Unity 2D vs Godot 2D

I think the development of FOSS has to be slow and cumbersome because everything contributed must go through process of peer review, right?

I guess theoretically if enough people contribute and decision making power isn't allowed to concentrate, you'd end up with some stellar software that way.

 

Commercial companies do the same.

 

In the end, organization matters. If the project is organized in a foundation or something similar - like Blender ("Blender Institute"), the contributors responsible for a component probably meet every morning (in voice chat) to briefly go over all tickets and especially open pull requests just like we do at my company for example.
If the contributors are just "random people on the internet", then I can well imagine that it goes slower however..

 

Company has incentive for profit though. Changes the decision making process.

 

Not really when it comes to peer review, aka code decisions. Management is not involved there (albeit we as devs tend to wish the management could at least understand a fraction of the code we write, lol).

It likely influences the general projects vision and long term decisions, yes. Though reality has many gray zones. It's too short sighted to think companies will always decide for profit and foss projects will always decide for user happyness, or something like that.
Foss projects can strive for popularity and commercial products profit from user happyness too as it binds customers to hopefully buy more or pay regularly if the product is deemed good.

 

How do you think I'm thinking?

 

How do you think I'm thinking?

 

You are referring to peer review which in the context of software development means someone vets the incoming code to verify that it won't cause any problems. If you're referring to something else that's on you for calling it peer review.

 

so when a community of passionate nerds review contributions to their pet project, that is essentially the same as shareholders in public company deciding what stays and goes in their product?

one team is on deadline and motivated to grow exponentially in any way possible. other has none of that baggage.

 

Okay, you're clearly misunderstanding what a peer review consists of. How about you explain what you're thinking?

 

Someone adds something to the project. Eventually everybody else involved gets around to trying it out. People gripe about things that don't work/cause problems.

Eventually problems get fixed or culled.

That's it.

If there is deadline and profit motive, that process may be expedited, foregone, or may a good product may even be stripped of its quality features in order to make it more profitable.

 

Shareholders wouldn't be involved in any of that. In fact the idea that they're entitled to a say at all is a myth.

https://www.washingtonpost.com/outl...d31f90-14b8-11eb-ad6f-36c93e6e94fb_story.html

The exception being in cases where one entity owns a majority share. Like Unity owning 3/4ths of ironSource.

 

That is not entirely wrong but fairly black & white thinking.
Also ANY project needs some sort of lead. I have not contributed to public FOSS projects yet, but I'd be very frustrated and leave the project if I worked on some set of features I thought of myself (because no leadership decided what features we want), and then those were rejected with a 60% majority or something.

 

that's part of being on a team though, isn't it?

Who canned Gigaya? Why so many unfinished projects from unity?

 

My guess is that's ol' Johny boy and his ilk. They're steering the ship. And they're not very good at it.

 

Unity management. Shareholders had nothing to do with any of that.

 

Who motivates unity management though?

 

In my world, leadership is always responsible.

If engineer bob farts and spills his coffee which frys the computer that held all of the projects code, it's managers fault for not preventing the causal chain of events which could lead to such a disaster, and whoever is above the manager is at fault for not properly setting the vision which included motivation and funding and whatever else is necessary to ensure a team which doesn't suffer problems like this.

 

Go read the article that I linked to you. While you're at it I recommend reading up on how businesses are run at the corporation level and how investing works.

 

i read it. dont understand what it has to do with what i'm saying.

you said i was wrong, i wouldnt be upset if you showed how but i'm not seeing how what you suggest is true: that a FOSS's incentives and a public companies incentives are fundamentally the same. You didnt say that, but you are saying my statement is wrong and not say what is right, so what are you saying then?

 

I'm saying you have a misconception when it comes to shareholders and who is affecting who.

 

i wish when people told me that I was wrong about something they'd make some effort to explain why. why bother just picking out a single detail and saying, "nope" without further explanation than that? How do we know I don't know more than you do, or vice versa? Whats the point of such stilted conversation?

if you link an article thats like, something, but if you can't at least paraphrase a few key points this is like the conspiracy nuts shouting, "do your own research!" which just means that they read a bunch of stuff made them feel special, they didnt actually reach a conclusion on their own, and they dont have the intellectual energy to actually challenge any of the conclusions they were spoon fed.

Now I know that's not Ryiah because ryiah is one of the heroes of the unity forums and a wealth of knowledge, but that's what it seems like if you just quote a post and say, "not true."

 

Since you appear to have missed it the first time here it is again. Two additional topics that it touches on are shareholders only focusing on profit and companies only focusing on increasing shareholder value.

https://www.washingtonpost.com/outl...d31f90-14b8-11eb-ad6f-36c93e6e94fb_story.html

 

yes, it is a lot of words to say, "some shareholders may have other priorities beyond profit."

So from that you suggest... what? development of public company products is fundamentally the same as FOSS?

 

Yes. You have a core team that manages the project that calls all of the shots and the developers that work on the project submitting it to the core team for review before integration. At least for the major projects like Blender. I'm not counting small time projects.

 

and you dont think the ever looming stress of needing to make a profit to feed people or keep share holders (just one factor in need driving profits) effects the decision making process?

 

Unity only somewhat recently started making a profit. During the first year of IPO they were still losing money but that's the time when they gained value. In September 2020 when they formed it was ~$60 a share. By Feb 2021 it was at it's previous typical sitting value of ~$90 a share. Again during this entire time they were in the red.

https://www.gamesindustry.biz/articles/2020-08-25-what-did-we-learn-from-the-unity-ipo-filing

 

Mh yeah no, doing work that's then scrapped is something that should be avoided when possible..
Naturally that happens often enough in companies too, but relying solely on code reviews as a way of decision making for the whole project sounds like a way to make scrapping work an inherent part of the process.

Are foss developers actually living off air and love?

Yeah that makes more sense and I'd tend to believe that many pure FOSS projects tend towards such structures once it's not purely the hobby of some few people anymore.
The Blender Institute surely has some parallels with regular companies.

By the way, Blender has a sort of membership program. It's officially donations but you pay regularly and that's a major part of what they live off.
So that naturally results in pressure to succeed and also don't disappoint by breaking delivery time promises (-> deadlines) so people do not cancel the subscriptions.
https://productmint.com/blender-business-model-how-does-blender-make-money/#:~:text=Blender makes money via donations,downloaded over 14 million times.


EDIT:
Also is this the wrong Godot or are they also a company of type "For Profit"?!
https://www.crunchbase.com/organization/godot

Honestly I'd be way less inclined to put all my cards on them vs Unity, knowing that. ("For profit") companies that do not have a product which on its own generates revenue, tend to either disappear because none lends them any more money, or monetize their product heavily once they have a large enough, dependent userbase (because that's why their investors did lend them that money in the first place).

 

They have to be listed as a For Profit because they aren't a non-profit, which comes with a wide variety of legal requirements and different tax structure. Blender.org is also a For Profit company.

 

According to their webpage they are a non-profit:

"Godot Engine is a not-for-profit, community-driven free and open source project. It is legally represented by Software Freedom Conservancy, a USA-based charity that helps promote, improve, develop, and defend FLOSS projects."

Source: https://godotengine.org/donate

 

A not-for-profit and a non-profit are different things.

 

so, you ever work on any teams?

currently i work for myself and i scrap probably half the stuff i make. You do everything right the first time?

 

Isn't this a strawman?

 

whats that got to do with incentives though?

the only thing i'm talking about this whole time how certain factors (liek the need for profit) drive decision making process.

you guys seem to suggest that in fact, a FOSS project like godot does share similar burden. I'm not convinced on that but it's kinda besides the point anyway.

it's more just a hair brained retort to a hair brained reply. Completely derailing the topic from its focus, drawing silly conclusions from my own experience and missing the larger picture.

 

This entire discussion started because you felt that FOSS was different from running a software focused business.

 

well you offered some evidences but i just dont see how it goes against what i'm saying. probably the first thing i wrote that you quoted i could put more explanatory details in but ya know, if you really care you could just ask or something.

anyway, i need to get back to work

 

Fear you have a slightly strange impression of what working in a team implies ._.'
It does not mean every member does what they personally think is right and then after 3 weeks all get together and decide whose code to keep, lol.
I am in an even above average sized one (which causes some overhead sadly), but while naturally there are change requests on my work during pull request, I never had to scrap more than maybe a few hours work since we have exchanges 2-3 times a week with the "product owner" and "technical lead" who keep the oversight over our part of the project. They are not the share holders by the way.

 

This "Godot is open-source" thing is just bullshit. It's a scheme to make money in different ways.

Actually, part of Godot's business model is to provide you with the ever incomplete product: a game engine that cannot work officially with consoles.
So one of its founders sells you console support via its company. That it's the only company mentioned in the official Godot docs.
https://docs.godotengine.org/en/3.0/tutorials/platform/consoles.html

Look what ridiculous excuses they have.

That would be like Blender would release software only on Nvidia GPUs and make you hire a company to port it to AMD GPUs. Or port to it to Mac because "MacOs require specialized hardware provided by Apple".

And we were complaining on Unity asking subscription to have a dark mode in the editor

 

To be fair, that's old documentation, later revisions added another company.

 

I think this needs to be emphasized more. They're not selling a service where they port the game for you and deal with the major headaches themselves. Like their FAQ states they simply "give you the code for you to port".

The FAQ adds that they "will stand by your side with [their] support" but I can't imagine it's much if they're able to start the price at just $3,000 USD. I can't find any information on their business aside from the website mentioning just three people and even then it doesn't say anything about their skills.

 

No, it is because it is mostly done by unpaid volunteers. The other lack of absolute leadership which has the final say. (even though there IS a core team, you can always attempt to fork the project).

Absolutely not. Shareholders are interested in profit only. Review is concerned with code causing trouble only. They're pretty much orthogonal to each other. Shareholders do not do code reviews.

If 60% of all contributions are rejected in a team, something is wrong and someone probably needs to be canned.

A skilled programmer with past history will get it mostly right on the first attempt. What he doesn't get right on the first attempt, he'll fix during bug testing, the rest goes to code review.

A team means that you're assigned with a task, and you deal with that task. Your submit your commit, it is reviewed by your senior, and if you screwed up, you're told where and it is explained to you why you must change your ways. The idea is not for multiple people to work in parallel until one of them "wins". It is a conveyer. And not a competition.

Also "everything right the first time" is an odd thing to say. A task requires an amount of time to fix it, and not getting it right on the first try. if you "throw away everything you did", then you scrap every change and start over. And this is not something that is common.

 

do you think that is by design?

i dont really care, never gonna use godot. Just curious.

But i can hardly even fault them - this is pretty much same thing all the games are doing nowadays too. Sell you something that requires you to pay more to get the complete product - and make sure not to mention that up front.

To even get approved to publish on consoles, you got to be like, a legit studio don't you? I had impression godot was just for tiniest 2d games - like real hobby level stuff. But i don't look it at all.

 

there is more than just write code to do a thing. The whole time the design of the game is changing, new things are discovered so you change to adapt to that... I can't imagine anybody making a game that isn't a direct clone or the fiftieth in a genre they make continually just writes a gameplay system in one go and never has to touch it again.

I doubt you could even really quantify how many times you touch something over the course of development.

Just my experience of course. Interesting to hear what others are doing.

the other stuff you quoted - not sure if you mean to or are just offering your take - but you are agreeing with what i was trying to say. I was paraphrasing what i think others were attemtping to say, because they weren't just saying it, rather just telling me I am wrong but not why.

 

Rejecting sixty percent still sounds insanely excessive.

Imagine you were given 10 million budget and then threw 6 million into the fire. That's what 60% being thrown away means.

To minimize amount of scrapped staff it is common to use iterative design. Meaning you start with an inexpensive prototype that is cheap to scrap, and only polish it when you've settled on the design.

Unreal demonstrated that in their level design docs for content examples.
https://docs.unrealengine.com/4.26/en-US/Resources/ContentExamples/LevelDesign/

 

Have you read a recent blog post concerning export to consoles? https://godotengine.org/article/godot-consoles-all-you-need-know

The blog post sounds reasonable. And the comparison with Blender is an ill fit. Blender has way more funding and Blender also has a legal entity representing their interests, something that Godot doesn't have and it appears also won't have in the future, which in of itself could be taken as a red flag, however. And Blender has the same constraints - open source code can't integrate proprietary SDKs. Blender's FBX implementation is reverse engineered, they can't implement Autodesk's FBX SDK. It's why FBX support in Blender has been problematic even up to this day. I don't think anyone's reverse engineering proprietary console SDKs anytime soon (or ever).

Also, the co-founder running the publisher/porter business does not participate in Godot's development anymore, I think? And porting to proprietary console SDK with full native support is expensive.

 

Godot is very much like Unity in that it started life as an in-house engine, but unlike Unity they decided they didn't want to profit off of the engine so they simply released it onto Github. Eventually the engine took off in a way that he didn't foresee and the donations enabled him to become a full time developer of it.

https://archive.fosdem.org/2019/interviews/juan-linietsky/

 

Official Spine runtimes are now available for Godot:

 

I appreciate you chiming in, we don't have enough people with proper hands on Godot experience here! It sounds like you have yet to get the full Unity development experience where you'll be fixing bugs in assetstore assets, trying to get the support team to acknowledge editor bugs etc. instead of contributing to the engine directly, is that correct? I'm hoping you'll give Unity a shot and will update this thread in a year or so with your experiences.

I still wonder why that one dev on Youtube (forgot the name) switched from Unity to Godot 6 months in if Godot is as bad as you (and others) say. I haven't used it yet and haven't really used the 2D feature set of Unity either, but from my research the pain just looks different, not greater on either side (unless you need console support). Maybe a matter of taste and unique project requirements?

I wonder how much sense it would make to start prototyping a game in Godot and then port it over to Unity once you're sure you want to commit to the project and have a good idea what and how you'd need to implement it. What do you think?



And a hypothetical question, if someone was to A/B test Unity vs Godot for a project that doesn't need to be finished any time soon, which version of Unity vs which version of Godot would make the most sense to test? Latest Unity vs Godot 4 alpha? Old 5.x Unity vs latest Godot 3.x?

 

Godot is pretty good for rapid prototyping, but it's different enough that you wouldn't use the same architecture in both engines, or rather you can't since Godot is not just another Unity clone and it's a lot more opinionated in how projects should be structured. So it's good for testing out ideas, but the implementations would be nearly completely different.

Since it's not time sensitive, Godot 4 beta is releasing soon in about a month. I'd go for that and compare to latest Unity tech stream. No reason to A/B test old Unity which is not supported on many platforms anymore due to outdated SDK support.

 

Take into account that he's in a personal crusade against Godot, and that his opinions stem more from ideological reasons than practical ones.

Now, I'm not saying he's neither right nor wrong, just pointing out that he's far from neutral in the subject. From reading his posts here, in Twitter and in the godot GitHub, he's as biased as your average fanboy, just in the other direction.