Unity is not worth using right now

I'm only one man, but you've failed to convince me of the latter. Picking between two graphical pipelines is, while certainly not ideal, not in my mind "too complex." I've already pointed out that most of the packages don't present new or confusing options, and that it's quite easy/normal to simply ignore options you're not using and don't understand (we all do this all the time in every other piece of software).

 

I think you're missing bunch of zeroes here.

That's normal, though?

Someone who is unable to search for necessary information online is not going to achieve anything in a computer related field.

It is prerequisite for making anything using any tool.

 

You really like throwing Combinatorics arguments at everything don't you? The problem is they're a fallacy. No one starting a project puts 13452 options on the table.

 

Unity could. It is called 'verified' packages and guess what? They're already doing that. I think, as usual, you don't know what you're talking about...

 

Isn't this roughly what they're doing with verified packages though? Maybe it's not as effective as it should but they're doing it, in theory.

 

So is there a full packages stack that are verified to work with each other and guaranteed to have bug fixes applied?

How does that work, are there verification cycles or alpha, beta, release, verified phases to a packages lifecycle?

And if your new to packages/Unity how does the UI/UX help ensure you only add verified packages?

 

I'm not sure, what we see from the viewpoint of the user is that there are at least three (four?) stages:
- "closed alpha" - when the packages only available through modifying the package manifest manually
- preview - it is publicly available but it is preview and you need to check the checkbox to install
- none - it is available without the checkbox, but it is not verified for your current Unity Editor version
- verified - verified for the current Editor version

AFAIK the difference between the "none" stage and the "verified" stage is that they finished all the unit- and other tests and everything is green in the given unity version.

IDK if they are testing packages together or not during the verification process, I guess yes, because the problems between preview packages is relatively frequent, meanwhile we haven't seen a single case with verified packages.

Have you seen the package manager UI? There is a checkbox either in the menu or in the preferences (depends on the unity version) where you can toggle off the preview packages. Otherwise it is your choice what you install or not, it is not Unity's job to save you from your own decisions.

But again, all of these are public, available information, if you had worked with packages, pay some attention, you would know.

 

Have you ever made non-trivial software outsideof Unity?

It's pretty normal for libraries to only be compatible with certain versions of other libraries. It's something any decent developer tries to minimise* but, for example, I can't test my library against a new version of your library which hasn't been released yet, so there are limits here.

Unity's own packages have the benefit of being in-house, and so far (even with a handful of "preview" ones) I've not had any issues so far just grabbing what I need when I need it. That said... I do grab just what I need when it's needed, which significantly reduces complexity in comparison to the "go on a shopping spree when you start the project" approach.

* You see Unity doing this when things are marked "legacy" and/or "deprecated" for significant periods of time before actually being removed. Being "deprecated" in particular usually means that something is being maintained specifically to give people time to re-work parts of their own stuff that's dependent on it so that it's no longer required.

Just how far back do you expect Unity to go? Mail me out a manual which tells me how to open my web browser and navigate to unity.com? Should it explan my keyboard and the power button, too?

It's perfectly ok to expect users to have some pre-existing knowledge. Game development is an advanced topic to begin with. WIthin it there are quire a few areas which are more complex or nuanced than others.

I do agree that Unity would benefit from having an info repository with a whole bunch of "foundational" knowledge for people to look up. I don't agree that they should design their tools to hold hands for people who don't know what they're doing. This stuff is easier and more accessible than it's ever been, in large part thanks to Unity's own advancements in availability.

 

Oh, right. The point here is that if Unity is self-aware enough to be able to tell you what to do then it may as well do it itself. The logical conclusion of some of what you're suggesting ultimately leads to a "make game" button.

That is of course all based on the incorrect assumption that there's "one right answer" as to how to make a given type of game in the first place. Just because Unity knows I want to make a 2D RPG with X, Y, Z features doesn't mean it can spit out the "one right answer" as to how to do that, because there isn't one right answer in the first place.

 

From my perspective, the major problem I've had with Unity the last couple years is hitting the moving targets Unity props up then pulls down. Post Processing stacks, render pipelines, dots, realtime GI, the list goes on.

My hope is that after the DOTS transition (Unity 2022?) and the completion of the RPs things will be in a much better spot. I'm also hoping it will settle down after the IPO when they don't have to convince future investors Unity is a tool for everyone in every industry.

Also, unlike some in this thread, I absolutely love the Package Manager idea. Some things could be a bit smoother and better presented, but overall it has been a pleasure.

 

No what I'm saying is that you are starting out on a game development journey with Unity and you have an ideal destination in mind and you want Unity to help ensure you start that journey with the best equipment and directions possible.

If Unity just goes hey here's some stuff with odd names just pack it all and see how you go, any direction is good...



How many developers then get lost on their first journey or tangled up along the way only to give up in frustration.

And if we view developing with the Unity game engine as a landscape then it's full of 'devoloution' game engine events that change the landscape and can make finding your destination game a nightmare or a pleasure.

Maybe Unity could even setup base camps that help you get to your target game by having the best tools for that style of game already setup.

Ideally they could have good prototypical generic builds of most game types already working. After all Unity have been a game engine technology for about 14 years now more than enough time to have great demos in every category of game.

 

You contradict yourself. If there was an objective "best equipment and directions" then that would be the "one right answer".

No. Ignore all of that. Just finish your darn game. With occasional exceptions you only have to care about stuff that helps you. Just because Unity does a thing doesn't mean you have to implement it into your project. Don't install the package, or don't download that update, or don't use that new feature.

If you're here to finish games then don't let the shiny new toys distract you. Be aware enough to know if things might help you, and from there only engage with the stuff that helps you get the job done.

 

In all probability, no change to the total number against the ones who would give up in frustration over something else. All the hand-holding in the world wont substitute for having the will to overcome obstacles in the first place.

Your core assumption is basically that as long as you provide all the right ingredients kits, then everyone can be a chef. But no, having flour, butter, milk, eggs weighed out for you wont help you make choux pastry. And you'll sure as hell fail at it a few times. Whether you learn to do it or not is not predicated on whether someone saved you the effort of sourcing the ingredients, its predicated on you learning from failure and not giving up in frustration.

Oh, and for the record, you'll find 'ingredients kits' for hundreds of kinds of games on the asset store. If you, personally, want all your projects to start from some sort of premade template, noone is stopping you, least of all UT who have provided a nice mechanism for you to get hold of them.

 

Using your analogy it would be here's a kitchen some of the things in it are prototypes and some are verified usable go ahead and make your dream Choux pasty have fun.



Wouldn't you want just the ingredients and utensils you need as well as some help a recipe and guidelines on how to make the best pasty?

 

That's an extremely unrealistic expectation.

I think that people who want the best equipment and directions are the ones that will usually achieve nothing in the end. Because "perfect is the enemy of good". You can continue looking for the best tool/book ever and waste months, or you can learn something usable right now. If you start right now, you'll be far ahead from the dude that still keeps looking for the "best tool" while having done nothing.

No. When you know exactly what you need, and you know that there's a recipe for that thing and you know that other people cooked it before, then you buy that thing in a bakery instead of spending time to make it yourself.

Cooking is for custom stuff you can't easily get.

Spoiler

Now I wonder how many people cook by dumping enough ingridients till the dough tastes right and then end up with a loaf of honey bread instead of pancakes they originally wanted to create.

 

You fundamentally misunderstood the point, didnt you?

Being able to to make the best pastry isn't about the ingredients or utensils.

But hey, if you ever need surgery, Im sure you wont care whether its the janitor that does it or the specialist, as long as they have the best tools and theatre, right?

 

Should I expect Sonar to have a template for every genre of music? Should I expect visual studio to have a template for my app that uses NetMQ to interface with Python on a Pi? Should I expect Microsoft Word to tell me how to write a climax and resolution?

Are you being serious right now?

 

I will leave this here. Just because.

Spoiler: Video about one guy who decided what to use and how he decided.

 

For mobile developers, they often have to upgrade in order to support their target platforms whether they want to or not. This is a problem because Unity isn't just adding new features, they're also changing or removing old features that are still perfectly fine. I still remember all the work I had to do when Unity deprecated the old particle system in favor of Shuriken, particularly because it was missing certain features the old system had, such as autodestruct.

Well here's a new example I only just found out about, apparently Unity is removing the Display Resolution Dialogue and there's no replacement.
https://forum.unity.com/threads/display-resolution-dialog-deprecation.793611/#post-5531455

Apparently the reasoning is that nobody uses it and if you want to do quick input rebinding, just use the new input system. Well what if I have no intent of using the new input system? Because as you say "don't install the package". And even if I was using it, I'd still need an external way for changing game resolution/graphics settings, since that was also part of the dialogue. And what about artists with no coding experience?

BHB: BioHazard Bot is a finished game released on Steam, my game. Although it is set to hidden, the Display Resolution Dialogue is still enabled and pops up if you hold alt or shift. This was necessary because if the player set the graphics settings higher than what their system could handle, there needed to be an option so players could change stuff externally. Skyrim does this for the same reason. This also made it much easier for me to test the game on different computers, as well as test out quick one-off prototypes. It works great, zero complaints.
So....what? if I want to keep using the Display Resolution Dialogue, the only solution is to never upgrade to Unity 2019+?

That'd be great if they'd told us that when 2019.1 first came out except they didn't.

So much for evolution. The problem is that because of Unity's switch to yearly releases and multiple updates per year and changing things so frequently, it's impossible to just make a list of all the work that will need to be done so we can plan around it. You think you're in the clear and then at the last minute Unity throws a curve ball. If Unity said "Ok guys, we're making Unity 7, it's coming out in 20XX, here's a big fat list of all the things we're adding/removing/changing", then ok that'd be one thing. We can plan around that and budget time and resources accordingly.

But right now? First it's ok we're adding this. Then the next update a few months later ok we're removing that. And then just when we thought we were in the clear, the next update a few months later, ok now we're changing this.
Already upgraded to Unity 2019 because everything was fine at the time? Well too bad brah! I know we told you one thing when 2019 came out. But we changed our mind and are now removing it entirely.

Even Autodesk doesn't pull this bullshit.

 

True. I was speaking largely towards Arowx's assertion that things don't "just work" any more, largely because there are too many packages and apparently that's confusing.

Different food requires different ingredients and utensils. If Unity provides me with just exactly what my project happens to need then chances are it's not a good fit for whatever you're doing.

 

That's actually a first-ever good Youtube comparison of these engines. Usually, it's dry comparison like they would read feature list on engine websites and not truly informative for beginners
Thanks! I've bookmarked to share with people later.

 

Yeah, I don't agree with him on a lot of things, but he has his preferences, I have mine. And he isn't afraid of articulating his. That's why I liked it.

 

Then why does Unity often produce near perfect demos that are exceeding large and way beyond protoypical level art and models? Wouldn't prototype lo-fi game demos be better as launching points for new developers (they would also be much smaller to download).

For instance could developers learn more from a low poly Heretic, Blacksmith, FPS Sample or other Unity demo and in less time than it would take to download the AAA version?

 

If you meant "for what purpose?", then it is for the advertising purposes.
If you meant "how do they do that?" then they do that by throwing a LOT of money and man-hours at those demos.

Speaking of which, they could've made those demos in ANYTHING.

I believe there's nothing to learn from those demos aside from "this can be done". "How it can be done" is in the manual.

Additionally, low poly heretic is not something worth showcasing, as the point of the demo is image quality. Low poly heretic would be just "Another World" intro and those could be done all the way back in 1991.

Spoiler

 

No. Unity's demos are not targeting new developers, but rather studios with the money to create content of that quality.

 

Yeah, that.

A beginner/newbie won't have a budget to make Heretic.

Now, if unity made a video showing how to create Heretic from scratch in 2 hours, and that's without using premade models... that would be a different deal.
Sadly, that would require sci-fi level technology.

 

And sadly, even with that sci-fi level tech making things super easy, it still wouldnt reduce the amount of "I have never used unity or made a game before, how do I make an mmo and why no mmo tutorial on unity website?!?!" threads.

 

Doesn't seem that a good comparison. No actual details, but lots of vague statements like "everything just looks better in Unreal by default" without explaining.

The reality is, they both open with basic scenes, or are empty. They both look bad. Unreal has some post-processing by default, but so do Unity's starting project options. They both look same.

I've yet to see a youtuber compare them properly.

 

Couldn't stomach it.

Actually this is awful and the information is incorrect.

Spoiler

Aside from the guy trying too hard to look hip and trying to make sure the interiors of his nostrils are visible in the frame, there is incorrect information right from the beginning.

For example, the characters in the opening shot do not look like they were made in the same renderer.
Unity has more flexible shader system, and actually unity supports more complex shaders.
.... too much noise, too much posturing, too little information.

I wrote a comparison before. It is outdated by now, though, and my opinion has changed on some details:
https://forum.unity.com/threads/i-have-to-let-go-of-my-frustration-with-unity.382308/#post-2487046

 

And even then a general comparison like that is kinda useless. In the end it matters how good are the features you actually want to use in your game and in the way you want to use them, which is an awkward thing since you usually have to spend a lot of time in an engine to figure that out.

I can say, that for the features I am interested in using, Unity seems to me like they stop polishing them way too early. When Timeline first released I tried it and thought "Wait, this isn't ready". Now 3 years later, I tried Timeline again and thought "Wait, this is almost the same as 3 years ago". At this point any complaints I have are more likely to be addressed in a new tool rather than in an improved Timeline, because... I guess just improving already existing tools isn't good marketing for Unity?

But to get back on topic /\ The above complaint took me years to figure out and can be summed up as : Every feature Unity produces feels like the team that made it insta-quits as soon as its made, and then someone else half-heartedly maintains it for a while.

Everything else, default shaders, post processing, Asset Stores, C# vs C++ don't really matter, I can adapt, they don't really matter, and those are usually what those Youtube comparison videos focus on.

 

Well. Literally the takeaway from that video is: "I dunno, try them for yourself on your problems and you will see".
So I have no idea WTF?

Also, keep in mind it is a video for absolute beginners from an absolute beginner. Basically a note from his journey. He told what his preferences were and how he found out which engine fulfilled (at least ostensibly) his needs and aligned more with his preferences.

LOL, and I don't even agree with him on most things...

 

What's the point of making a video if the message is "try them yourself"?

------

Either way to me that the dude in the video feels very fake, and doesn't tell a lot of useful info.

 

Well, for people who feel intimidated by the choice or have choice overload, it is useful to show how he decided. You probably don't really care about this, you're working with complex systems for so long that you don't feel it anymore.
Me too. But I have a daughter and she's learning *some* computer-related stuff in college, I see this first hand on her. She is intimidated by the choices among unknown stuff. It's rather natural to feel these things, we can feel these too when we are up to do the same choice on fields we're not familiar with.
For example I have none to very little knowledge on cars. Now of course I have a vague idea how they work. But even when I was forced to choose a new car I felt a little bit uncomfortable because I don't know them. There were guys who told me how much time it takes from 0mph to 60mph measured in seconds. Great! Data point. Do I care about it? No. So useless crap. Mileage? Finally something! How many average person fit in the car? Useful! "It does look good" - Useless (Unreal looks better out of the box is exactly like someone tells me that the Teslas look good... matter of taste and I don't care about that).
But I watched some videos how guys actually chose their vehicles and it was good, got some ideas what I can care about and how their thought process was when they did. I didn't care what they chose, but how they chose it and what was the reasoning behind it.
Does it make more sense?

Well, the good thing is that this guy speaks to his hipster audience on his very "fake", nostrils-show-off way, so you don't have to.

 

I upgraded from 2019.2 from 2018.x a while back, and so far have been pretty impressed with the stability of the engine since then. Only weird issue I've had is sometimes the editor would hang when I closed it, but this seems to be gone now.

I'm on 2019.4 and quite happy, and I've built fine for iOS, Android, WebGL and Windows IL2CPP.

For me its very well worth using right now, I haven't touched any of the SRP or DOTS stuff, because I'm focused on releasing v2 of my game, but will definately tinker with it after release. Excited to see what my game can look like with HDRP on desktop, and of course if the lightweight pipeline can help mobile performance.

Getting my game to look good on even low-end mobile devices would be great!

 

One of the things I dislike most about the upgrades is compatibility, iOS xcode and android studio get insanely more larger by the day and this introduces small bugs, it would be nice to have a strip down version especially without all the emulators under the sun.

 

Actually, this Youtuber's approach is very reasonable. He does explain that you can't make a game without programming. Visual scripting is coding, etc.

He insists on learning basic concepts of programming with Unity and C#, which is excellent advice. C++ (even with Unreal framework + Resharper++) is intimidating and pushing for total beginners who understand concepts of class and float. And any visual scripting allows you to code without understanding what you're doing, so any scripting language or C# in Unity is a good choice to learn basics on your own, at home. "And then you can try Unreal if already understanding basics of programming".

This guy presents a very good approach, try things, think about what you need and how you want to work. It's way better than telling people "all engines are good, here are features, try it for your own" and then beginners aren't closer to understand what's better for them. Or how to judge, what's more suitable for their needs.

He's even like "yes, it's possible to make Firewatch in Unity, but it's not possible to me now, so I'm making the first game with Unreal". Another person could be like "OK, I but want to learn how to handle Unity like a programmer, I'd like to focus on that and build/assemble my tools". Which he also said, in some way...

Nice "voice in discussion", providing more informations than just "don't use Unreal for 2D, use Unity for mobile games, use anything for 3D" which basically beginners can get from the most of such Youtube comparisons

 

I suppose you have a point here. I can't quite relate to "being intimidated" part, as "being annoyed" is something I experience more often.

I understand that this is a stage persona, the issue is that I see mask nearly falling off. I personally find that irritating. Personal preferences.

 

What do you need XCode and Android studio for?

You don't need either. There is an iOS Project Builder on the Asset Store which is all you need to build and deploy iOS apps from Windows, I don't even open my Macbook anymore, and thank goodness for that.

 

Do you still need an IOS developers license or is it free now?

 

You need the developer license still.

 

Just wanted to share that I personally find it worthwhile to use unity. My main reason for using it is that it is possible to do a lot in it despite not being a coder, or very gifted at figuring out technical things. Like it's even possible for someone with almost no prior knowledge of technical things to play with something like machine learning in it. That's pretty awesome.

 

@neoshaman: I think the old thread where you mentioned Godot is closed by now, so I'm tagging you here. I started reading through the docs and watching a couple videos about Godot and it sounds like for the pure 2D project that I have in mind it's actually a better fit than Unity. I'll let you know if I get anywhere with it.

 

The thing is though, I agree with @Arowx on it being confusing. Old input system vs new input system. Shuriken vs VFX graph. HDRP vs URP vs Built in. There's mecanim and then like 2 other new animation systems in development, instead of just integrating them into mecanim. Unity is working on a visual scripting system while also acquiring Bolt. Resources/Asset Bundles vs Addressables. 3ds Max has the same problem where if you want to rig a character, you can use Biped, CAT, or make a custom rig instead of having one unified system and it's confusing to a lot of people.

And speaking of Addressables, some of the newer things in Unity like the Playables API and Addressables are like trying to decipher the necronomicon. I recently spent a couple days trying to figure out Addressables to create a streaming system similar to the one used in Metroid Prime. Eventually I gave up and used Resources folder and got it up and running in a few hours and it works flawlessly--but Addressables is supposed to eventually replace Resources and Asset Bundles.

The number of packages is a problem/confusing because what happens if Resources gets removed like Display Resolution Dialogue in an update or two? Am I making a mistake if I build a game around Resources folder even though it works perfectly and at the moment there's no immediate talks of it being removed? And if it doesn't get removed, does that mean I might be wasting my time if I were to stick with Addressables?

I don't know about you, but before Riccitiello took over I never had these kinds of concerns with Unity.

 

I don't have them now. I agree that there are some pain points at the moment, but the ones you've listed aren't the same as mine.

Personally I'd much rather that they replaced things from time to time and didn't feel obliged to keep carrying their old baggage along with them. A single unified system that does everything isn't necessarily better than choosing between more specialist systems. I don't find those decisions confusing.

Yes. The documentation is pretty clear that we should avoid it except for in very specific, limited use cases. I wouldn't want to ship a game which made heavy use of Resources for large assets.

Personally I found Addressables nice and easy to get up and running with. The catch is that there's no blocking (ie: non-asynchronous) way to load stuff, so in the rare cases where you need something loaded at the start of the first frame of your game... it's just not an option.

 

And there's no way at all to call synchronous load when you need the asset loaded before proceeding with the gameplay?

 

Not that I'm aware of. The developers did provide an example of how to use it to load stuff synchronously, but my experience with it was similar to that reported here.

If we'd designed with it in mind in the first place we could have worked around it, but it would have added unnecessary complexity even then. The developers seem to think it's a non-issue because asynchronous is better, which is true... in the cases where it's an option.

It's definitely one of the areas where Unity is causing pain at the moment. Don't use Resources, Addressables are the way of the future... but Addressables don't cover some valid cases, so I roll my own thing to make up the difference. Please note that I'm not saying that choosing between them is confusing. It's not. My concern is with the design of Addressables not covering certain important use cases.

 

Hi @scoat2 ! first Wellcome to this forum. I'm a user like you. Yes, Unity is full of features of any kind. And at the beginning, it can be overwhelming. More advanced ones are in development. Usually, I try to use the built-in default ones. Because they are robust, tone of tutorials, assets, and with fewer surprises workarounds. Universals and HD are options that are in development and you will not find that many assets or youtube video tutorials as the default PBR. But if you are an advanced user then it is recommended to jump into it!

I will try by using 4 RTX 3090Ti one for each quarter of the screen monitor!

 

Especially if the artist is coming from Cry Engine.

 

I wonder if things would be in a better state if Unity hadn't switched to the subscription model and instead had focused on major releases to introduce new tech and deprecate old features, while still providing bugfix support for the last major version(s) so that people can finish their multi-year projects on the old version.

The whole "with a subscription they don't need to hold new features back for major releases" doesn't sound like such a compelling argument to me anymore when those features are still so rough when they are being released, and might be abandoned again before they are stable.

 

Possible, but I don't think so. Plenty of major software titles are sub based and haven't had problems with the transition. Many of the things that folks bring up here don't seem to be related to those issues. The problems seems bigger than one thing. It largely feels like a lack of clear vision, or (and) the ability to implement that vision in a complex eco-system.

In some fairness to Unity, they certainly helped make the market what it is today. Before Unity, game development was a much smaller space, and "indie" game developers were a very different beast. (and hobbyists were very rare). Besides that, the market/space has significantly changed in time that Unity has been around. Whole markets have come and gone, gaming tools have moved into the tv/film and real time space. And just in general the nature of what a gamer is and how they consume is always changing. It a constantly moving target... historically very few companies have been able to navigate this kind of change over long periods of time, let alone short ones.