even if "revoking" licenses for all prior releases were allowed, it could only work if all copyright holders agreed, which is not practical for most projects.
Copyright laws are all completely out of balance and stupid. As a rule, corporations can not retroactively change contracts. If this one is legal (I wouldn't try to guess), it will be because of the EULA rules.
>> While Unreal engine currently does not have terms like Unity’s, there’s nothing stopping them from doing something similar. In fact if Unity manages to get away with this it seems likely they will follow suit.
Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
Tim Sweeney explicitly mentioned this often + the fact that they are trying to break the Gplay/Apply monopoly shows that the Epic games leadership are not corporate piranhas like Unity's.
But what about Godot? He says as if "it's open source so no issue". Yeah but what if the devs stop supporting it? This "community will continue to work on it" is BS: in reality it's usually one or two guys who actually do the actual work.
So if Godot devs stop working on it you're stuck with a project with zero support. Good luck developing it further instead of focusing on your game.
You're contradicting your own points. If it is not a problem that Unreal can change their license for next versions because you can use the old one (which is now unsupported and not worked on) that is fine.
For Godot, if we were to stop developing it (and I'd like to note that Godot is one of the most active projects on Github right now) you would still have the version you have now.
What is the difference between Epic not working on a game engine you use or an open source project not working on a game engine you use? Except that with Godot at least you could work on it yourself if you wanted to.
> Except that with Godot at least you could work on it yourself if you wanted to.
let me play the devil's advocate - unreal's source is available (despite it not being actual opensource licensed). This means if Epic ever abandons unreal, you could theoretically also just make the changes you need to support whatever your project required - as long as you didn't distribute those changes (except perhaps the run-time? Not quite sure how unreal engine and the runtime are licensed).
If you'd amended the source, then it wouldn't be the same version and the license would be invalid. If you want to use the perpetual license, you can't do anything with the source.
> Epic grants you a non-exclusive, non-transferable, non-sublicensable license to privately use, reproduce, display, perform, and modify the Licensed Technology in accordance with the terms of this Agreement (the “License”). This means that as long as you are not violating this Agreement or applicable law, you can privately use the Licensed Technology however you want. If you want to share the Licensed Technology or anything you make with it, Sections 3 and 4 below address when and how you can do that.
In section 3/4 it goes to that you can compile your game and give the output to outsiders if you pay the roylaties when applicable or share them royalty free on epics github (basically you make a pr to merge your stuff upstream) or on unreal marketplace (sell it)
I'm not intending to be defensive (I'm sure there's loads of people on here with more knowledge about licensing terms), but usually a corrective post is something that we can learn from. Just saying something is wrong with no more info seems against the spirit of HN.
If unreal decides to abandon all older versions support and development, to just focus on something new that requires payment, I don't think you can keep fixing or updating the old unsupported code, even if you have the source
You just can't redistribute the source to anyone else but using it internally for your own projects to build executable that you share (and sell as long as you pay the royalties to Epic) is fine.
The last UE3 game released in 2021. UE4 came out in 2014 and UE5 in 2020.
Unreal Engine license are per engine version and perpetual. This is something Sweeney has been pointing out for years. And again a few days ago when Unity started this shit storm. And as Sweeney points out the big studios/publishers usually negotiate even better terms.
Yes you can. But unless you're a AAA developer studio, it's ridiculously infeasible. At least for Godot, there would be a community effort to maintain it. Such a community would be impossible with any closed-source (even if readable) engine.
In general one very rarely upgrade engines after release. Even during development you usually don't upgrade if there isn't a new feature in the new version you really need.
Really it is only an issue for new projects and at that point if the license of the newest version is not to your liking pick a different engine.
Also the last UE3 game I know of (Them and Us in 2021) was made by a small indie studio not some big AAA studio with massive publishers backing. At that point the engine had been in "end of life" state for 5+ years.
>This means if Epic ever abandons unreal, you could theoretically also just make the changes you need to support whatever your project required - as long as you didn't distribute those changes
No it doesn't, you still don't own the code so you can't just modify it and use it to develop games.
IANAL but I am fairly certain you are legally allowed to modify the source code of proprietary software and use that modified version, so long as you do not distribute it (as you couldn't distribute the unmodified version either).
The key difference between proprietary and Free Software is not actually that you can or cannot modify the source, it's that you are guaranteed access to the source in order to modify it or not. Since you usually cannot access the source of proprietary software, you usually have no legal way to modify the source for your computing.
It all depends on the license. There’s no umbrella ability to modify the source code of a software binary, open-source or proprietary, unless explicitly allowed.
Depends on the market, for making the next pixar movie probably not (yet). However blender usage is way, way up and my game studio (Prehensile Tales) has no difficulty finding extremely talented blender users.
An open source project doesn't have to be the very most used thing from the beginning in order to eventually eat everyone's lunch :)
> Exactly. Blender still isn't the first choice for most animation and VFX jobs
And it doesn't need to be, it just needs to be a viable alternative. Everyone always looks at these projects like it's a winner takes all scenario, but it's not. There can be multiple programs that accomplish the same thing, some open-source, some not, so long as they're sustainable and have something unique to offer that's not a problem.
Sure, if we were living in my ideal fantasy all software would be fully open and free, but in this reality I'm just happy we have alternatives that are actually sustainable and don't feel like you're actively gimping yourself.
I have lived from my VFX freelance work for 4 years and worked soley with Blender for the 3D and much of the 2D part.
Blender has some parts where it is the best (the tracker for example easily beats all commercial trackers I have ever used) and other parts where it doesn't shine as much (e.g. fluid simulation — which is a non-issue because it integrates well with other tools).
I have been coming from 3dsMax and Maya and never have been looking back. Blender also has been getting so much in the past 5 years it feels ridiculous and makes you wonder what the likes of Adobe and Xo are doing with all their money.
> Exactly. Blender still isn't the first choice for most animation and VFX jobs.
It doesn't really need to be; being self-sustaining while being open-source beats Unity's model of being funded to cover operations while being closed-source.
TBH, it's only a matter of time before Blender is a choice for most animation and VFX, and then only a little more time before it's the first choice.
I'm afraid not. Its not as simple as that. We have a heavy competition in automobile industry but when it comes to gaming industry, it is(I mean was) mostly unreal or unity despite decades of technological advancement. The reason for this is because using a game engine to make games is no where as simple as driving a vehicle. It takes a lot of skill, knowledge(sometimes things are engine specific only to make things even more annoying) and the time to attain both of them with the engine they are working on to make good games.
You are grossly underestimating the complexity involved in game engines. It is not like a web app where devs don't have to worry about constraints like memory or frame rate and chill. Things need to happen in real time. A delay of even half a millisecond is not acceptable. And these "Things" involve changing of 3d objects' position w.r.t player's movement, calculating zero or tens or hundreds of NPC AI characters' position and finalizing their animation state, calculating the lighting on all the objects and a lot more. All this just to finally render and present one frame. Yes, hardware has gotten better over the years and memory constraints might not seem like an issue but that is not the case for games. Improved hardware only helps with improving the overall quality of the game. Game now will be able to afford to look better and do more things than games from 2003 and that's it. Games still need heavy optimizations.
Thanks for the better hardware, making 2D games now is neither expensive nor hard. So, Godot being more friendly than unity for making 2D games is quite possible to happen. But that is not the case when it come to comparing Godot with unreal. Unreal is already at a league of its own. I don't think Godot can integrate something like nanite or lumen inside its engine anytime soon. In the past few years, only unreal has been introducing ground breaking computer graphics tech inside a game engine. Unity is having a hard time to even keep up with unreal's tech like meta human. It doesn't matter how many google, amazon or apple contribute to godot, it'd be a big surprise if Godot is at least able to hold its ground against O3DE IMHO.
> 4. How You Can Share the Licensed Technology When It Isn’t Part of a Product
> You may only Distribute Licensed Technology (including as modified by you) outside of a Product as expressly permitted by this Section 4.
> a. Sharing of Engine Code
> i. Sharing Engine Code with Another Licensee
You may Distribute Engine Code (including as modified by you) in Source Code or object code to a third party who is separately licensed by us to use the same version of the Engine Code that you are Distributing.
>Any public Distribution of Engine Tools (e.g., intended generally for third parties who are separately licensed by us to use the Engine Code) must take place through a marketplace operated by Epic such as the Unreal Engine Marketplace (e.g., for Distributing a Product’s modding tool or editor to end users) or through a fork of Epic’s GitHub UnrealEngine Network (e.g., for Distributing Source Code).
Yes you can but still need to mention the original licence and copyright notice. But your end product can be proprietary, or released under other license such as the GPLv3.
But you own the modifications of your code so you can still license your derived work as closed as you want as long you as you obey the original MIT license terms which consists only mentionning it and keeping original copyright notice.
That is what all company selling proprietary products that include MIT and BSD licensed codes do. Usually the jist include a file called "third party copyright notice" with the product ad well as an entry in the "about" section of their gui.
It does matter, since the license of the fork you can make has a huge effect on the long-term sustainability of that fork.
If Unreal screws you, then you can fork it to build some features you need for your current project. If Godot screws you, then you can fork it to build some features you need for your current project, cooperate with others on the features they need which also help you, and start a community for Engine-formerly-known-as-Godot-v2 and invest in it as a thriving basis for projects 10 years down the road.
In an age of locked down mobile OSes and forever changing graphics drivers and such, I don't think "FOREVER" is very long any more.
It's like a perpetual license to a specific version of JetBrains Rider. It's sold as if I own a general purpose tool FOREVER and can opt out of an endless subscription but no, it doesn't mean much at all. It will not work with the next runtime update and past runtimes are obsoleted after a couple of years. They have managed to outsource their subscription lock-in as the release cycles of a third parties while pretending to be holier than thou. If they shipped the volatile parts as open source plug-ins, I would feel differently.
I gave you the alternative: plug-ins that could be maintained by the community. They already have the module system but all the version sensitive .net features are in the core.
For Unreal, I don't think there is a solution to closed platforms if it requires a large/sophisticated team to maintain compatibility. We can only point out that any notion of "forever" is a fiction - you have got in bed with a commercial dependency you may come to regret if they choose to change the rules.
Unity had the same in its license. But they changed it without notifying and when you updated the engine the new license applies even if it's a minor version so Unreal can do the same thing.
> the fact that they are trying to break the Gplay/Apply monopoly shows that the Epic games leadership are not corporate piranhas like Unity's.
Epic is not doing this out of the goodness of their heart, they're doing it because it is beneficial to them. Epic is a multi-billion dollar corporation part owned by a massive conglomerate.
> BS: in reality it's usually one or two guys who actually do the actual work.
Over the last week, Godot had 32 authors pushing 52 commits. Over the last month it had 135 authors. This is not "one or two guys".
> So if Godot devs stop working on it you're stuck with a project with zero support.
Guess what, if a closed-source company decides to stop working on their product you're also stuck, but now you're even more stuck because you don't have access to the source to make your own changes!
It's hilarious that their arguments for the commercial engine are somehow both "you can use one version forever" and "you might be forced to use one version of something else forever."
If you understood the problem then you’d know that their argument is about how Unity’s new pricing change is retroactive, affecting all Unity versions and charging fees to devs for games made years ago.
Unreal uses perpetual licenses for their versions meaning this kind of bullshit behavior is not possible
Perhaps it is you who doesn't understand. We all get that part, and it makes sense in isolation. Yes, that is better.
Then it's goofy af to say Godot might make a license change that necessitates using one version in perpetuity-except a version you and others are allowed to modify and distribute.
You can access the source code to unreal engine when you get a license, but the terms are not by any stretch of the imagination open source. You wouldn't be able to freely share your modifications for instance.
I was going to comment that it’s viewable by anyone on their GitHub, but you are correct: you have to be part of the Epic GitHub organisation to access the repo. Getting access is easy, but you do have to agree to their licensing terms to do so. So yes, what you said.
You don't have to get a "license" (at least in the sense of paying anything), IIRC you just connect your EGS account to your Github account, accept an EULA (which technically is a "license" I guess) and then have read access to the Unreal Engine GH repositories.
(currently there seem to be around half a million users with access)
I'm hailing from an era where "getting a game engine license" involved several in-person meetings between top-level management of both companies, followed by multiple technical meetings between engineers from both companies, followed by exhaustive due diligence investigations from both sides, followed by intense haggling and shady backroom deals for several months, and finally signing a proper "contract", handing over absurd amounts of money upfront and then paying equally absurd amounts of royalties after release.
That's basically what I understand as "getting a license for a game engine" (this is also why Unity got popular in the first place, because they skipped all this nonsense).
OTH I accept probably 5..10 EULAs a week without thinking or even reading the text (most of them are not enforceable anyway).
AFAIK in Germany the TL;DR is that if there's anything in the EULA which violates "German Civil Code (BGB)", then either parts or all of the EULA are invalid. I remember that in Germany an EULA cannot prohibit reselling the software to somebody else, or making your own backup copy (kinda tricky nowadays though where everything is just a "service").
"AFAIK in Germany the TL;DR is that if there's anything in the EULA which violates "German Civil Code (BGB)", then either parts or all of the EULA are invalid. "
There is nothing unique about German law here. The same is true in the USA and most other countries. If a EULA or any other contract (whether agreed verbally, signed physically, digitally, or via a "click") violates the law, it can be considered invalid/unenforceable. It has nothing to do with how the agreement was agreed to, but with what the agreement contains. (There may be some legal theory that may carry some weight that a contract agreed to via a "click" is more likely to be unconscionable that one agreed to with a physical signature, but that does not automatically make all "click" agreements unenforceable.)
I used to work for Epic so I have some inside knowledge, but modifying the engine and sharing it with other licensees (via the UDN forum usually) Its actively encouraged too.
Epic is mainly owned by its founder. They are not primarily corporate though they are influenced by it. They are 100% more trustworthy than Unity or other corporate game companies.
EDIT: To elaborate, this is an article about a company fucking users over after an aggressive growth / dumping business model phase. Your response is yes but this other company would never fuck users over after an aggressive growth / dumping business model phase.
> Your response is yes but this other company would never fuck users over after an aggressive growth / dumping business model phase.
Well, Epic didn't do that when it went through it's aggressive growth and the last time it dumped it's business model. Instead, they changed their licensing to lower fees, and when you look at what they did across the board, made things better for customers and game developers.
Watching this industry for a long time, I've learned that I can never trust a company. Companies are made up of people but the people within them change, and their incentives change. This is very much the case with Unity.
People, on the other hand, I can trust. Not often, but in cases where a person has made a long series of decisions over a decade or more, you can get a feel for what their value system is. Tim Sweeney is in this category. He was involved in Unreal back when I was in undergrad, more than 20 years ago. So I sort of put him in the same category as Gabe Newell and John Carmack: relatively enlightened game business leaders that understand the true value that gamers and developers derive from the ecosystem. None of them are in to make another dollar in the next quarter: they are focused on long term success and the are passionate about games themselves.
So it's not that you're wrong, it's just that your argument applies cynicism uniformly, and I'm not sure that's fair given the history of those involved.
Until Tim Sweeney sells the company, or dies, or whatever. Enter any contract you like, but know that those contracts outlast the people who executed them.
> Epic is not doing this out of the goodness of their heart, they're doing it because it is beneficial to them. Epic is a multi-billion dollar corporation part owned by a massive conglomerate.
Holy straw man! And there’s certainly no self-interest involved with me using their engine without giving them a cut under 1 mill right?
You can make this argument about literally anything.
Godot must have no self-interest in releasing their engine for free and making $30k a month to develop their project.
Who cares what the motivation is if it gives real benefits to devs. If anything it tells me they’ve picked a business model that’s mutually beneficial unlike Unity
While you can make this argument about literally anything, it seems apt to make it about a multi-billion dollar company chasing a trillion dollar company.
It's fine to support them while they are the underdog and are saying and doing the right thing, but don't pretend like they will say and do the right thing forever.
I don't think the parent was ever saying they'll be good forever. More that if they do become evil, their licensing lets you continue using a revenue structure that made sense at the time you chose it (even if new versions have super awful terms).
> Who cares what the motivation is if it gives real benefits to devs.
Motivations matter, because they will decide in which direction a project moves over time, or if situations change. Epic is the "good guy" now, but only because they are an underdog. If they become the dominator, they might become the predator which Unity tries to be at the moment.
But if we are honest, such uncertain possibilities don't matter if they are so far in the future. This might be a problem for future games, 10, 20 years down the river..
> Epic is not doing this out of the goodness of their heart
Tim Sweeney founded Epic and wrote the Unreal engine himself. John Riccitiello’s whole career is in management. Even if they’re both entirely motivated by profit, they have different perspectives on how to get there.
Also, Unity is a public company, while Epic is private. Even though Tencent owns a considerable share of the company, Sweeney still holds over 50% ownership. That gives them very different incentives.
Sometimes we forget that executives are people too and they have their own personalities. Tim Cook famously[0] told climate change denialists in a shareholder meeting that “If you want me to do things only for ROI reasons, you should get out of this stock”. Sure enough, shareholders are also not solely motivated by profit, and voted with Cook on that occasion. It’s useful to remember that cynicism isn’t the same thing as realism.
Eh. Part of apples brand is “doing the right thing” and “being the good guy”. I would wager Apple’s brand would be hurt more by not being a green company. ROI is more complicated than simple fist level cost. Going green and doubling down on recycling has generated a much larger ROI than doing nothing.
Further more, Tim is bound by law to do what is best for the shareholders. Simply put, if Tim favored environmental concerns over profit he would be removed.
"Further more, Tim is bound by law to do what is best for the shareholders."
This stupid meme needs to die already. There is no such obligation, he only has a fiduciary duty to not trash the company and spend the earnings on cocaine. "companies are legally forced to maximise profit" has never been true and this piece of misinformation has been going around for ages now.
> "companies are legally forced to maximise profit" has never been true
It's more like too hard to be proven in any way. Unless you live in an simulator it's really hard to say which set of decisions is better than another. People often say it is obvious or in hindsight but fact is there are no such hard proofs.
Even then, if the shareholders approve of trashing the company with ice cream parties (had to get rid of the illegality of cocaine for this point) there's nothing inherently wrong or illegal with that.
As long as the executives are behaving generally how the shareholders want, it's not a problem.
There's still argument about this. The oft mentioned Dodge v. Ford Motor Company 1919 covers much the argument for and against. But it's clearly not straightforward.
My (IANAL) reading of it is that maximising shareholder value is probably the law, but it's practically unenforcible. Being practically unenforcible doesn't stop CEOs and boards from using it as a guiding principle.
Dodge v. Ford Motor Company was a Michigan State decision, so even if one thinks it means maximizing shareholder value is the law (it really didn't say that, broadly), it only applies in Michigan.
Cook isn't bound by that true enough, because he remains the majority stakeholder, but that is increasingly not the case as it becomes more and more regular that companies bring in new CEOs from entirely different companies if not entirely different industries, who do not own that much stake. In those cases, the board and shareholders can and do exert a lot of influence, up to and including firing them if they do not do their jobs correctly, which to shareholders is invariably some form of "make line go up."
And that's just civil influence, there are legal mechanisms indeed in place if a CEO "trashes a company" and what that means is different depending on the company.
And, insofar as such an obligation to “maximise shareholder value” might exist, that obligation doesn’t necessarily translate into “maximise profit”.
The shareholders of a theatre company might care more about breaking even while getting an interesting assortment of plays produced with a great cast than they do about making a bunch of money out of the venture, so an executive who makes a bunch of money by running productions of uninspired cash grab shows won’t actually be maximising value. Likewise, I’m sure that Rob McElhenney and Ryan Reynolds care more about Wrexham AFC’s managers getting good athletic results than they do about making a bunch of money.
>And, insofar as such an obligation to “maximise shareholder value” might exist, that obligation doesn’t necessarily translate into “maximise profit”.
And even where it does translate into "maximise profit" because it's what shareholders of a particular company may want, there is no timeframe for it, and there is no way to tell whether any particular decision by the CEO runs contrary to the goal of eventually maximising profits.
Companies can spend all their revenues plus a constant stream of new capital on growing market share or revenue, on charitable activities or the happiness of employees, on huge research and development projects or on restructuring after restructuring and still credibly claim that all of it is ultimately meant to maximise profits.
The point where CEOs and CFOs have to be careful is when the company faces solvency issues. That's where legal limits of freewheeling decision making kick in, because it's where it's no longer about shareholders but about creditors.
"Best interests" is deliberately vague. Is it in their "best interests" that you fire the CEO and call the cops because he raped an intern? Or maybe that you agree to accelerate his $40M "bonus" if he agrees to resign without saying why? Or maybe it's in their "best interests" that you pay PIs to develop evidence that the intern has a drug habit and "leak" this "shocking revelation" to the media if they go public?
You can argue almost anything meets this criterion, in some egregious scenarios a court won't buy it, but they will give you enormous leeway.
People write this sentence seeming to imply it means "CEOs and management have a responsibility to be as ruthless and sociopathic as possible to deliver the highest returns, and any consideration of the people or communities they trample beyond legal requirements is itself borderline illegal."
There is a lot - A LOT - of room for ambiguity and debate on the specifics of shareholder value and "best interests." The "legal constraints to act in the best interest" is not some set of corporate rules and KPIs codified into our legal code write large. It's not about maximizing a specific KPI over a fixed timeframe.
Not to mention Sweeney is the majority shareholder in Epic's case.
It’s a private company… to whatever theoretical extent he’s beholden to the majority of the shareholders with potential board of directors and shareholder meeting shenanigans … that majority of shareholders is himself… and I’m pretty sure he’s ok with his own decisions…
If Tim Cook had calmly said what he did, I’d believe it was just toeing the party line. The fact that he actually got angry, though? He doesn’t strike me as the sort of person who’d fake that, I don’t think that’s the image he wants if nothing else.
He doesn't have to fake it. Dude probably believes that there's no connection between right to repair, unsustainable consumption, and planetary destruction. That might be worse than faking it.
iPhones have long support cycles. iOS 17 supports the iPhone XS/XR, which were released 5 years ago. So iPhone users don't have to replace their phone as much, reducing environmental impact. Plus with the recycling and other stuff used to make iPhones in the first place, their environmental impact is going to be lower than some other options on the market. Combine these two things together (longer device life, smaller impact for making a new device) and I can believe iPhones have the smallest CO2 footprint.
Phew, there I was thinking Apple do their best to make sure customers buy the new phone every year, but glad to hear it's not happening thanks to Apple.
I mean, Apple should just release a new model every five years and really get that footprint down...
> Over the last week, Godot had 32 authors pushing 52 commits. Over the last month it had 135 authors. This is not "one or two guys".
Yeah, right now (and I hope it keeps increasing thanks to the Unity news).
I got burnt once (and almost twice) from this in the past. cocos2d-iPhone used to be huge, and Zynga even contributed to it. I released two games using it, and started another one. And then it stopped getting updated (and Apple keeps breaking things like Apple likes to do), so it died on the vine and I had to port to something else.
Currently making a game in Monogame, and while it did get a significant update a little over a year ago, it's had very little activity on it since then and no other releases besides a hotfix shortly after, and zero communication about what's going on with any of their official channels at least. Not great especially since there seems to be spotty support for .Net 6 outside of Windows still, and I keep running into various issues with its 3D support (which it is mainly supposed to be a 2D game, but it does support 3D to a certain extent).
Also while it claims to support platforms like PS4 and Switch, I see almost no documentation on it, and very little documentation on getting Steamworks working with it (I have some basic things working, but I'm having to figure things for Unity first and then porting that knowledge over). Also most of its multiplayer libraries seem outdated, at least the ones I looked up. And the forums/Discord still have some activity on them, but not a ton. Also I'd love to make a game that supported VR (almost switched to Unity just for that alone).
I actually compared its Github activity to Godot a week ago, and Godot looks SO much better supported than Monogame at this point, that I was already considering either porting or making my next game in Godot before this Unity news was announced. But maybe that's just me hopping onto another platform that will have the same problem in 5 years after I've gotten pretty invested into it.
I also know Unity (worked on a game professionally years back) and considered switching to that for my game too, but that mostly got killed by this announcement.
Personally I prefer code-based development (I don't like using an editor too much), so if Monogame was better supported and had much better 3D support I'd probably just stick with it. I thought I was going to when I first started using it seriously three years ago. But I hesitate to keep dumping time into it if they don't maintain it.
Also I made a small game with Phaser.js and was going to do more with that, but even that creator is pretty much the only one maintaining it and they got sidelined by life for about six months (which is fine, but get someone to help keep it going if you can! I know that's hard though). Phaser at least is already close to what I want from it, and doesn't seem to need much else for the foreseeable future.
> the Epic games leadership are not corporate piranhas like Unity's.
the Epic games leadership are bribing developers for exclusivity, including kickstarters where i was promised steam or gog
the Epic games leaderhip makes money from free to play grinding/gambling simulators
the Epic games leadership also comes from Tencent that made all their gaming money before trying to get into the western market from asian gacha
The only reason their terms are more reasonable now is because they're the underdog and throwing a lot of money made from free to play crap at fixing that.
Bribing implies an illegal act. Giving money in exchange for goods and services is completely normal: I don't bribe a restaurant for food. Exclusivity is just a way of competing for games, a sign of a good competitive market.
Kickstarter is kickstarter. Half of them probably don't fulfill all their promises. That's the risk you're taking, and why it's not a preorder.
Epic don't have gambling. They've even removed gambling from Rocket League after purchasing them too.
> Bribing implies an illegal act. Giving money in exchange for goods and services is completely normal: I don't bribe a restaurant for food. Exclusivity is just a way of competing for games, a sign of a good competitive market.
It doesn't have to be illegal, merely immoral.
> Kickstarter is kickstarter. Half of them probably don't fulfill all their promises. That's the risk you're taking, and why it's not a preorder.
I don't expect them to finish a project when i back them. I expect them to deliver my fucking GoG key when they finish the project and ask me what do I want it on and i say GoG! I don't expect them to say "Epic paid us a ton of money and you can get it on their gacha financed store".
True story. So long, Julian Gollop.
> Epic don't have gambling. They've even removed gambling from Rocket League after purchasing them too.
They're free to play. That means the game is designed to keep you playing forever and keep you buying IAPs. It is not designed to entertain you.
Releasing on a store is not immoral. Is not releasing on GoG what you believe to be immoral? Even GoG has paid for exclusives, they charge store fees like everyone else. They're a business too at the end of the day.
That sounds like an unfinished feature to me then.
Why are you playing games forever that aren't entertaining?
> So if Godot devs stop working on it you're stuck with a project with zero support. Good luck developing it further instead of focusing on your game.
>
> The same goes for his very own product.
> the fact that they are trying to break the Gplay/Apply monopoly shows that the Epic games leadership are not corporate piranhas
That’s a very unique take. I would call a company that lets others take startup a buildout risks and the uses the courts and regulators to swoop in with a low cost competitor a pirhana.
It’s a good strategy (for Epic), but it is extremely predatory. It’s no secret they have their eyes on the console market next; let HW makers popularize platforms using a business model of low margin HW and high margin SW, then get governments to mandate alternative stores so Epic can undercut the SW.
Low risk, low effort, high return. It’s a solid business strategy but if it’s not pirhanic, I don’t know what is.
> Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
If I understand correctly, Unity had that too. And then they changed it anyway.
I don't know the details, but apparently Unity maintained their license in a git repository for the explicit purpose that everybody could easily track changes to their license. Just before the license changes, they removed that repository, and later put it up again, but without the clause that you could always use an old version forever without new restrictions applying to you.
Any product can end up no longer having work done on it, or the work can start getting bad. Such is life. With open source at least you have an option, if you run into one bug or missing feature, there is a chance you can patch it yourself and get the game done. In the distant past I wanted to use Unity and ran into a breaking bug in the particle system that wasn't fixed for at least 2 years, if ever.
>>>"""But what about Godot? He says as if "it's open source so no issue". Yeah but what if the devs stop supporting it? This "community will continue to work on it" is BS: in reality it's usually one or two guys who actually do the actual work.?"""
Exactly.
We like to hate corporate greed and lionize the open source developers. I also hold open source devs in high regard.
BUT.
Open Source Dev have to eat too.
If we want to keep the Open Source ecosystem moving, we do need to find a way to pay them. Even today, a lot of open source projects are supported by individual corporations that keep the devs on payroll to give them time, but that can also lead to influence and lock in, and even abandonment if desired.
Yes, unless you build everything from the game engine to the distribution channel yourself, there is a cost to doing business. IIRC Epic Games Store offers a more reasonable 88%/12% revenue split, while Steam uses the same rapacious split as Apple and GOG at 70%/30%.
>Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
Maybe _you_ didn't research how Unity had this same exact clause and decided to just... remove it. The author is implying Unreal could do the same at any point in time.
> Tim Sweeney explicitly mentioned this often + the fact that they are trying to break the Gplay/Apply monopoly shows that the Epic games leadership are not corporate piranhas like Unity's.
Haha this made me laugh. If you think epic isn’t piranha then you’re living in a bubble.
Epic's still remains. Yes, if Epic removes this feature of the license, we can be concerned. But, Epic's license means I can remain on the previous license that has these more favorable terms.
> Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
Isn't part of the current furore that unity used to have similar terms and they removed them?
> Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
> Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license.
Unity also had a similar clause in their ToC, they removed that now and retroactively applied the new pricing model on ALL previous versions of Unity. Unreal can do it too.
This still seems to be an unresolved legal question. Unity will almost certainly be getting sued for this by parties still using versions that shipped with that clause.
> Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license.
Even if this is 100% legally true, they can absolutely say that they are revoking all old licenses, and then if you used one, it's up to you to take Epic to court to prove that they can't.
> So if Godot devs stop working on it you're stuck with a project with zero support. Good luck developing it further instead of focusing on your game.
Exactly I'm glad someone is pointing out the problems with these piddly open source projects. Companies are too reliant on open source software, I mean it's not like open source tooling has ever taken off. Except for Emacs, and Vim, and VS Code, and maven, and ...
But seriously though the over reliance people have on open source projects is staggering, I mean really what happens when the guy who maintains Linux gets bored and wanders off, everyone is boned.
Therefore I agree with OP all open source projects are inevitably doomed to failure and can never work.
There are already well over one or two Godot devs working on the engine. The community funding is already enough that it supports multiple full time devs salaries. If the current devs quit, this funding would be used to hire new devs. At this point, I think the Godot community is already well over the tipping point where it will continue no matter what the original devs decide to do.
> There are already well over one or two Godot devs working on the engine. The community funding is already enough that it supports multiple full time devs salaries. If the current devs quit, this funding would be used to hire new devs. At this point, I think the Godot community is already well over the tipping point where it will continue no matter what the original devs decide to do.
Yeah. They've reached critical mass to, from this point on, be self-sustaining.
Why would you need godot to be "developed further"? If your game is shipped, it's developed. You can continue doing this. When the source is private and behind a license you are forced to agree to the new terms when the old ones expire, or if you can't agree to them then you can no longer ship the game. These are not equivalent circumstances.
I'm less concerned about Unreal pulling a Unity. The bigger issue is just a lack of attention. The last numbers I saw was that Unreal was pulling in $100m a year for Epic. That's a drop in the bucket compared to what Epic makes off of Fortnite. In my experience, products that matter a lot to the consumer but not the company tend to stagnate. Just look at Google outside of search.
The worst case scenario for Godot is the same as the worst case scenario for Unity: you can keep using the last release that worked for you. As far as I can tell Unreal has no concept of an LTS version, so if you don't like the new license you're not getting any updates at all.
This is where FOSS comes in: If support ends, you can keep using the latest version just like with proprietary software, but you can also fork it and fix pressing issues. It's not even a question of relying on the community continuing to create major new releases. If that fails, your org could fork it alone and patch any blocking bugs.
My company already does this for several pieces of legacy software that we haven't had the time to migrate away from. We don't make any major changes, but we can and do fix things that get in our way. We'd get no such benefit from proprietary software that changes its terms to be unfavorable.
I kinda feel bad for Unity. The company is in a bad place finacially. They're burning almost $1 billion a year and they have no good solutions to this. Either they try to increase pricing and piss off their customers, or they cut costs (which they've been doing with mass layoffs) and risk losing ground to other game engineers like Unreal.
For years the only thing really holding the company a float was their valuation. So long as investors were willing to give them money for growth and future profits they could continue to fund their loses. But now their growth is slowing and investors are less willing to pay up for growth generally what does Unity do?
I've felt for a while that they'd be a likely victim of this most recent tech rut and that seems to be playing out. They have no moat, slowing growth and are burning huge amounts of cash. Unless the macro changes in their favour it's hard to how they get out of this and stay on top.
Not Weta FX but Digital’s Tools, Pipeline, and Engineering.
"Weta Digital’s Academy Award-Winning VFX teams will continue as a standalone entity known as WetaFX under majority ownership by Sir Peter Jackson and helmed by CEO Prem Akkaraju."¹
Because to make the next Fortnite takes not just a lot of money, but luck. What if they made a game and it flopped (a techinically impressive game can definitely flop)? People will be like "oh see Unity engine is so bad, even the game from the first party doesn't sell".
as someone who has played so much of these two games: both are incredible games, in their own way - creating/testing maps, the weapons, the movements, the game-modes, the moddability (skins, sounds, etc)...so good
UT99 and all the Unreal games have been delisted from stores and they're shutting down the game's master server. You can still play solo or in LAN but online multiplayer is gone and there's no (legal) way to get new copies.
In some ways this is backwards, you could argue Fortnite was a flop and became a success because of their license conditions (pivot to cloning licensee PUBG).
Unreal has had much more favourable license terms since ue4 in 2014, which is before Fortnite and the Store. To the best of my knowledge, unreals terms have always been "good", which is one of the reasons it's so popular.
Honestly, if they just added revenue share like Unreal does and tweaked their subscriptions a little, the backlash wouldn't have been nearly as bad. The uproar is a combination of a really badly announced system and details that don't seem to be ironed out properly leaving devs ripe for abuse. Plus you have one of the most tight knit communities of artists who all reacted almost as one to this announcement and have experience with abuse from gamers in terms of piracy, review bombs, refund waves and so on. I guarantee most developers would have grumbled but given in to a revenue share
I'm also completely baffled how poorly thought out this whole thing has been. Unity has been used by the likes of Nintendo, Microsoft and other massive game development studios. Do they seriously think a fee applied on retroactive sales and revenue numbers would be accepted without issue from them?
> Do they seriously think a fee applied on retroactive sales and revenue numbers would be accepted without issue from them?
Unity have not proposed a fee on retroactive sales. They have proposed a from-this-point-forward fee that applies to new sales/installs of any game made with Unity, including new sales/installs of back catalog games.
Sales and installs don't happen at the same time. If I buy a game once on Steam, I might install it on any number of unique devices. Per the Unity FAQ, every one of those installation events results in an additional bill for the game's developer.
In other words: This is absolutely a retroactive rug-pull on new installs of old sales.
Update: Because of the way they're rolling this out, the only way to avoid the retroactive license change is to immediately stop using Unity's development tools. If I were running a game studio, that's what I would do.
> Honestly, if they just added revenue share like Unreal does and tweaked their subscriptions a little, the backlash wouldn't have been nearly as bad.
They ran the numbers. If it was going to work they would have gone that route, but I think the problem is that the clear majority of Unity-users (indie devs) make no money anyway.
Revenue sharing with someone making $0 in revenue is pointless.
Okay, but when we're told to share our work for free "for the exposure", we're supposed to buy that and just give it away? They wanted people to use their product for free for the exposure. Changing their mind is one thing, making it retroactive is another.
I wouldn't feel bad for their executives. They are the ones who chose the 'growth at all costs' path, and this is what you get.
They could've probably been sustainable and profitable if they mainly serviced their core audience of indie developers with a smaller amount of employees and a simpler product, but it seems they really wanted to brute force themselves into the AAA market.
I suppose. There's quite a few companies in Unity's position right now for a reason though.
For better or worse the "growth at all costs" strategy was the strategy companies like Unity had to employ for the last decade to attract investor capital so I tend to blame the low interest rate "easy money" environment more than companies like Unity being reckless.
Although that said, it does seem Unity made some stupid moves in recent years. I don't follow the company that closely, but I'm aware they made some large acquisitions funded in-part with debt during the pandemic. It's one thing taking on debt if you have a strong balance sheet, but it seems a little short-sighted for a company losing billions to use debt to fund an acquisition – especially at the elevated prices they paid during the pandemic.
> I tend to blame the low interest rate "easy money" environment more than companies like Unity being reckless.
As if the founders had no other options in life than to start companies relying on business models where you capture the market with investor money and then once you’re customers are locked in, you squeeze them as much as possible. Unreal is in the same market. They also took in investor money. They’re not in the same situation.
The people behind Unity chose to play the game. The c-suite is handsomely rewarded, and we praise their business acumen when things go right, but if things go wrong suddenly they’re victims of circumstance.
You can’t stay small if you took a lot of investor cash. This is why self-funded companies are usually the ones who avoid these awful growth/layoff cycles.
"core audience of indie developers" I thought that the engine was hugely popular beyond games? from what I recall indie games were just a small portion that used Unity.
Why feel bad? They're a victim of deliberate bad choices. The buck stops at the top. Nobody told Riccitiello to hit the gas pedal, nor to enrich himself by selling shares prior to a disastrous announcement, he did that all on his own.
I feel much worse for the thousands of developers he's holding a metaphorical gun to with this awful policy, who now have to stress about swapping engines.
How many such game development companies actually are there? I see a lot of aging engines out there, and companies that jumped to unreal, but I have by no means broad knowledge
Certainly not an expert but top studios like Bethesda and CDPR maintain their own engines with an order of magnitude less employees. They make games too.
They also produce games that are very similar to their other games, with similar mechanics, while Unity has to support a far more diverse set of games (basically anything 2d or 3d), and presumably has to continually offer a wider set of cutting-edge features.
Given the state of Creation Engine, I wouldn't say that Bethesda maintains their engine well. CDPR is switching to Unreal. CP2077 launch issues were mainly due to their engine not being capable of handling such large games.
Gaming companies get a lot of crap for this - it seems unfair. The OS I'm using to write this is a relative newcomer at only 30 years old. Obviously, it has changed a lot in that time, but so have the engines. Most software doesn't get rewritten, it evolves.
To be fair, the difference here is that Unity is doing more than just an engine for a specific type of game. Rather, they are building lots of different tools for lots of different games.
For example, Larian (BG3) and Wube (Factorio) each have their own engine. They are specially built specifically for the games they are making.
Also, you say "have been kept modern" but even that is questionable. "Modern?" What does that really mean? It's "modern" enough for the game. Starfield was just released, and it was released without ray tracing.
And we are just talking about games. Unity and Unreal do more than just working on the engine for games. So yes, while it's a lot, it's not fair to compare the bespoke engine use by game companies and engines like Unity and Unreal.
Epic Games only had about 2,000 in 2020 and they also develop games, store front (although really slowly), so it's at least a 4x more than they need.
But also their previous monetization seems to try charge per professional developer which is a limited audience compared to consumers, especially if you want to maintain a AAA engine (which it seems Unreal is favoured still) given it seems fast pace techniques & improvements. Maybe by selling cloud servers for networking, which I think maybe they were too expensive compared to alternatives.
I’d probably bet that the majority of games on steam have less than 1000 sales, so being a popular game on Steam probably doesn’t mean much compared to being a popular game by daily active users viewing unity ads.
The methodology on the site only includes games with a minimum popularity, so the long tail of tiny indie games isn't included. By the way, since steam has a fee to be included, aren't most of those tiny indie games over at itch.io anyway?
> Unless stated otherwise, we filtered out unreleased games, free games, those that launched with less than a $4.99 price point, and those that have fewer than 50 reviews.
That's not their focus at all. They're trying to be an all-things-entertainment company, currently, they are focusing on movie animation and digital character creation.
May I ask you if you ever heard of Rust? The game, not the programming language. Online survive game, which has thousands of active players every single day since like a decade already, full 3D and quite awesome graphics I'd say. It's written in Unity.
While I don't know the Rust code, this isn't necessarily solely a Unity problem. Unity's C# API provides a lot of footguns that game developers tend to not think about in the near term. Lots of code bases I've seen in Unity don't do a good job of caching component instances, or just never opened the profiler to see their GC is off the charts. Unity can run really well if the proper thought and care is put into it.
> They're burning almost $1 billion a year and they have no good solutions to this
Maybe they should have something to show for this $1B/year and take a cut of this $250B+ industry to cover the costs? They appear to have about %30 market share, so they need to capture less than %2 of the value created with their tool to break even and if they can't do that or they are providing tech for the less than average profitable part of the industry they should shift focus or reduce costs.
AFAIK it's only natural for businesses to go out of business if they can't capture more value than they consume.
None of this is attributable to engines, it's the size of the products built using the engines and it appears that the ad revenues are not included and that appears to be another $80B.
The game engine market is the market of supplying the game developers with the tech to build their games. That's where the game engine makers that spend 1B per year and hold %30 of the engine market need to charge about %2 of the products made using their engines to break even.
Unity seem like a very valuable acquisition target. Any of the American or Chinese big tech companies could put Unity to good strategic use plus a few medium sized tech companies like Adobe and Sony.
My money is on Meta. Unity powers the vast majority of VR games and is (was) the favored API for Meta's dev tools, and even Apple chose it as their only officially supported 3rd party toolkit for the Vision Pro.
Meta seems to enjoy throwing money into the VR space to subsidize it at the low-cost end, I could easily see them doing a hostile takeover of Unity "for the good of the VR space"
Apple “chose” it might be a bit misleading. While I’m sure they could’ve gone with Unreal, the outstanding lawsuits alongside the very loud negative press from Epic made it a non starter. Outside of Unreal and Unity, there aren’t a lot of options.
Losing a billion means they spent more than that, but on what? I haven’t used unity since 2016 and never in a professional capacity but I can’t imagine anything significant was improved or added in the last year.
I'd love to read more about "game engine development, in general, moves at breakneck speed," if you happen to know of a blog post or something or might be willing to share more.
Some of the stuff looks pretty cool that they are adding but would probably take a retooling of many workflows to use correctly. They had some very compelling items they were adding in that would make people pick it over others. But with those license terms. That will be a hard pass by many. My guess is they are hurting financially but took cues from their weta tool stack for installs. Where in that industry per install cost is decently normal and just rolled into a production budget. But on the games side that is not going to fly. As it is a one time cost vs recurring. I feel bad for the shops where they are 2 years in with this thing and just had the rug pulled on them.
Maybe Godot is similar enough where you'd be able to consider this. Otherwise this move is more harmful in the long term of Unity and it's about future games and not current ones. Long term matters more, always.
It may also make AAA invest away from Unreal as it looks like a monopoly now.
Do feel bad for them but this is a problem of their own making. Many parts of their current situation were avoidable.
That doesn't justify what they've tried to do here though. This isn't just a price increase, it's a significant price increase, and a poorly thought out "revshare" model that applies retroactively people who signed deals with them and built a business model off of the deal they signed.
Blender is also managed exceptionally well for a FOSS project, systematically polishing up rough edges and paying close attention to the needs and desires of its userbase which no doubt inspires larger donations from more donors than FOSS projects usually have.
Small nitpick: AIUI it is possible to sell binaries of a GPLv2 program, provided you keep providing source for free. (I think; IANAL, there's maybe some caveat about exactly when you have to give source and to who.) That can actually work if your users don't want to compile stuff themselves.
In theory. Show me one example (not service or support, but selling actual software). You have a better chance of winning a lottery, happens every day.
In theory, I can just walk through solid wall, using a quantum tunneling effect.
The closes thing you will find are things like blender release under GPL (i.e. pay me a money to release my commercial software under GPL, but that's not selling GPL software),
Maybe they should stop acquiring a continuous stream of startups that have nothing to do with their primary mission? I would bet only a very small proportion of Unity employees work on or even adjacent to the Unity engine.
They spent like 6 billion on acquisitions that don't help make their core product better. Literally just don't do that and you'd be in the green right now.
Yes, and now they are burning tons of money due to that debt and having to make ridiculous decisions to appease the stockholders. You just stated the exact reason this whole thing is a problem.
Unreal is technically orders of magnitude ahead of Unity, and they have around 4000 employees spread across multiple continents. And I'm referring to the entirety of Epic, so that's also the Fortnite teams and everybody else. Unity, by contrast, has about 8000 employees. Many companies seem to be hiring far more employees than they realistically need - often to the point of their own detriment, and I don't entirely understand why. Even for successful companies like Google, it seems unlikely that they need anywhere remotely near 180k employees.
Everything you wrote paints a familiar picture: a company whose leadership made poor financial decisions over and over. I won't judge anyone for mourning the loss of their favorite corporate entity, but I personally find it hard to feel pity knowing that they dug their own grave.
As Game developer myself, and unity user, I was concerned about the new pricing model, however, actually reading the actual changes[1], you see that 99% of the developers are not affected, and in some cases, for those affected, still cheaper to go with Unity than with Unreal: There is a video explaining it well https://www.youtube.com/watch?v=ENoVL68z9PU
For sure an open source solution would be more than welcome, but sounds like impossible, because either those out there are half-backed, or are just finding a way to monetize it too..
I've been watching a few gamedev streamers port their games to Godot and it's clear that it's not there yet in terms of feature parity. The good thing is that the Godot devs are in the chats, posting (unreviewed) PRs that could be used as workarounds but some major things (profiling?) are miles off still.
> By using an open source engine you can be sure that whatever that “next thing” is, the engine won’t keep you from taking advantage of it. Nor would the engine be able to dictate your monetization strategy for you.
We really ought to separate the fact that "open" does not always mean "free" here. Especially given that it is fairly straight-forward to change licenses on a whim.
Imagine you wake up one day and suddenly the company/OSS maintainers put out a dual-license on the project. You now have to pay $0.20 per install after your app has made $200,000 over the past year. Now you're still using open source, but still legally held to the terms of that license.
Licenses tend to be permissive enough that if such thing were to happen or other factors(discontinued development/support/etc), community forks can be created(which has its own set of problems).
That's not possible with something that is open source. If it has a "dual license" that means you can pick which license to apply. You just pick the one that doesn't make you pay.
There are a number of projects doing this today with dual-license models(Permissive/Commercial). I'm not a lawyer and have no idea if they hold up in court/can be enforced but it absolutely is a thing today.
These are a little different. So these "dual licensed" products work like this:
You get a GPL version of the code, which is free and nobody can charge you for, ever.
If you want to use the software in a way that is not GPL compatible you can opt for the other, paid for, license.
This works in some cases, but in the case of Godot (assuming we were to do something like this) is the MIT license, which already gives you the rights to do whatever you want.
And even in the GPL case, if the other license terms became too odious you could simply switch to the GPL version, and not pay.
This is specific to GPL-style licences. You would have to meet the terms of the GPL anyway (i.e. provide your modified source code under the same licence).
The commercial licence is just an additional option for companies willing to pay to not adhere to the GPL terms.
There are certainly cases where maintainers have started licensing _new_ versions of a project under different, non-open source terms (Terraform, ElasticSearch etc.). But you're free to continue using any code that was released under the old licence.
There are some badly written licences which make it ambiguous whether the licence can be revoked in future (e.g. Wizards of the Coast with the OGL), but I have rarely seen this raised as a concern in a software context.
It might exist but open-core model ( https://en.wikipedia.org/wiki/Open-core_model ) is much more commmn. I went through QT project licensing FAQ and I still don't see any situation when I would be forced to pay for commercial license when creating commercial software. Some SO answers suggest that complying with LGPL is harder on mobile OSes. Anyway - if you release something on open license, you can't really limit its use, no matter how your second license is constructed. That is probably why paid add-ons are so popular.
No, that doesn't work. You can't retroactively change a license like that. What could happen is that future versions get licensed differently (if either all contributors agree, the project used a CLA or is permissively licensed).
Either way however then you could either stop using it or fork the project.
>You now have to pay $0.20 per install after your app has made $200,000 over the past year. ... Now you're still using open source, but still legally held to the terms of that license.
You can't retroactively pull an open source license ... whatever the new license is, if it forces you to pay, it no longer meets the definition of 'open source' [1]
Unity was a dream, or a lie, the lie that you can make games without a somewhat competent software team.
Now TFA gives hope to another dream, or a lie, the dream that magical open source fairies can let you keep getting away without a competent software team. It won't.
Find those programmers, build a team, build a company where this team has a place, adjust your business model (i.e. raise your prices) to where you can afford to pay them.
This is written based on my quite poor understanding of
the game business. I am sincerely wanting to learn more
about how it works and how the new license structure impact
developers.
My questions below might be idiotic.
(but not intentionally so)
I have no idea how many apps reach above 200.000 installs total
or $200.000 income per year.
How common is it for apps to meet the minimum requirements
when the fees kick in?
I would guess that if you base the application on a subscription
model that this will not be a major problem?
Also doesn't Apple App Store or Google Pay charge far more than this
on income?
The problem here is that most platform tax your income, not your userbase size. If you think of big free to play Unity games like Genshin Impact (or even smaller stuff like VRChat) the amount of paying user is probably in the single digit percentage, yet with Unity terms you are on the hook for everyone who installs your game.
Look at the millions of free to play games in the App-Stores.
Normally you have 98% of free playing users paying nothing, 2% pay.
Imagine now you have 1 million installs. 2% (20.000) pay you 200.000 / year.
980.000 players pay nothing. Now Unity want 1.000.000 x 20 Cent = 200.000, you made ZERO with your game... crazy. That is an uncalculatable risk for a small company.
> For many developers it is now cheaper to switch the engine than to pay for Unity.
Only for a (possibly very) small minority.
I find the the fact that they had no qualms about retroactively applying this to already released game much more infuriating (if they did this, what can they do next?). The pricing itself seems fairly reasonable if make more $1.5-2 or so per user (compared to Unreal anyway).
You'd only pay above a million user and need to have above in 1 million revenue.
Previously you weren't even allowed to use Pro/Personal if you made over $200k (not just per game).
> Now Unity want 1.000.000 x 20 Cent = 200.000,
Not saying the whole price model is not stupid but only those developers who are very bad at basic math would pay this. Everyone else would upgrade to pro.
> Now Unity want 1.000.000 x 20 Cent = 200.000
Again, not really. Even in this case only a subset of user would pay 20 cent.
Another example demonstrating the incompetence of Unity's marketing department, they should've understood that most people can't really memorize more than a single number or be expected to spend over 30 seconds reading something (now I'm not saying that this change overall was not a terrible decision but even a significant proportion of people commenting here don't seem to understand how their per install pricing is going to work).
They would probably qualify under enterprise pricing of $0.01 or $0.02 on Pro. To release a game without Unity branding you also need at least 1 year of Pro subscription. They also would count more than 21M maybe e.g. multiple devices.
They also charge more at less scale and can't detect things like pirated installs, bypass Steam DRM, and you're still gonna hit the Unity servers. It's just checking device HWIDs.
But Marvel Snap is a card game and like many others has terrible monetisation, you can't keep up with card releases at all without paying. It's more like the pay to win games.
One way way to put this into perspective is salaries. 200k usd means any game studio with at least 2-4 employees must earn that yearly to break even, so they are affected.
> reach above 200.000 installs total or $200.000 income per year
No developers/companies who made more than $200k per year (overall revenue of the company, not per game) were even allowed to the use the Personal/Plus tiers and were required to upgrade to pro (which has 1 million install/revenue limits).
I don't think the itself cost would be unreasonable for at least 90-95% of all developers and if you average it out across everyone the proportion of revenue Unity get's would be still pretty low (not much more than 1% or so).
Most people seem to be upset because how they applied these changes retroactively on currently released games and because the whole model seems way to convoluted and not really thought trough.
Why not charge $1 for your game? Problem solved. If you're making a game with a fancy commercial engine I would hope what you're offering is worth a buck. Let's get some quality in this world.
The difference in conversions between free and $1 is not a matter of price. The hard part is getting someone to pay in the first place.
That's why a model of free game with paid addons/extras/pay-to-win/etc. works so well – once someone already tried your game and gets invested in it, they're more likely to spend anything.
And in mobile gaming there's also the matter of looking for "whales". Basically, while majority of people will never spend anything, there's a minority willing to pay hundreds or thousands of dollars. Then, the difficulty comes from showing your game to enough people to find whales.
(Whether that model is ethical in the first place is a whole different matter. I don't think it is, so maybe eliminating it will be a net positive in the world.)
With $1 paid to the developer the player only needs to (re)install the game 6 times on their computer to cost the developer more than they earned.
This pricing is per install, not per player. Installations are an unrestricted functionality which can be abused. If you care about quality games: this pricing causes the opposite.
> We treat different devices as different installs. We don’t want to track identity across different devices.
This allows malicious actors to fake the device identity, which leads to the exact same problem.
> We are not going to charge a fee for fraudulent installs or “install bombing.” We will work directly with you on cases where fraud or botnets are suspected of malicious intent.
This roughly says "trust us bro". They will "work directly with you" in cases of fraud? How nice, but that could just mean an automated response email with generic advice, followed by radio silence. No promise that they will resolve the issue, no promise they will work on it before your dev studio is bankrupt, this is just a collection of words built on trust they don't have right now.
Because mobile games that aren't free do not sell (except Minecraft). In mobile F2P space you can end up with tons of downloads, with less than 20 cents per download, because majority of the players are free casual ones, who play few hours or even minutes, and 99% of your revenue comes from the so called 'whales' (players who spend a LOT of money). If you hate this business model - good for you, I do too and I don't do mobile gamedev, but it's a reality there and Unity knows it.
Unity also doesn't give a crap about quality unless by quality you mean microtransactions, milking players and ads/adware. The engine is associated (partly unfairly) with lagginess, low performance, high ram usage, and crappy games and assets flips. Unity is famous for how slowly new features come and how buggy it is in parts. They also scrapped their own game on Unity that was supposed to show how to use Unity right, it's an engine company that never made games (unlike Epic or id Software, who made games first, engine seconds, and even when being in engine business they kept making games). Unity users who just slapped few assets together basically created the asset flip genre, the opposite of quality. As CEO of EA Riccotello supported always online online DRM that shut down singleplayer games on release day due to servers dying, as CEO of Unity he said you are a "fucking idiot" (direct words) if you don't want to maximize your monetary return from your players. Unity bought an adware company, Unity has an ad network (and if you use it to put ads in your game, they waive the install fee, curious..).
The author seems to misunderstand the new pricing model (as did many others; this is Unity's fault) [1]
The $0.20 is not retroactive, nor do you need to hope your users see 10 and impressions to cover a cost to the game developer -- you need to earn $200,000 _and_ have 200,000 installs. A free game would not be charged, and if you don't believe you can muster 10 ad impressions per user (to cover the unity install cost of $0.20), the base cost of the game can be increased by < $0.01 across your 200,000+ sales starting next year when the policy takes effect on new installs.
I think when most people are, in this context, talking about the "retroactive" part, it's about if it will affect existing games. The pricing structure will be applicable to everyone starting next year.
The other issue is how they measure "installs", as it has been a constant problem in the mobile community for years on end (a problem especially affecting the ad industry as you may imagine).
I personally don't see how they will accurately measure the cases mentioned in the tweet without forcing everyone to have some authentication.
It is retroactive, games that are already made today and already exist today on the various app-stores will be impacted.
10 ad impressions per user might not sound like much, but you're going to get a lot of people installing your app, not liking it, and uninstalling it. The way the mobile game business works the "install count" means almost nothing, you need many, many, many installs to start making any money at all because so few people 'stick around' to actually make you any money.
Regardless of how they spin it, this model is very bad for the FtP games, which is the vast majority of games now.
Looks like you know a lot about F2P games, can you help me understand this fee in the context of other variables such as cost of customer acquisition? Is that usually a much lower amount than 20 cents? What about 2 cents, which would be the cost on a pro license, which most successful F2P devs would assuredly be able to afford?
I’m hoping you can answer these with concrete data. If that’s not possible, could you share your references?
> What about 2 cents, which would be the cost on a pro license,
2 cents with the pro-license only applies after the first 1M installs, tallied per month. If a game is under 100K installs per month, a pro license only lowers the install fee from 20 cents to 15.
Basing the fee on number of installs instead of number of buyers is a problem, it's an unpredictable cost that developers have no control over as it also counts repeated installing of the game. Their FAQ suggests that malicious "install bombing" is something they cannot prevent automatically but instead have to resolve via their support. In other words you depend on them having good support while they profit from bad support, all while their CEO is the kind of manager who wanted Battlefield players to pay for each weapon reload.
If Unity just counted number of buyers they wouldn't have to deal with malicious installs and users of their engine would have more predictable costs. But for some reason they don't want that, and that alone is suspicious.
No, it is different - and it crosses line. There is DRM built into the Unity engine that informs Unity (the company) of an install, and companies pay via this Unity derived number. Not reported game sales, or even self reported game installs.
If I got 3 ads impression the moment I install and open and app. I can promise you that I will never know if the app has 3, 4, 10 or 100 ads impressions.
Unity is saying they can send developers a bill for whatever they want, and if it's not paid their game can't be installed.
Unity has published what "whatever they want" currently is. It doesn't matter if people get the details wrong on that because Unity can change its mind.
That being said, you're still more wrong than the author on the details. The change is retroactive in that it applies to existing applications. The author was very clear what they meant by "retroactive":
> this change will be done retroactively on existing applications as well.
I was crowing to my game dev buddy about how he (a Unity C-sharp developer) ought to check out Bevy and Rust because the Rust type system is friggin’ awesome!
switch to godot(http://godotengine.org/), gamedevs. you reap what you sow. it's better to support open source projects like godot instead of unity whose control lies with the company whose actions probably might not lie with you/your companies' interest. There are hundreds of instances in the past where open source projects replaced corporate projects(ex: blender, git). so, better use godot imo.
All other 3d games I found made with Godot look like they came from the 90s. 2d capabilities look great though. There's probably also some selection bias, it's probably not the first choice for most who want to make a 3d fps.
not quite an apples to apples comparison when your document can be open and read by other spreadsheet applications. You cant take a unity game and run it on unreal.
Until recently, Godot was practically unusable for 3D. They're still actively adding features that other engines already have. It also exports for consoles, where Godot doesn't.
Having said that, I've been looking at it recently and it looks like it has everything I need. I'm definitely going to try it out before deciding between it and UE5.
Seems to me, the bit where all these game frameworks get a bit tricky is when you want to target iOS. Apple keeps moving the goalposts (you must use some version of Xcode, or you must target some version of iOS) and so the open-source frameworks have to constantly faff about to support the latest apple things. A few years ago I was building using LibGDX which used RoboVM to target iOS but then Xamarin bought RoboVM and shut it down and etc etc etc...
What Unity offers is a solid way to build cross platform and target iOS. They're able to offer that because they have the resources to keep up with iOS. Open-source alternatives understandably struggle to keep on the iOS treadmill.
Yup, in fact just 2.5 years ago everyone was switching from Unreal to Unity. The reason? Apple threatened to revoke Epics dev accounts that it needed in order to support ongoing upgrades for ios.
Epic released statements through their attorney's that they would consider the public version of Unreal a dead product if they lost the ability to maintain ios.
That was this summer, not last summer. It's seems that there's a lot of discussion even today about usage going down: https://www.reddit.com/r/ModSupport/comments/16icvv2/1_month... . I have noticed a continuing decrease in quality on reddit for many years (as it attracts more people), like all sites that grow and depend on ad revenue, and activity seems to have gone down substantially lately in the few subreddits I still watch.
Developers / professionals are more judicious in where they spend their time / money / effort than most people. When I select a technology for professional use, I look at the long term prospects for the company. Open source is always a positive, as I don't have to worry about a company raising rates and trying to extract more money from us. Feature completeness can be a concern, as is ease of use, but we really only need our needs met and a path to meet our future needs.
For the record, I am not a game developer but I don't think that changes my opinion on open source technology.
Open Source software is great, but that also comes at the cost of support when you are running into a problem. Particularly if you are on a critical timeline, especially if trying to fix a bug after something is released.
You are at the mercy of the community being able to help (and most likely only being able to give them a limited amount of information) vs the possibility of the company you are licensing from sending you engineers to actually help you fix the problem.
For business critical this is an important thing that needs to be looked at, and from what I can tell there is not anything like this for GoDot (I don't see an enterprise option similar to redhat).
This is not in any way shape or form defending Unity, what they are doing is horrible and will likely (hopefully) cause a migration to Unreal. But just saying open source is not a cut and dry option either. Especially not for something as critical as the engine.
Not really in a position to say anything conclusively either but I don't think you are at the mercy of the community. First, it's open source, so your company can look and fix the code themselves. If it's restaurant software this is likely complicated but for gamedev it is much more likely the company already has the expertise to do that. Second, Godot doesn't have a formalized process but one of the developers can be willing to offer paid support.
> While we don't have a formal structure for offering commercial support, many Godot developers can already help you, work per hour, or relocate to your company. For such inquiries, contact support@godotengine.org. Let us know about your needs and we'll try to find the right person for you.
Not sending engineers, but when I worked for a smaller dev studio we did need to contact Unity and get support a couple times when we got stuck on something. I think we ended up paying for access to the source code and the development manager ended up making a couple of changes in the source code to get something working better as well (I don't remember the specifics, I was working in Unity but on a different game at the time).
This was like 10 years ago though. That guy worked on multiple arcade games in Assembly back in the day too (you've likely played games he worked on if you've played more than a handful of popular arcade games), he was no slouch.
Not only get it out but actively promote it as an opensource project instead of proprietary and make revenue through an asset store. Theres so few games that use source that theres no way its commercially viable on its own.
Valve doesn't do it that way. Instead of some crummy asset store with generic and terrible looking assets they literally just let you use the assets from their games. See Garry's Mod, Sven Co-op and Insurgency for an example of this.
I wish game development wasn't so monolithic. You're either all in on Unity, or all in on Unreal, or all in on GoDot. They are designer forward, code later. Each has a marketplace of sorts, but in my experience it's substandard.
Compare that to web development (different beast I know, but there's no reason gaming can't be more componentized). Web development is so simple these days because I spend most of my time gluing stuff together, and sanding the edges.
Game development has concepts like this, and I have no doubt I'll get replies showing me some nifty component system, or talk about how I just don't understand ECS etc, but they seem tacked on to a monolithic walled garden. There's something magical about "npm install x" and adding a whole feature to your app.
Maybe it's because "it's art" and game development is just a passion project.
> Web development is so simple these days because I spend most of my time gluing stuff together, and sanding the edges.
This does exist in the gaming industry, it is called "shovelware". "Shovelware" is when you just glue existing assets purchased for cheap and "shovel" them into a game for a quick buck. Usually not very complex games, low effort stories, etc. Also usually very cheap to download.
I assume the parent comment meant gluing parts of the game engine together like renderers and physics engines. You can make shovelware in a monolithic game engine.
No, it's not because it's art, it's because it's okay if all websites look the same, but regarding games, mostly 1 or 2 or its kind will survive, the others have to innovate or will be seen as sub-copies. On top of that, the complexity of making a game makes it so that it's more "coupled" to the hardware, so it's harder to generalize. If you want more component, then you can use the assets store, but at some point you will have to put in some work if you want your game to not be considered as a shovelware
The problem is, that making a 3D engine is extremely tough, and most 3D engines are completely proprietary as a result. There are plenty of 3D libraries that make 3D game development for small games easy, but when you compare a Python library next to a system that includes a 3D engine, a level editor, a file management system, and the ability to easily compile for different systems, its no contest for game companies. A fully-fledged development environment scales better for these multi-million dollar projects than a singular library.
It's why in the 90s every game was mostly built on an engine specifically designed for that game, and in the 2000s they switched to prebuilt game engines. Saves a lot of time AND money.
You better understand the value proposition of game engines than you think.
Spin your view a little: NPM is your web development engine. Somebody had to write all of that code that undergirds NPM as an engine as well as the particular library you want to install.
Game engines are no different. They manage a bunch of (mostly boilerplate) game development components (sound, textures, rendering, animation, menus, collisions, physics, resource I/O, etc etc) so game developers can spend time "gluing stuff together, and sanding the edges" of their game. Monolithic engines like Unity have a plugin system for installing 3rd party libraries, and the engine tooling itself acts as its own level, effect, cutscene (etc etc) editor.
At the end of the day games are just like any other programming task. You have a limited amount of effort/time/budget, so you have to make decisions about where you want to optimize. Game engines, even monolithic ones like Godot and Unity (and, really, every 3rd party game development tool) are all designed to let you focus effort/time/budget on the unique aspects of your game.
If you are thinking about why the game engines aren't broken down into finer reusable components (e.g. one module for rendering and another for physics, etc.), it probably has to do with how these things are tightly integrated, and how every game engine has a different design philosophy on how the world should be organized. The different organizations might also mean that you have to design the games to fit a specific game engine, which is unfortunate.
Maybe you are lamenting why the marketplaces or asset stores are so closely tied to the game engines? For some plugins, they have to be tailored to the specific engines. But for arts and sounds, there are many free and paid options to choose from, although not necessarily of the same quality and cost as one might get from the asset stores that are sponsored by the game engines (e.g. Unreal Megascans).
There are still various tools associated with game development (e.g. Blender for 3D, your favorite text editor for scripting), and those aren't tied to specific game engines.
It's not as monolithic as you'd think. There are lots of engines out there but their communities aren't very vocal compared to Unity, Unreal, and especially Godot's community.
If you look at both of these you'll see just how many engines there are and neither of these cover everything. There are plenty of engines popular in the Python community that no one outside of it are aware of. Such as Arcade [0], Python-Tcod [1], Ursina [2], UPBGE [3], and Panda3D [4]. But based on your description you'd really like https://gdevelop.io/. It embraces exactly what you're describing where you can build a game but just installing entire features others have made and put online into your game.
732 comments
[ 2.9 ms ] story [ 477 ms ] threadFor programming platforms (like game engines) this is even MORE of an advantage, as with something like terraform you could conceivably rewrite your stuff in a matter of weeks if you have reasonable testing. For a game that is not possible.
Nobody can take away your rights under the MIT license, there's no legal mechanism to do that. You are protected, fully, from shit like this.
[1] https://news.ycombinator.com/item?id=37499731
Like going from GPLv2 to GPLv3 or straight to AGPL, so now you are stuck with an old version forever? Yeah, that would never happen!
That's a massive overstatement of the current state of OpenTF right now. Plus they're still dependent on Hashicorp's hosting which also changed their license terms in response to the project.
You're right that because it's OSS you can do this kind of thing where it wouldn't be possible at all in a proprietary system but "just fork" requires a community to organize around it. Without a bunch of backing orgs this wouldn't have happened.
I got burned by Steam, I have a super old laptop with some old games on it, Half Life, so one day I got the laptop out, Steam updates and f** itself, the old system is no longer supported but they had to f** things up s the games won't work.
I agree, open source is not required, but we need to own our software not rent it.
That's just a start but also require freezing your OS and have lots of spare hardware or have rights to emulate it :)
And same with data - buyed songs for example :) Big industry already is killing media you can own - cd, dvd, blueray depends on outdated cpu, pendrives decay before becoming usefull backup media...
In the age of asholess open sources and resources are best. Let's bring more viral licenses then GPL !
_Some_ kinds of Open Source might help there, since _some_ open source licenses are very easy to comply with and very business-friendly. On the other hand, all the game studios now complaining would also howl and whine when forced to be GPL-compliant and release their source code.
And even with business-friendly take-what-you-want-and-never-give-back Open Source licenses, there is always the risk that the project you are using does what Hashicorp did with Terraform and stuff: Change the license for all future releases to something you won't like, cutting you off from your necessary updates and fixes. Maybe there will be a community maintaining a fork, maybe there won't.
How long would it take for someone to take over the project (if ever).
One could even imagine scenarios like an originally MIT-licensed software splitting into a commercial company offering commercial paid licenses, plus a community (or even the company itself) offering a GPL-licensed fork. Of course one could then still maintain an additional MIT-licensed fork, but if the rest of the community is happy with GPL and all the development just happens there, your MIT fork will "starve"...
Given the project itself is still strong, this might not be a problem, but then I see no reason why it has chosen it in the first place if not for that specific option.
The law often makes use of the reasonableness standard. I'm hard pressed to believe that carefully reading the hundreds of TOS and EULA's hoisted upon us is reasonable.
If you're reading my comment you must have read the TOS/EULA for ycombinator, firefox (or chrome), your wireless or ISP, the keyboard app on your phone, odds are you have an email, perhaps a google account, remember the OS licence. Ever listened to a music streaming service, watched youtube? Messaged using an app? Banked online? Have some managed passwords? Like games, how many? For the common people, what about social media?
That's about a dozen "contracts" and that's lowballing it, multiplied by each update to the "agreement" (pray they don't alterate further) multiplied by the requirement to also read and acknowledge the privcy policy. All this for services that have become when not essential ubiquitous and constantly shift under you. You'd need 8 figures to run that by a lawyer, or a part time job to carefully consider. That's not a reasonable arrangement.
So, no, your request is not reasonable.
I'm confused what you mean by saying that the employees of a business don't need to read these legal texts but that the "business" should read them. The "business" is not a physical life form that has the ability to read, only the employees of a business have that ability, since they are humans (unlike the business entity itself, which is not human). Perhaps your idea was that businesses should hire a team of lawyers whose only job should be to read through the various ToS and EULA legal texts that their other employees merely click-through?
If you as a buisness sign up to amazon web services to host your entire backbone on, you better read the contract. If you don't, you are irresponsible.
Someone, somewhere at the buisness should take the time to read a simple little document before using it as the backbone of everything they do at a company, yes. It does not have to be a lawyer...
I'm extremely sympathetic to the idea that there are too many eulas for free products, things like games or whatever where it really doesn't matter for 99.9% of users. But for the 0.1% that make their livelihood from the program, they should take the 25 minutes to read through it at least once.
I'm not saying this because im a big bad lawyer that hates you (I am not a lawyer at all), im saying it because it's a tiny thing you can do to save a whole lot of heartbreak.
You just gotta decide when its important to read contracts and when you don't care. I don't care about the contract when I get new lenses for my glasses, or when I pay for netflix. But if im moving into a new house? I read the bloody contract.
Yes, but there are two sides to "resonableness": The "what" side and the "who" side. "What" is the thing that should be resonable. But the overall resonableness just as much depends on the question: "for whom is this reasonable?". While a consumer cannot possibly be expected to really read all the ToS everywhere, a business maybe can be expected to do so. Especially for things that are very critical and integral to the business, like the license of that one framework you are building all your software upon. So I do think requiring a business to read and understand the Unity ToS is totally resonable.
While TOS are rarely worth the toilet paper they could be printed on, I'm curious about whether arguments could be made about whether existing subscribers from 2019 could now sue for breach of contract and costs associated with (re)development.
The big question is: to what extent to Unity games need to be able to talk to Unity's servers? If they're looking at number of installs (and apparently that includes pirated copies even?), serve ads, and probably provide other services, that sounds like the games need a connection to the server. In which case they may be able to disable your game if you don't pay. And then even if you could sue them, the real damage is already done.
I would feel very uneasy if my product was based on a framework, where the provider makes no commitment and reserves the right to change the licensing terms at any time.
The last part I do not agree: a license cannot be changed retroactively if there is no provision in the original license for doing so, and no open source license have that. Even if "revoking" licenses for all prior releases were allowed, it could only work if all copyright holders agreed, which is not practical for most projects.
Unreal has licenses which allow you to use one version of their engine FOREVER and they cannot revoke that license. So the author didn't fully research this. It is true that for NEWER version Epic can change this.
Tim Sweeney explicitly mentioned this often + the fact that they are trying to break the Gplay/Apply monopoly shows that the Epic games leadership are not corporate piranhas like Unity's.
But what about Godot? He says as if "it's open source so no issue". Yeah but what if the devs stop supporting it? This "community will continue to work on it" is BS: in reality it's usually one or two guys who actually do the actual work.
So if Godot devs stop working on it you're stuck with a project with zero support. Good luck developing it further instead of focusing on your game.
The same goes for his very own product.
For Godot, if we were to stop developing it (and I'd like to note that Godot is one of the most active projects on Github right now) you would still have the version you have now.
What is the difference between Epic not working on a game engine you use or an open source project not working on a game engine you use? Except that with Godot at least you could work on it yourself if you wanted to.
let me play the devil's advocate - unreal's source is available (despite it not being actual opensource licensed). This means if Epic ever abandons unreal, you could theoretically also just make the changes you need to support whatever your project required - as long as you didn't distribute those changes (except perhaps the run-time? Not quite sure how unreal engine and the runtime are licensed).
https://www.unrealengine.com/en-US/eula/unreal
> 2. How You Can Use the Licensed Technology
> Epic grants you a non-exclusive, non-transferable, non-sublicensable license to privately use, reproduce, display, perform, and modify the Licensed Technology in accordance with the terms of this Agreement (the “License”). This means that as long as you are not violating this Agreement or applicable law, you can privately use the Licensed Technology however you want. If you want to share the Licensed Technology or anything you make with it, Sections 3 and 4 below address when and how you can do that.
In section 3/4 it goes to that you can compile your game and give the output to outsiders if you pay the roylaties when applicable or share them royalty free on epics github (basically you make a pr to merge your stuff upstream) or on unreal marketplace (sell it)
The additional sections that allow distribution are the important bits and I guess the devil is in the details.
You do need a license to modify the source, use it internally, and then sell a binary you've produced with the modified source.
You just can't redistribute the source to anyone else but using it internally for your own projects to build executable that you share (and sell as long as you pay the royalties to Epic) is fine.
The last UE3 game released in 2021. UE4 came out in 2014 and UE5 in 2020.
Unreal Engine license are per engine version and perpetual. This is something Sweeney has been pointing out for years. And again a few days ago when Unity started this shit storm. And as Sweeney points out the big studios/publishers usually negotiate even better terms.
https://twitter.com/TimSweeneyEpic/status/170161922085161792...
Really it is only an issue for new projects and at that point if the license of the newest version is not to your liking pick a different engine.
Also the last UE3 game I know of (Them and Us in 2021) was made by a small indie studio not some big AAA studio with massive publishers backing. At that point the engine had been in "end of life" state for 5+ years.
No it doesn't, you still don't own the code so you can't just modify it and use it to develop games.
The key difference between proprietary and Free Software is not actually that you can or cannot modify the source, it's that you are guaranteed access to the source in order to modify it or not. Since you usually cannot access the source of proprietary software, you usually have no legal way to modify the source for your computing.
There's zero risk of this happening.
Godot is going to become Blender for gaming and eventually eat into Tim's margins. (Unreal isn't even his cash cow.)
Unreal might be significantly ahead now, but when Google, Amazon, Apple, etc. contribute to Godot, it's game over.
Amazon did make a game engine. It shows "Amazon money" doesn't magically solve every problem in the world.
> Godot is going to become Blender for gaming
Exactly. Blender still isn't the first choice for most animation and VFX jobs.
An open source project doesn't have to be the very most used thing from the beginning in order to eventually eat everyone's lunch :)
Also, it let me have a cute monkey as a logo : https://prehensile-tales.com :) (I'm not selling anything there, but you can see the logo)
And it doesn't need to be, it just needs to be a viable alternative. Everyone always looks at these projects like it's a winner takes all scenario, but it's not. There can be multiple programs that accomplish the same thing, some open-source, some not, so long as they're sustainable and have something unique to offer that's not a problem.
Sure, if we were living in my ideal fantasy all software would be fully open and free, but in this reality I'm just happy we have alternatives that are actually sustainable and don't feel like you're actively gimping yourself.
Blender has some parts where it is the best (the tracker for example easily beats all commercial trackers I have ever used) and other parts where it doesn't shine as much (e.g. fluid simulation — which is a non-issue because it integrates well with other tools).
I have been coming from 3dsMax and Maya and never have been looking back. Blender also has been getting so much in the past 5 years it feels ridiculous and makes you wonder what the likes of Adobe and Xo are doing with all their money.
We need standards, sure, but we desperately need better competition between high-quality products.
It doesn't really need to be; being self-sustaining while being open-source beats Unity's model of being funded to cover operations while being closed-source.
TBH, it's only a matter of time before Blender is a choice for most animation and VFX, and then only a little more time before it's the first choice.
You are grossly underestimating the complexity involved in game engines. It is not like a web app where devs don't have to worry about constraints like memory or frame rate and chill. Things need to happen in real time. A delay of even half a millisecond is not acceptable. And these "Things" involve changing of 3d objects' position w.r.t player's movement, calculating zero or tens or hundreds of NPC AI characters' position and finalizing their animation state, calculating the lighting on all the objects and a lot more. All this just to finally render and present one frame. Yes, hardware has gotten better over the years and memory constraints might not seem like an issue but that is not the case for games. Improved hardware only helps with improving the overall quality of the game. Game now will be able to afford to look better and do more things than games from 2003 and that's it. Games still need heavy optimizations.
Thanks for the better hardware, making 2D games now is neither expensive nor hard. So, Godot being more friendly than unity for making 2D games is quite possible to happen. But that is not the case when it come to comparing Godot with unreal. Unreal is already at a league of its own. I don't think Godot can integrate something like nanite or lumen inside its engine anytime soon. In the past few years, only unreal has been introducing ground breaking computer graphics tech inside a game engine. Unity is having a hard time to even keep up with unreal's tech like meta human. It doesn't matter how many google, amazon or apple contribute to godot, it'd be a big surprise if Godot is at least able to hold its ground against O3DE IMHO.
Your answer: you can use the old version.
What if Godot stops the development?
Your answer: good luck (???)
You can also use the old version. Also others can take it from where they left off, since it’s open source, as it's has been pointed.
Then others can take it from where they left off, since it’s open source. As long as it has a user base, it is guaranteed to live on. Unlike Unreal.
edit: it doesn't unless Epic Games grants you a specific license to do that. https://www.unrealengine.com/en-US/eula/unreal
source:
> 4. How You Can Share the Licensed Technology When It Isn’t Part of a Product > You may only Distribute Licensed Technology (including as modified by you) outside of a Product as expressly permitted by this Section 4.
> a. Sharing of Engine Code
> i. Sharing Engine Code with Another Licensee You may Distribute Engine Code (including as modified by you) in Source Code or object code to a third party who is separately licensed by us to use the same version of the Engine Code that you are Distributing.
>Any public Distribution of Engine Tools (e.g., intended generally for third parties who are separately licensed by us to use the Engine Code) must take place through a marketplace operated by Epic such as the Unreal Engine Marketplace (e.g., for Distributing a Product’s modding tool or editor to end users) or through a fork of Epic’s GitHub UnrealEngine Network (e.g., for Distributing Source Code).
You can't do that for Godot either.
That is what all company selling proprietary products that include MIT and BSD licensed codes do. Usually the jist include a file called "third party copyright notice" with the product ad well as an entry in the "about" section of their gui.
If Unreal screws you, then you can fork it to build some features you need for your current project. If Godot screws you, then you can fork it to build some features you need for your current project, cooperate with others on the features they need which also help you, and start a community for Engine-formerly-known-as-Godot-v2 and invest in it as a thriving basis for projects 10 years down the road.
By hitting the fork button on github.
>but the license doesn't allow changing it legally afaik
It does allow you to make own changes to it and share it with other licensees of the version you forked from.
In an age of locked down mobile OSes and forever changing graphics drivers and such, I don't think "FOREVER" is very long any more.
It's like a perpetual license to a specific version of JetBrains Rider. It's sold as if I own a general purpose tool FOREVER and can opt out of an endless subscription but no, it doesn't mean much at all. It will not work with the next runtime update and past runtimes are obsoleted after a couple of years. They have managed to outsource their subscription lock-in as the release cycles of a third parties while pretending to be holier than thou. If they shipped the volatile parts as open source plug-ins, I would feel differently.
For Unreal, I don't think there is a solution to closed platforms if it requires a large/sophisticated team to maintain compatibility. We can only point out that any notion of "forever" is a fiction - you have got in bed with a commercial dependency you may come to regret if they choose to change the rules.
I can still run my first 3d engine from 2003 on Windows. Windows actually does a really good job with backwards compatibility.
Epic is not doing this out of the goodness of their heart, they're doing it because it is beneficial to them. Epic is a multi-billion dollar corporation part owned by a massive conglomerate.
> BS: in reality it's usually one or two guys who actually do the actual work.
Over the last week, Godot had 32 authors pushing 52 commits. Over the last month it had 135 authors. This is not "one or two guys".
> So if Godot devs stop working on it you're stuck with a project with zero support.
Guess what, if a closed-source company decides to stop working on their product you're also stuck, but now you're even more stuck because you don't have access to the source to make your own changes!
Unreal uses perpetual licenses for their versions meaning this kind of bullshit behavior is not possible
Then it's goofy af to say Godot might make a license change that necessitates using one version in perpetuity-except a version you and others are allowed to modify and distribute.
Isn't Unreal's source open?
I was going to comment that it’s viewable by anyone on their GitHub, but you are correct: you have to be part of the Epic GitHub organisation to access the repo. Getting access is easy, but you do have to agree to their licensing terms to do so. So yes, what you said.
(currently there seem to be around half a million users with access)
That's basically what I understand as "getting a license for a game engine" (this is also why Unity got popular in the first place, because they skipped all this nonsense).
OTH I accept probably 5..10 EULAs a week without thinking or even reading the text (most of them are not enforceable anyway).
There is nothing unique about German law here. The same is true in the USA and most other countries. If a EULA or any other contract (whether agreed verbally, signed physically, digitally, or via a "click") violates the law, it can be considered invalid/unenforceable. It has nothing to do with how the agreement was agreed to, but with what the agreement contains. (There may be some legal theory that may carry some weight that a contract agreed to via a "click" is more likely to be unconscionable that one agreed to with a physical signature, but that does not automatically make all "click" agreements unenforceable.)
Interestingly, Unity runs afoul of many of those.
You're still welcome to contact them directly and try to get some kind of contract directly if you can't agree to the EULA for access.
Have you read the unreal engine license agreement? [0]. Section 3.a covers modifications
[0] https://www.unrealengine.com/en-US/eula/unreal
Edit: I see some language covering this further up.
EDIT: To elaborate, this is an article about a company fucking users over after an aggressive growth / dumping business model phase. Your response is yes but this other company would never fuck users over after an aggressive growth / dumping business model phase.
Well, Epic didn't do that when it went through it's aggressive growth and the last time it dumped it's business model. Instead, they changed their licensing to lower fees, and when you look at what they did across the board, made things better for customers and game developers.
So, all evidence is to the contrary.
Edit: Also, nice strawman.
People, on the other hand, I can trust. Not often, but in cases where a person has made a long series of decisions over a decade or more, you can get a feel for what their value system is. Tim Sweeney is in this category. He was involved in Unreal back when I was in undergrad, more than 20 years ago. So I sort of put him in the same category as Gabe Newell and John Carmack: relatively enlightened game business leaders that understand the true value that gamers and developers derive from the ecosystem. None of them are in to make another dollar in the next quarter: they are focused on long term success and the are passionate about games themselves.
So it's not that you're wrong, it's just that your argument applies cynicism uniformly, and I'm not sure that's fair given the history of those involved.
Holy straw man! And there’s certainly no self-interest involved with me using their engine without giving them a cut under 1 mill right?
You can make this argument about literally anything.
Godot must have no self-interest in releasing their engine for free and making $30k a month to develop their project.
Who cares what the motivation is if it gives real benefits to devs. If anything it tells me they’ve picked a business model that’s mutually beneficial unlike Unity
It's fine to support them while they are the underdog and are saying and doing the right thing, but don't pretend like they will say and do the right thing forever.
Motivations matter, because they will decide in which direction a project moves over time, or if situations change. Epic is the "good guy" now, but only because they are an underdog. If they become the dominator, they might become the predator which Unity tries to be at the moment.
But if we are honest, such uncertain possibilities don't matter if they are so far in the future. This might be a problem for future games, 10, 20 years down the river..
More then 50% of the shares/votes is owned by a single person. The founder and CEO Tim Sweeney.
Which doesn't really matter unless you'll lose control. If 1 entity has >50% it's pretty safe.
Tim Sweeney founded Epic and wrote the Unreal engine himself. John Riccitiello’s whole career is in management. Even if they’re both entirely motivated by profit, they have different perspectives on how to get there.
Also, Unity is a public company, while Epic is private. Even though Tencent owns a considerable share of the company, Sweeney still holds over 50% ownership. That gives them very different incentives.
Sometimes we forget that executives are people too and they have their own personalities. Tim Cook famously[0] told climate change denialists in a shareholder meeting that “If you want me to do things only for ROI reasons, you should get out of this stock”. Sure enough, shareholders are also not solely motivated by profit, and voted with Cook on that occasion. It’s useful to remember that cynicism isn’t the same thing as realism.
0. https://www.theguardian.com/environment/2014/mar/03/tim-cook...
Further more, Tim is bound by law to do what is best for the shareholders. Simply put, if Tim favored environmental concerns over profit he would be removed.
This stupid meme needs to die already. There is no such obligation, he only has a fiduciary duty to not trash the company and spend the earnings on cocaine. "companies are legally forced to maximise profit" has never been true and this piece of misinformation has been going around for ages now.
It's more like too hard to be proven in any way. Unless you live in an simulator it's really hard to say which set of decisions is better than another. People often say it is obvious or in hindsight but fact is there are no such hard proofs.
As long as the executives are behaving generally how the shareholders want, it's not a problem.
My (IANAL) reading of it is that maximising shareholder value is probably the law, but it's practically unenforcible. Being practically unenforcible doesn't stop CEOs and boards from using it as a guiding principle.
https://en.wikipedia.org/wiki/Dodge_v._Ford_Motor_Co.
And that's just civil influence, there are legal mechanisms indeed in place if a CEO "trashes a company" and what that means is different depending on the company.
And, insofar as such an obligation to “maximise shareholder value” might exist, that obligation doesn’t necessarily translate into “maximise profit”.
The shareholders of a theatre company might care more about breaking even while getting an interesting assortment of plays produced with a great cast than they do about making a bunch of money out of the venture, so an executive who makes a bunch of money by running productions of uninspired cash grab shows won’t actually be maximising value. Likewise, I’m sure that Rob McElhenney and Ryan Reynolds care more about Wrexham AFC’s managers getting good athletic results than they do about making a bunch of money.
And even where it does translate into "maximise profit" because it's what shareholders of a particular company may want, there is no timeframe for it, and there is no way to tell whether any particular decision by the CEO runs contrary to the goal of eventually maximising profits.
Companies can spend all their revenues plus a constant stream of new capital on growing market share or revenue, on charitable activities or the happiness of employees, on huge research and development projects or on restructuring after restructuring and still credibly claim that all of it is ultimately meant to maximise profits.
The point where CEOs and CFOs have to be careful is when the company faces solvency issues. That's where legal limits of freewheeling decision making kick in, because it's where it's no longer about shareholders but about creditors.
eBay v Newmark
https://h2o.law.harvard.edu/cases/3472
https://onlinelibrary.wiley.com/doi/abs/10.1111/basr.12108
Businesses are legally bound to act in the best interest of their shareholders. This is quite an open ended precedent.
You can argue almost anything meets this criterion, in some egregious scenarios a court won't buy it, but they will give you enormous leeway.
There is a lot - A LOT - of room for ambiguity and debate on the specifics of shareholder value and "best interests." The "legal constraints to act in the best interest" is not some set of corporate rules and KPIs codified into our legal code write large. It's not about maximizing a specific KPI over a fixed timeframe.
Not to mention Sweeney is the majority shareholder in Epic's case.
Businesses are legally bound to follow the official decisions of shareholders at official meetings. Anything beyond that is merely "a good idea".
If they are legally bound, there must be either a law or contract, can you cite either?
Don’t fall for feelgood greenwashing.
I mean, Apple should just release a new model every five years and really get that footprint down...
https://www.ifixit.com/News/64865/iphone-14-teardown
https://news.ycombinator.com/item?id=37573332
Yeah, right now (and I hope it keeps increasing thanks to the Unity news).
I got burnt once (and almost twice) from this in the past. cocos2d-iPhone used to be huge, and Zynga even contributed to it. I released two games using it, and started another one. And then it stopped getting updated (and Apple keeps breaking things like Apple likes to do), so it died on the vine and I had to port to something else.
Currently making a game in Monogame, and while it did get a significant update a little over a year ago, it's had very little activity on it since then and no other releases besides a hotfix shortly after, and zero communication about what's going on with any of their official channels at least. Not great especially since there seems to be spotty support for .Net 6 outside of Windows still, and I keep running into various issues with its 3D support (which it is mainly supposed to be a 2D game, but it does support 3D to a certain extent).
Also while it claims to support platforms like PS4 and Switch, I see almost no documentation on it, and very little documentation on getting Steamworks working with it (I have some basic things working, but I'm having to figure things for Unity first and then porting that knowledge over). Also most of its multiplayer libraries seem outdated, at least the ones I looked up. And the forums/Discord still have some activity on them, but not a ton. Also I'd love to make a game that supported VR (almost switched to Unity just for that alone).
I actually compared its Github activity to Godot a week ago, and Godot looks SO much better supported than Monogame at this point, that I was already considering either porting or making my next game in Godot before this Unity news was announced. But maybe that's just me hopping onto another platform that will have the same problem in 5 years after I've gotten pretty invested into it.
I also know Unity (worked on a game professionally years back) and considered switching to that for my game too, but that mostly got killed by this announcement.
Personally I prefer code-based development (I don't like using an editor too much), so if Monogame was better supported and had much better 3D support I'd probably just stick with it. I thought I was going to when I first started using it seriously three years ago. But I hesitate to keep dumping time into it if they don't maintain it.
Also I made a small game with Phaser.js and was going to do more with that, but even that creator is pretty much the only one maintaining it and they got sidelined by life for about six months (which is fine, but get someone to help keep it going if you can! I know that's hard though). Phaser at least is already close to what I want from it, and doesn't seem to need much else for the foreseeable future.
the Epic games leadership are bribing developers for exclusivity, including kickstarters where i was promised steam or gog
the Epic games leaderhip makes money from free to play grinding/gambling simulators
the Epic games leadership also comes from Tencent that made all their gaming money before trying to get into the western market from asian gacha
The only reason their terms are more reasonable now is because they're the underdog and throwing a lot of money made from free to play crap at fixing that.
Kickstarter is kickstarter. Half of them probably don't fulfill all their promises. That's the risk you're taking, and why it's not a preorder.
Epic don't have gambling. They've even removed gambling from Rocket League after purchasing them too.
It doesn't have to be illegal, merely immoral.
> Kickstarter is kickstarter. Half of them probably don't fulfill all their promises. That's the risk you're taking, and why it's not a preorder.
I don't expect them to finish a project when i back them. I expect them to deliver my fucking GoG key when they finish the project and ask me what do I want it on and i say GoG! I don't expect them to say "Epic paid us a ton of money and you can get it on their gacha financed store".
True story. So long, Julian Gollop.
> Epic don't have gambling. They've even removed gambling from Rocket League after purchasing them too.
They're free to play. That means the game is designed to keep you playing forever and keep you buying IAPs. It is not designed to entertain you.
That sounds like an unfinished feature to me then.
Why are you playing games forever that aren't entertaining?
Did I say I play endless games? I said free to play games have to be endless to get you to pay.
Lol you can literally pay people to keep supporting it if the code is Free.
The same goes for Unity and Unreal.
That’s a very unique take. I would call a company that lets others take startup a buildout risks and the uses the courts and regulators to swoop in with a low cost competitor a pirhana.
It’s a good strategy (for Epic), but it is extremely predatory. It’s no secret they have their eyes on the console market next; let HW makers popularize platforms using a business model of low margin HW and high margin SW, then get governments to mandate alternative stores so Epic can undercut the SW.
Low risk, low effort, high return. It’s a solid business strategy but if it’s not pirhanic, I don’t know what is.
If I understand correctly, Unity had that too. And then they changed it anyway.
I don't know the details, but apparently Unity maintained their license in a git repository for the explicit purpose that everybody could easily track changes to their license. Just before the license changes, they removed that repository, and later put it up again, but without the clause that you could always use an old version forever without new restrictions applying to you.
Exactly.
We like to hate corporate greed and lionize the open source developers. I also hold open source devs in high regard.
BUT.
Open Source Dev have to eat too.
If we want to keep the Open Source ecosystem moving, we do need to find a way to pay them. Even today, a lot of open source projects are supported by individual corporations that keep the devs on payroll to give them time, but that can also lead to influence and lock in, and even abandonment if desired.
For a 5% cut. Add that on top of taxes and steam and you’re basically an employee.
The extra 18% doesn't feel as unreasonable when you get to make a lot more sales.
Maybe _you_ didn't research how Unity had this same exact clause and decided to just... remove it. The author is implying Unreal could do the same at any point in time.
Isn't that valid for any software including Unity or Unreal?
With open source at least you can try it yourself
Haha this made me laugh. If you think epic isn’t piranha then you’re living in a bubble.
I thought the unity TOS had this too?
Epic's still remains. Yes, if Epic removes this feature of the license, we can be concerned. But, Epic's license means I can remain on the previous license that has these more favorable terms.
Isn't part of the current furore that unity used to have similar terms and they removed them?
so did unity, until they removed that clause
Unity also had a similar clause in their ToC, they removed that now and retroactively applied the new pricing model on ALL previous versions of Unity. Unreal can do it too.
Even if this is 100% legally true, they can absolutely say that they are revoking all old licenses, and then if you used one, it's up to you to take Epic to court to prove that they can't.
Exactly I'm glad someone is pointing out the problems with these piddly open source projects. Companies are too reliant on open source software, I mean it's not like open source tooling has ever taken off. Except for Emacs, and Vim, and VS Code, and maven, and ...
But seriously though the over reliance people have on open source projects is staggering, I mean really what happens when the guy who maintains Linux gets bored and wanders off, everyone is boned.
Therefore I agree with OP all open source projects are inevitably doomed to failure and can never work.
Glad to see someone else who sees reason.
Yeah. They've reached critical mass to, from this point on, be self-sustaining.
I don't think Unity ever got to that point.
This is where FOSS comes in: If support ends, you can keep using the latest version just like with proprietary software, but you can also fork it and fix pressing issues. It's not even a question of relying on the community continuing to create major new releases. If that fails, your org could fork it alone and patch any blocking bugs.
My company already does this for several pieces of legacy software that we haven't had the time to migrate away from. We don't make any major changes, but we can and do fix things that get in our way. We'd get no such benefit from proprietary software that changes its terms to be unfavorable.
No mater how slim the chance Bevy actually becoming a mainstream engine is.
For years the only thing really holding the company a float was their valuation. So long as investors were willing to give them money for growth and future profits they could continue to fund their loses. But now their growth is slowing and investors are less willing to pay up for growth generally what does Unity do?
I've felt for a while that they'd be a likely victim of this most recent tech rut and that seems to be playing out. They have no moat, slowing growth and are burning huge amounts of cash. Unless the macro changes in their favour it's hard to how they get out of this and stay on top.
Unity acquired Weta at 1.6B. The solution is not to do this.
"Weta Digital’s Academy Award-Winning VFX teams will continue as a standalone entity known as WetaFX under majority ownership by Sir Peter Jackson and helmed by CEO Prem Akkaraju."¹
1- https://investors.unity.com/news/news-details/2021/Unity-Com...
The only reason Unreal engine has so favorable license conditions is because Epic Games earn billions from Fortnite and Epic Games Store.
[1] https://en.wikipedia.org/wiki/Unreal_Tournament_(cancelled_v...
I'm also completely baffled how poorly thought out this whole thing has been. Unity has been used by the likes of Nintendo, Microsoft and other massive game development studios. Do they seriously think a fee applied on retroactive sales and revenue numbers would be accepted without issue from them?
Unity have not proposed a fee on retroactive sales. They have proposed a from-this-point-forward fee that applies to new sales/installs of any game made with Unity, including new sales/installs of back catalog games.
In other words: This is absolutely a retroactive rug-pull on new installs of old sales.
Update: Because of the way they're rolling this out, the only way to avoid the retroactive license change is to immediately stop using Unity's development tools. If I were running a game studio, that's what I would do.
Note, it's not sales, it's installs. And it's not new installs.
That you keep saying sales really means you aren't fully informed about the changes and should spend some time researching this.
They ran the numbers. If it was going to work they would have gone that route, but I think the problem is that the clear majority of Unity-users (indie devs) make no money anyway.
Revenue sharing with someone making $0 in revenue is pointless.
They could've probably been sustainable and profitable if they mainly serviced their core audience of indie developers with a smaller amount of employees and a simpler product, but it seems they really wanted to brute force themselves into the AAA market.
For better or worse the "growth at all costs" strategy was the strategy companies like Unity had to employ for the last decade to attract investor capital so I tend to blame the low interest rate "easy money" environment more than companies like Unity being reckless.
Although that said, it does seem Unity made some stupid moves in recent years. I don't follow the company that closely, but I'm aware they made some large acquisitions funded in-part with debt during the pandemic. It's one thing taking on debt if you have a strong balance sheet, but it seems a little short-sighted for a company losing billions to use debt to fund an acquisition – especially at the elevated prices they paid during the pandemic.
As if the founders had no other options in life than to start companies relying on business models where you capture the market with investor money and then once you’re customers are locked in, you squeeze them as much as possible. Unreal is in the same market. They also took in investor money. They’re not in the same situation.
The people behind Unity chose to play the game. The c-suite is handsomely rewarded, and we praise their business acumen when things go right, but if things go wrong suddenly they’re victims of circumstance.
They can take ownership of their decisions.
The early investors and founders cash out first and the public that bought into the unsustainable growth trajectory are left with a failing company.
This is so common I truly don't understand why people still buy into these companies.
I feel much worse for the thousands of developers he's holding a metaphorical gun to with this awful policy, who now have to stress about swapping engines.
https://arstechnica.com/gaming/2022/03/the-witcher-will-retu...
Source: game dev using unreal
Gaming companies get a lot of crap for this - it seems unfair. The OS I'm using to write this is a relative newcomer at only 30 years old. Obviously, it has changed a lot in that time, but so have the engines. Most software doesn't get rewritten, it evolves.
For example, Larian (BG3) and Wube (Factorio) each have their own engine. They are specially built specifically for the games they are making.
Also, you say "have been kept modern" but even that is questionable. "Modern?" What does that really mean? It's "modern" enough for the game. Starfield was just released, and it was released without ray tracing.
And we are just talking about games. Unity and Unreal do more than just working on the engine for games. So yes, while it's a lot, it's not fair to compare the bespoke engine use by game companies and engines like Unity and Unreal.
Even so, 8000 employees does sound a lot. And if they're losing a billion per year, it sounds like they have no revenue at all.
But also their previous monetization seems to try charge per professional developer which is a limited audience compared to consumers, especially if you want to maintain a AAA engine (which it seems Unreal is favoured still) given it seems fast pace techniques & improvements. Maybe by selling cloud servers for networking, which I think maybe they were too expensive compared to alternatives.
> Unless stated otherwise, we filtered out unreleased games, free games, those that launched with less than a $4.99 price point, and those that have fewer than 50 reviews.
May I ask you if you ever heard of Rust? The game, not the programming language. Online survive game, which has thousands of active players every single day since like a decade already, full 3D and quite awesome graphics I'd say. It's written in Unity.
It's wiki: https://en.wikipedia.org/wiki/Rust_(video_game)
A gameplay of a rather popular youtuber that does daily uploads: https://www.youtube.com/watch?v=WSXCI0rLvHg
I would have picked a different game.
I remember the minutes of loading of a new world..
That said I'm in the "dozens of hours" bracket and not the "thousands of hours" one, so maybe I just haven't hit that yet.
Unity supports basically all gaming platforms. Windows, Linux, Mac, PS5, Switch, XBox, etc.
There are AAA desktop games that use Unity.
Maybe they should have something to show for this $1B/year and take a cut of this $250B+ industry to cover the costs? They appear to have about %30 market share, so they need to capture less than %2 of the value created with their tool to break even and if they can't do that or they are providing tech for the less than average profitable part of the industry they should shift focus or reduce costs.
AFAIK it's only natural for businesses to go out of business if they can't capture more value than they consume.
None of this is attributable to engines, it's the size of the products built using the engines and it appears that the ad revenues are not included and that appears to be another $80B.
The game engine market is the market of supplying the game developers with the tech to build their games. That's where the game engine makers that spend 1B per year and hold %30 of the engine market need to charge about %2 of the products made using their engines to break even.
Bought by Apple or Microsoft maybe?
If you haven't seen it already: https://sriramk.com/memos/zuck-unity.pdf
Meta seems to enjoy throwing money into the VR space to subsidize it at the low-cost end, I could easily see them doing a hostile takeover of Unity "for the good of the VR space"
How easy is for a studio to change engine? Isn't that a decent moat?
It may also make AAA invest away from Unreal as it looks like a monopoly now.
If you were switching Unity -> Unreal you'd honestly be better off firing your team and hiring Unreal devs
That doesn't justify what they've tried to do here though. This isn't just a price increase, it's a significant price increase, and a poorly thought out "revshare" model that applies retroactively people who signed deals with them and built a business model off of the deal they signed.
https://www.blender.org/about/people/
In theory, I can just walk through solid wall, using a quantum tunneling effect.
The closes thing you will find are things like blender release under GPL (i.e. pay me a money to release my commercial software under GPL, but that's not selling GPL software),
This whole line of thinking is disingenuous.
>You have a better chance of winning a lottery, happens every day.
Agreed here, though.
[0] https://www.simplemobiletools.com/
It's not like they bought these with a pile of $6B cash they had lying around
Yes, and now they are burning tons of money due to that debt and having to make ridiculous decisions to appease the stockholders. You just stated the exact reason this whole thing is a problem.
Epic maintains their far more advanced engine with 2200 people, and they also run a AAA live service game.
ON WHAT AND WHY???
I don't. They could have sold to Meta and didn't. If they didn't have a plan that didn't involve fucking their users, that's on them.
For sure an open source solution would be more than welcome, but sounds like impossible, because either those out there are half-backed, or are just finding a way to monetize it too..
[1] https://unity.com/pricing-updates
We really ought to separate the fact that "open" does not always mean "free" here. Especially given that it is fairly straight-forward to change licenses on a whim.
Imagine you wake up one day and suddenly the company/OSS maintainers put out a dual-license on the project. You now have to pay $0.20 per install after your app has made $200,000 over the past year. Now you're still using open source, but still legally held to the terms of that license.
Licenses tend to be permissive enough that if such thing were to happen or other factors(discontinued development/support/etc), community forks can be created(which has its own set of problems).
https://opensource.stackexchange.com/questions/5599/any-succ...
You get a GPL version of the code, which is free and nobody can charge you for, ever.
If you want to use the software in a way that is not GPL compatible you can opt for the other, paid for, license.
This works in some cases, but in the case of Godot (assuming we were to do something like this) is the MIT license, which already gives you the rights to do whatever you want.
And even in the GPL case, if the other license terms became too odious you could simply switch to the GPL version, and not pay.
https://github.com/SixLabors/ImageSharp/blob/main/LICENSE#L2...
https://news.ycombinator.com/item?id=33301518
Again, nothing stops someone from forking and maintaining.
Are you saying this is not legally enforceable?
The commercial licence is just an additional option for companies willing to pay to not adhere to the GPL terms.
There are certainly cases where maintainers have started licensing _new_ versions of a project under different, non-open source terms (Terraform, ElasticSearch etc.). But you're free to continue using any code that was released under the old licence.
There are some badly written licences which make it ambiguous whether the licence can be revoked in future (e.g. Wizards of the Coast with the OGL), but I have rarely seen this raised as a concern in a software context.
(IANAL, this is not legal advice, etc.)
Either way however then you could either stop using it or fork the project.
You can't retroactively pull an open source license ... whatever the new license is, if it forces you to pay, it no longer meets the definition of 'open source' [1]
[1]https://opensource.org/osd/
Now TFA gives hope to another dream, or a lie, the dream that magical open source fairies can let you keep getting away without a competent software team. It won't.
Find those programmers, build a team, build a company where this team has a place, adjust your business model (i.e. raise your prices) to where you can afford to pay them.
People think that making good games is cheap and anyone can do.
Making games, maybe. Good ones? I don’t think so.
My questions below might be idiotic. (but not intentionally so)
I have no idea how many apps reach above 200.000 installs total or $200.000 income per year.
How common is it for apps to meet the minimum requirements when the fees kick in?
I would guess that if you base the application on a subscription model that this will not be a major problem?
Also doesn't Apple App Store or Google Pay charge far more than this on income?
Imagine now you have 1 million installs. 2% (20.000) pay you 200.000 / year. 980.000 players pay nothing. Now Unity want 1.000.000 x 20 Cent = 200.000, you made ZERO with your game... crazy. That is an uncalculatable risk for a small company.
For many developers it is now cheaper to switch the engine than to pay for Unity.
Only for a (possibly very) small minority.
I find the the fact that they had no qualms about retroactively applying this to already released game much more infuriating (if they did this, what can they do next?). The pricing itself seems fairly reasonable if make more $1.5-2 or so per user (compared to Unreal anyway).
Previously you weren't even allowed to use Pro/Personal if you made over $200k (not just per game).
> Now Unity want 1.000.000 x 20 Cent = 200.000,
Not saying the whole price model is not stupid but only those developers who are very bad at basic math would pay this. Everyone else would upgrade to pro.
> Now Unity want 1.000.000 x 20 Cent = 200.000
Again, not really. Even in this case only a subset of user would pay 20 cent.
Another example demonstrating the incompetence of Unity's marketing department, they should've understood that most people can't really memorize more than a single number or be expected to spend over 30 seconds reading something (now I'm not saying that this change overall was not a terrible decision but even a significant proportion of people commenting here don't seem to understand how their per install pricing is going to work).
https://unity.com/case-study/marvel-snap
> Generated $100M+ revenue and 21M+ downloads on Android and iOS since October 2022
At $0.20 per install after meeting both conditions that would be a lot of money.
Note: I don't know the full terms but just took what most articles are reporting on.
i.e. $4,160,000 USD (# of total downloads - 200k * $0.20)
They also charge more at less scale and can't detect things like pirated installs, bypass Steam DRM, and you're still gonna hit the Unity servers. It's just checking device HWIDs.
But Marvel Snap is a card game and like many others has terrible monetisation, you can't keep up with card releases at all without paying. It's more like the pay to win games.
I personally think its dumb for Unity to even have the $200k install pricing. It's created a ton of bad examples and drama.
Should just be upgrade to pro at $200k revenue then only after $1m do they do install pricing.
No developers/companies who made more than $200k per year (overall revenue of the company, not per game) were even allowed to the use the Personal/Plus tiers and were required to upgrade to pro (which has 1 million install/revenue limits).
I don't think the itself cost would be unreasonable for at least 90-95% of all developers and if you average it out across everyone the proportion of revenue Unity get's would be still pretty low (not much more than 1% or so).
Most people seem to be upset because how they applied these changes retroactively on currently released games and because the whole model seems way to convoluted and not really thought trough.
That's why a model of free game with paid addons/extras/pay-to-win/etc. works so well – once someone already tried your game and gets invested in it, they're more likely to spend anything.
(Whether that model is ethical in the first place is a whole different matter. I don't think it is, so maybe eliminating it will be a net positive in the world.)
This pricing is per install, not per player. Installations are an unrestricted functionality which can be abused. If you care about quality games: this pricing causes the opposite.
> We treat different devices as different installs. We don’t want to track identity across different devices.
This allows malicious actors to fake the device identity, which leads to the exact same problem.
> We are not going to charge a fee for fraudulent installs or “install bombing.” We will work directly with you on cases where fraud or botnets are suspected of malicious intent.
This roughly says "trust us bro". They will "work directly with you" in cases of fraud? How nice, but that could just mean an automated response email with generic advice, followed by radio silence. No promise that they will resolve the issue, no promise they will work on it before your dev studio is bankrupt, this is just a collection of words built on trust they don't have right now.
[0] https://unity.com/pricing-updates
Unity also doesn't give a crap about quality unless by quality you mean microtransactions, milking players and ads/adware. The engine is associated (partly unfairly) with lagginess, low performance, high ram usage, and crappy games and assets flips. Unity is famous for how slowly new features come and how buggy it is in parts. They also scrapped their own game on Unity that was supposed to show how to use Unity right, it's an engine company that never made games (unlike Epic or id Software, who made games first, engine seconds, and even when being in engine business they kept making games). Unity users who just slapped few assets together basically created the asset flip genre, the opposite of quality. As CEO of EA Riccotello supported always online online DRM that shut down singleplayer games on release day due to servers dying, as CEO of Unity he said you are a "fucking idiot" (direct words) if you don't want to maximize your monetary return from your players. Unity bought an adware company, Unity has an ad network (and if you use it to put ads in your game, they waive the install fee, curious..).
The $0.20 is not retroactive, nor do you need to hope your users see 10 and impressions to cover a cost to the game developer -- you need to earn $200,000 _and_ have 200,000 installs. A free game would not be charged, and if you don't believe you can muster 10 ad impressions per user (to cover the unity install cost of $0.20), the base cost of the game can be increased by < $0.01 across your 200,000+ sales starting next year when the policy takes effect on new installs.
[1] https://twitter.com/unity/status/1702077049425596900
The other issue is how they measure "installs", as it has been a constant problem in the mobile community for years on end (a problem especially affecting the ad industry as you may imagine). I personally don't see how they will accurately measure the cases mentioned in the tweet without forcing everyone to have some authentication.
10 ad impressions per user might not sound like much, but you're going to get a lot of people installing your app, not liking it, and uninstalling it. The way the mobile game business works the "install count" means almost nothing, you need many, many, many installs to start making any money at all because so few people 'stick around' to actually make you any money.
Regardless of how they spin it, this model is very bad for the FtP games, which is the vast majority of games now.
I’m hoping you can answer these with concrete data. If that’s not possible, could you share your references?
2 cents with the pro-license only applies after the first 1M installs, tallied per month. If a game is under 100K installs per month, a pro license only lowers the install fee from 20 cents to 15.
If Unity just counted number of buyers they wouldn't have to deal with malicious installs and users of their engine would have more predictable costs. But for some reason they don't want that, and that alone is suspicious.
No matter what is the app.
This sales model will ruin indie devs who aren't so eager to swallow adtech's shaft that they're content with rolling out malware to their users.
Unity has published what "whatever they want" currently is. It doesn't matter if people get the details wrong on that because Unity can change its mind.
That being said, you're still more wrong than the author on the details. The change is retroactive in that it applies to existing applications. The author was very clear what they meant by "retroactive":
> this change will be done retroactively on existing applications as well.
This is true.
Code examples: https://bevyengine.org/learn/book/getting-started/ecs/
License is Apache 2.0 OR MIT:
Apache 2.0: https://github.com/bevyengine/bevy/blob/main/LICENSE-APACHE
MIT: https://github.com/bevyengine/bevy/blob/main/LICENSE-MIT
All other 3d games I found made with Godot look like they came from the 90s. 2d capabilities look great though. There's probably also some selection bias, it's probably not the first choice for most who want to make a 3d fps.
It's not a binary thing and we can argue semantics, but MS is definitely holding some keys here.
Big network effects
Having said that, I've been looking at it recently and it looks like it has everything I need. I'm definitely going to try it out before deciding between it and UE5.
What Unity offers is a solid way to build cross platform and target iOS. They're able to offer that because they have the resources to keep up with iOS. Open-source alternatives understandably struggle to keep on the iOS treadmill.
Epic released statements through their attorney's that they would consider the public version of Unreal a dead product if they lost the ability to maintain ios.
https://i.imgflip.com/5aw95n.jpg
Not saying it's realistic that those types of things will get very far, but the Apple monopoly and their activities is worse than DC in some ways.
$U upgraded to Buy from Neutral @ BofA; PT raised to $56
Apollo, Sync for Reddit, BaconReader etc
Developers / professionals are more judicious in where they spend their time / money / effort than most people. When I select a technology for professional use, I look at the long term prospects for the company. Open source is always a positive, as I don't have to worry about a company raising rates and trying to extract more money from us. Feature completeness can be a concern, as is ease of use, but we really only need our needs met and a path to meet our future needs.
I'm in general sketchy about OSS revenue generation if the devs don't collect $$ proactively but rely on goodwill.
Open Source software is great, but that also comes at the cost of support when you are running into a problem. Particularly if you are on a critical timeline, especially if trying to fix a bug after something is released.
You are at the mercy of the community being able to help (and most likely only being able to give them a limited amount of information) vs the possibility of the company you are licensing from sending you engineers to actually help you fix the problem.
For business critical this is an important thing that needs to be looked at, and from what I can tell there is not anything like this for GoDot (I don't see an enterprise option similar to redhat).
This is not in any way shape or form defending Unity, what they are doing is horrible and will likely (hopefully) cause a migration to Unreal. But just saying open source is not a cut and dry option either. Especially not for something as critical as the engine.
> While we don't have a formal structure for offering commercial support, many Godot developers can already help you, work per hour, or relocate to your company. For such inquiries, contact support@godotengine.org. Let us know about your needs and we'll try to find the right person for you.
How often do you think that happens for unity? _Maybe_ for some massive developers, but I don't think this would be very common for most customers.
This was like 10 years ago though. That guy worked on multiple arcade games in Assembly back in the day too (you've likely played games he worked on if you've played more than a handful of popular arcade games), he was no slouch.
Have you ever tried to use Microsoft "support"?
I'd love academics working on 3D to submit things that can roll into godot.
Using unity as the example there are some crummy assets, but there's also some incredibly popular and well made assets, many used by AAA games.
Compare that to web development (different beast I know, but there's no reason gaming can't be more componentized). Web development is so simple these days because I spend most of my time gluing stuff together, and sanding the edges.
Game development has concepts like this, and I have no doubt I'll get replies showing me some nifty component system, or talk about how I just don't understand ECS etc, but they seem tacked on to a monolithic walled garden. There's something magical about "npm install x" and adding a whole feature to your app.
Maybe it's because "it's art" and game development is just a passion project.
This does exist in the gaming industry, it is called "shovelware". "Shovelware" is when you just glue existing assets purchased for cheap and "shovel" them into a game for a quick buck. Usually not very complex games, low effort stories, etc. Also usually very cheap to download.
It's why in the 90s every game was mostly built on an engine specifically designed for that game, and in the 2000s they switched to prebuilt game engines. Saves a lot of time AND money.
What do you mean by monolithic?
Spin your view a little: NPM is your web development engine. Somebody had to write all of that code that undergirds NPM as an engine as well as the particular library you want to install.
Game engines are no different. They manage a bunch of (mostly boilerplate) game development components (sound, textures, rendering, animation, menus, collisions, physics, resource I/O, etc etc) so game developers can spend time "gluing stuff together, and sanding the edges" of their game. Monolithic engines like Unity have a plugin system for installing 3rd party libraries, and the engine tooling itself acts as its own level, effect, cutscene (etc etc) editor.
At the end of the day games are just like any other programming task. You have a limited amount of effort/time/budget, so you have to make decisions about where you want to optimize. Game engines, even monolithic ones like Godot and Unity (and, really, every 3rd party game development tool) are all designed to let you focus effort/time/budget on the unique aspects of your game.
If you are thinking about why the game engines aren't broken down into finer reusable components (e.g. one module for rendering and another for physics, etc.), it probably has to do with how these things are tightly integrated, and how every game engine has a different design philosophy on how the world should be organized. The different organizations might also mean that you have to design the games to fit a specific game engine, which is unfortunate.
Maybe you are lamenting why the marketplaces or asset stores are so closely tied to the game engines? For some plugins, they have to be tailored to the specific engines. But for arts and sounds, there are many free and paid options to choose from, although not necessarily of the same quality and cost as one might get from the asset stores that are sponsored by the game engines (e.g. Unreal Megascans).
There are still various tools associated with game development (e.g. Blender for 3D, your favorite text editor for scripting), and those aren't tied to specific game engines.
Take a look at: https://itch.io/game-development/engines/most-projects
And
https://www.gamedeveloper.com/blogs/the-generous-space-of-al...
If you look at both of these you'll see just how many engines there are and neither of these cover everything. There are plenty of engines popular in the Python community that no one outside of it are aware of. Such as Arcade [0], Python-Tcod [1], Ursina [2], UPBGE [3], and Panda3D [4]. But based on your description you'd really like https://gdevelop.io/. It embraces exactly what you're describing where you can build a game but just installing entire features others have made and put online into your game.
[0] Beginner friendly 2D library:
[1] Rougelike: https://python-tcod.readthedocs.io/en/latest/
[2] Beginner friendly 3D engine (built on Panda3D): https://www.ursinaengine.org/
[3] Blender Game Engine Fork: https://upbge.org/
[4] Highly flexible code first 3D engine: https://panda3d.org/