Search Unity

  1. Unity 2019.1 is now released.
    Dismiss Notice

A warning to new indie developers in general

Discussion in 'General Discussion' started by arkon, Apr 21, 2019.

  1. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Ok, I've been using Unity since V3 and have created and released many games now so feel I should now take the time to warn new developers of one of the major problems I've encountered over and over again using Unity as a tool. Whilst it has allowed rapid game development it has come at a huge cost that is not immediately noticeable until years later.

    Basically it's a combination of the Asset store assets and Unity's constant depreciating of features.
    A case in point is a Game I wrote back in U3, later converted to U4 not sure when now, maybe around 2014.
    It has just been delisted first by Apple for not updating it in a while and then by Google for some ridiculous reason I won't go into. Anyway the game was still actively played and makes me approx $300 per month. So I felt the need to update it and re release it now in 2019. I'm currently using Unity2017.4.25 LTS version so still way behind the current unity release so everything I update will already be out of date.

    I'm currently staring at a console with 254 warnings and 30 errors. as I fix the errors it seems to unlock Unity to import more stuff in the project which just makes the warning list grow and grow. Most are caused by either Unity depreciations or Asset store assets I used that need updating, and here's the rub many of them are abandoned by their authors so there are now 2017 updates! So while the asset store seems at first a good idea, if you use it you will be in my position in a few years time with a ton of make work that adds nothing to the game and just uses huge amounts of time to fix or work around.

    Just for fun here is the current state of play on this game:

    Realistic FPS Prefab The entire game is based around this and it's depreciated causing dozens of warnings
    Detonator framework, not updated since U4
    T4M, Gone leaving me with a what the hell do I do now type moment.
    On top of this is all the image effect errors, which if I delete and import the new ones cause a buck ton of errors and effects not working anymore.
    Prime 31 plugins, ALL need re paying for to get the latest versions just to keep the game alive on stores.

    Anyway so depressing, too 3 months to write the game, and I've spent at least that amount of time again just maintaining it and here I am 4 years later still doing it. Just needed to let off steam guys sorry.
     
    recon0303, JBR-games, Stardog and 9 others like this.
  2. dadude123

    dadude123

    Joined:
    Feb 26, 2014
    Posts:
    783
    I'm interested, could you estimate what fraction of the asset-store-assets you are using in that game fall into the code or art category?
    Like how many things you bought are full "systems" or stuff that contains non-trivial code?
    And how many things are "pure art" like lets say just 3D characters with textures and animations?

    It sounds like in your case more than 50% of the things you have trouble with is non-art stuff, is that right?
    Actually, how many "purely-art" assets did you buy from the asset store for your game, if any? And how many of those have given you any trouble while upgrading?

    Unfortunately I don't have much advice to offer here, but I can understand your situation and I can see how frustrating it must be :(

    Ah one last thing though: If one were to compare engines like unity, unreal, godot, whatever..., then I can assure you that when it comes to deprecation and generally "engine devs trying not to break stuff when possible" Unity would definitely come out on top of that list.
    Other engines are much more aggressive when it comes to removing stuff, forcing updates to APIs, ...
    Just throwing that out there.
     
    neoshaman likes this.
  3. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Mostly code assets but there are some art assets. I seem to have less trouble over the years with art assets.

    I understand why they depreciate things, and why asset store developers eventually lose the will to live and stop supporting their assets. A strong reason to write your own always no matter how much longer it adds to the project.

    I think Unity should never depreciate anything unless they are prepared to maintain previous releases indefinitely.
    We are constantly forced to update the engine and it would be good if half our code didn't need fixing each time.

    Oh I should mention I have more than 20 games released and this problem is magnified by that number. It's like a constant juggling on each game to keep it alive on the stores over the years and it's been ages since I've written a new one due to the time needed to maintain them all, all the time!

    Currently I'm in the middle of ditching Prime31 assets from all my games and replacing them with the Unity built in options. All my GUI stuff is replaced with NGUI (Hope he never stops supporting that one!), Detonator Framework is all being replaced with ParticleSystem stuff. Swapping out all the Pool code purchased over the years and creating my own. Who knows one day I might free enough time to actually create something new!
     
  4. Ryiah

    Ryiah

    Joined:
    Oct 11, 2012
    Posts:
    13,872
    Everything you've described only further backs up my decision to avoid mobile game development. I can't remember a single instance where I had to update my game development tools (Unity or otherwise) to continue making games for standalone. Being able to stick with old releases means my assets don't break either.
     
    Last edited: Apr 21, 2019
  5. hippocoder

    hippocoder

    Digital Ape Moderator

    Joined:
    Apr 11, 2010
    Posts:
    25,183
    My advice is don't chain the fate of your projects to third party on asset store. If you will use it then make sure it's something that you can potentially maintain...
     
    deliquescator, Rich_A, Lu4e and 10 others like this.
  6. Antypodish

    Antypodish

    Joined:
    Apr 29, 2014
    Posts:
    4,575
    No you are not forced. Unless by yourself.

    We would have still boo and Unity JavaScript. Which is pointless now, and would slow down Unity progress.
     
  7. Ryiah

    Ryiah

    Joined:
    Oct 11, 2012
    Posts:
    13,872
  8. Antypodish

    Antypodish

    Joined:
    Apr 29, 2014
    Posts:
    4,575
    But how old app would have to be, to be only 32-bit?
     
  9. Ryiah

    Ryiah

    Joined:
    Oct 11, 2012
    Posts:
    13,872
    mtornio, JBR-games, Aldo-V and 4 others like this.
  10. Antypodish

    Antypodish

    Joined:
    Apr 29, 2014
    Posts:
    4,575
    JBR-games likes this.
  11. angrypenguin

    angrypenguin

    Joined:
    Dec 29, 2011
    Posts:
    11,769
    For mobile platforms you can be. The 32/64 bit thing is a big one, but there's a bunch of smaller things that have happened along the way as well. I can't give recent examples because I haven't done mobile development for quite some time now, but back when I did there were multiple occasions where we upgraded things just because we had to for some reason or another.
     
    Kiwasi and Martin_H like this.
  12. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    Your problems seem to be with them more than with Unity, unless your game is no longer playable on new devices.

    And, well, games have an expiration date too... Maybe you should completely remake, and maybe improve it, instead of updating it? :)
     
    SparrowsNest likes this.
  13. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    I never just update a game, I always use the opportunity to add something for the player. It's just the time left to add new features etc, is getting smaller as more and more time is taken just to get the thing buildable again.

    As Ryiah rightly pointed out, You are forced by external influences to continually update your projects unless you just want to give up and pull a game from the stores. I see my games as works of art, I took time, love and care to create them, so want to see them available and playable for as long as humanly possible. The fact they are still downloaded and played 5 years or more after publishing them is pretty amazing I think.

    Google just forced me to do updates on all of them, not for the 64bit thing although that will come, but because every one of my games had the same in build mechanic for players to see my other games by pressing a "View more of my games" button. Google for some reason recently decided this was deceptive advertising! So ALL of them need this functionality taken out.

    My games all use the Unity Particles (Pre particleSystem) and most have GUI stuff in them, both are gone in U2018.
    Have you every tried to convert from one particle system to the newer one? It's doable but takes huge amounts of time and I rarely get them looking like they did before.

    Apple force updates, first was the 32-64bit thing, this meant some of my games had to go from U3 to U5 and the rest U4 to 5. Unity5 introduced a terrible performance hit on mobile games, both app size got bigger and frame rates plummeted. It took huge amounts of effort just to get the FPS near back to what it was on each one. Apple also periodically send me a "Your game is too old, you haven't updated it in a while, you have 30 days to upload a new one" type email.

    All this would be fine if the engine and the continual dropping of support on Assets wasn't a thing. I am an Asset store publisher too, I have assets on the store and I know first hand the drivers for giving up on them, and I don't blame the authors. My missive was meant as a warning that if you use assets from the store you are going to get burnt more than once. Like I say if NGUI ever stops updating it I will surely need to find a bridge to jump off.

    So my lovely indie developers, I wish you all the luck in the world, and Unity you really should give more thought to this as it may be fine to use it for the initial game development, but mobile games have very long legs and stay relevant for years after and you don't make it easy to support our old stuff.

    I am dreading moving to Unity2018 !! And 2019 is another future nightmare waiting.....
     
    Jingle-Fett, ihgyug and Ony like this.
  14. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Just for kicks, I've just gone through all the assets I've ever bought, the stats are: Purchased 171, Depreciated 45 or 26% of the assets I've used from the store are now depreciated. Amazingly some of them art assets! Not sure how to calculate how much wasted money that was as on a depreciate asset you can't click on it to go to its page and see how much it was.
     
    JBR-games and Ony like this.
  15. DoctorShinobi

    DoctorShinobi

    Joined:
    Oct 5, 2012
    Posts:
    71
    You can't expect Unity to not deprecate things and support them forever. They'd be wasting way too many resources just to maintain support for those things instead of developing new features. It comes down to a matter of balancing how much time they should support obsolete features before they become fully deprecated. At a minimum they usually wait a couple of major patch releases before fully deprecating completely.

    The best thing I can suggest to new(and old) developers using Unity is to avoid using plugins as much as possible. Especially if it's made by some random person and not a company that is more likely to support it for years. This is something I believe is relevant to other engines too.
     
    dadude123 likes this.
  16. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    4,402
    So time goes by and the world changes. Unity advances, mobile phones advance, part time asset developers move on etc. Nothing stays still.

    Game development, like any software development, is a question of building something to align with multiple changing variables. You aim for a moment in the future where you believe you can most clearly forecast the right circumstances - the right idea, the right need from the market you're aiming at, the right tools, the right time in your life to spend the necessary time on the task - and you go for it. But 4 years later, each of these variables has changed.

    The thing is, it's the same deal for asset developers. And for Unity themselves. Sometimes, like riding a wave, you make a product that succeeds for a long time, and if you are fast and fortunate, you can change exactly the right things at exactly the right time to keep it at the top. Often you are not so fortunate, and you either forced to abandon it, or you make a big change, hoping for the better. Both of these create costs, for you and the people using your product. But that's life in business - and nobody can benefit from you if you don't survive.

    My advice, frankly, would be to build a new game. $300 per month is significant, but it seems to me that 4 more years of experience would give you the ability to do even better.
     
    JBR-games and Socrates like this.
  17. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Throw the baby out with the bath water you mean. As much as I'd like to abandon things like this I personally think if players or in our case developers have paid for something then you should have the ability to always use it. If a player has spent money on my game I'd like to not pull the rug out from him. I know other devs do as many a game I've paid for in IOS is no longer available, and it kind of sucks, and I'd like not to be adding to that.
     
  18. Billy4184

    Billy4184

    Joined:
    Jul 7, 2014
    Posts:
    4,402
    If it's multiplayer, then sure, but a singleplayer game you've bought is on your device, no one can take it. If Google pull it down for not having updates, well that's where the issue is frankly. There's enough unwanted garbage on there that there's no need to be deprecating a $300/month game just because it's not been updated in a while.

    I guess the way I see it is that game development is a difficult and complicated business, and there are a range of things that make it a lot easier than it otherwise would be. But these products come at a great cost. And in a lot of cases, the creators of these products have not found a way to properly factor in the costs of making, maintaining and upgrading it while keeping the price 'acceptable'. And when this happens, these products always die, usually sooner or later. This is the case with 99% of the asset store.

    The thing is, arithmetic doesn't lie. If you make a baseline cost analysis of what it would cost for you to learn all the necessary skills to not depend on any tools at all, that cost is extremely high. Years of experience and dedication, and hundreds of thousands of dollars in man-hours. That's the reference point.

    But when you can download a few assets from the store for a few hundred dollars total, it makes it look like everything should be low-cost, when it's actually not. Capitalism is a beautiful thing, but it isn't magical - there's a great saving to be made from the hard work and experience of someone else offering a product on the market, but the price cannot be reduced too low without increasing the risk of abandonware and reduction of quality.

    The thing is, it takes a lot of work to create and maintain a great product on the asset store. I think that with the asset store, like newbie freelancers, people often create things without knowing really how good they are at what they do, and so something good appears at an attractively low price. Or they are aware of their capabilities, but they think the asset store market is of a different nature than it is. Sooner or later, the market with balance things out - people realize what they are worth, or they realize their business model is a failed one (or both) and they quickly move on.

    In short, if you would like to create stability in your business, I think it would be best to learn things yourself or hire skilled people. There are a few assets on the store that are long-lasting and successful (because they have a good business model) but for the most part, it is a flea-market of products of all different prices and quality levels, with the only thing generally in common (according to my perspective) being a lack of clear business model at all.
     
  19. AndersMalmgren

    AndersMalmgren

    Joined:
    Aug 31, 2014
    Posts:
    4,293
    I'm worried for that built in support will drop. Other than that we dont have much third party code assets. Obi rop being the biggest one.

    I get were you are coming from, but what you did wrong was to wait to update. You should instead do constant updates. Always have two branches other than your production branch. One with latest LTS and one with latest and greatest. Time of writing that is 2017.4.25 and 2019.1.0f2.

    Software development is a agile business, with constant small pushes forward.
     
    Amon, xVergilx, pcg and 1 other person like this.
  20. Murgilod

    Murgilod

    Joined:
    Nov 12, 2013
    Posts:
    5,411
    This problem is not really exclusive to Unity.
     
  21. AndersMalmgren

    AndersMalmgren

    Joined:
    Aug 31, 2014
    Posts:
    4,293
    Btw, never listen to those that say you shouldn't update the engine during development. They do not know what they talk about.
     
    xVergilx likes this.
  22. Ippokratis

    Ippokratis

    Joined:
    Oct 13, 2008
    Posts:
    1,504
    Hi @arkon,

    Some ideas that might help you in your current task:
    - Consider getting some help with the ui - particles conversions, they can take some time.
    - Do not add new features to the existing game now, hold until a later update.
    - Re-buy the asset store packages that got updated, the cost to do it by yourself/hire someone to create them from scratch will probably be higher.
    - In case of abandoned unity assets, consider hiring the author of the asset, it will be easier for him and less expensive - time-consuming for you than hiring someone else.

    Some other thoughts:
    - Targeting the latest LTS seems a good idea. You have already made some games and some assets. It is tough to stay up-to-date with everything but being up-to-date with the latest LTS might worth the trouble.
    - It is beneficial to upgrade some projects and kill others. Like it or not, you get to choose what to support. Sometimes it is necessary to press the kill button if you wish to survive, it sucks, but living in a carton box sucks more.

    I am sure that you have already thought most of this stuff, anyway.
    Kind regards.
     
    betovena likes this.
  23. Lars-Steenhoff

    Lars-Steenhoff

    Joined:
    Aug 7, 2007
    Posts:
    2,028
    It's good to be aware of,
    I have been though the same path, made a game in unity 3 updated to 4 and it was still 32 bit, then later apple ditched 32 bit and it had to be updated.

    I have still not had the time to update to unity 2018 for that project because so much had changed, and so many plugins were deprecated. Just a few examples.

    Old particle system,
    A plugin for timeline before timeline even existed in unity. ( now deprecated )
    All the global elimination that was done in beast. ( takes a long time to redo in enlighten )
    Old terrain shaders ( T4M )
    Some editor tools I was using also got depreciated with made things more labour intensive.

    And with all that the new versions are not any faster than unity 4 was for the type of project I have.
    But because I need to use the latest Xcode I also needed to upgrade my OS.
    And in the latest MacOS Nvidia 1080ti card is not supported anymore so no more light baking with GPU.

    Anyway long story short, It's difficult to maintain a game for a long period.
     
    Last edited: Apr 21, 2019
    JBR-games and arkon like this.
  24. hippocoder

    hippocoder

    Digital Ape Moderator

    Joined:
    Apr 11, 2010
    Posts:
    25,183
    Edited the title because it's pretty clear it's external forces, and marketplaces != engine itself, but people. Not a warning for indies using unity, more "things I learned about game dev".
     
  25. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    I kniew the title had changed... :p
     
    hippocoder likes this.
  26. hippocoder

    hippocoder

    Digital Ape Moderator

    Joined:
    Apr 11, 2010
    Posts:
    25,183
    :) just a courtesy for the original poster as he's a long term member of the community so this letting off steam is fine. I edited title so that it's less inflammatory and wanted to explain.
     
  27. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Erm, not sure that's totally true. The main and probably the only reason the assets on the store become un-usable and unsupported is exactly because the engine (Unity) changes so much that the assets are no longer compatible with it.

    This one I'm updating at the moment uses RTFPS asset, now un supported and throws so many warnings it makes it hard to see the warnings that are more significant. I've had to #pragma it to no show the depreciated warnings. The asset makes huge use of the old GUI system, it's in nearly every file, if GUI gets removed after Uinty2017 then this asset will totally break and the fix is just not worth it, way too much work for no result.
     
  28. Murgilod

    Murgilod

    Joined:
    Nov 12, 2013
    Posts:
    5,411
    Things get deprecated in other engines too.
     
    RecursiveFrog likes this.
  29. AndersMalmgren

    AndersMalmgren

    Joined:
    Aug 31, 2014
    Posts:
    4,293
    Again software is a agile business. Constant updates is super crucial. With assets you sadly get dependencies that third party is doing this work.

    I have seen that this agile approuch is also alot less standard than in corp biz. Probably because of the high amount of hobbyists
     
    xVergilx and Billy4184 like this.
  30. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    Your problem is not with Unity, it's that companies no longer want to sell it on their platforms. As I already said, games have an expiration game; there is nothing wrong with that.

    Is the original game, the one that Apple and Google rejected, still working when you play it?
     
  31. AndersMalmgren

    AndersMalmgren

    Joined:
    Aug 31, 2014
    Posts:
    4,293
    Unmaintained games have an expiration date, look at games like original Doom, it's still being managed and published on numerous platforms.
     
    Ony and angrypenguin like this.
  32. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Not if the engine was structured better, there would be no need. I was the architect of a game development engine in the Gambling industry for 20+ years, The last engine I wrote lasted more than 10 years with never needing to make anything depreciated. Well not strictly true, I would depreciate things but never remove something that any game had used or needed in the past. The code was structured in a way that if something wasn't used then it wasn't included in the build. A true library if you will. It meant a 10 year old game could always be built on the latest version of the libs and would still work exactly the same as it did 10 years prior.

    Take OnGui for example, if they really want to remove it, then they should spend some time making it so that if the game uses it then it is included and still available. Not just rip it out and say tough, re write all your game code!

    Same with the old particle system. I'm not asking Unity to maintain it and add features, just keep it in, make it a true library component, not just rip it out.
     
    deliquescator, Meltdown and Ony like this.
  33. Lars-Steenhoff

    Lars-Steenhoff

    Joined:
    Aug 7, 2007
    Posts:
    2,028
    One problem is that when unity depricates some API it does not always do an auto conversion to the new API, The be fair in many cases it can auto update the scripts with new API if its only a name change, But bigger changes are not always auto updated. It could save a lot of time on he user side if things can be auto updated to newer versions.

    For example the new postprocessing v3 is not backwards compatible with v2.
    It would be nice if its auto updated to v3
     
    Last edited: Apr 21, 2019
  34. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Only on older IOS and older devices. Apple change Xcode & MacOS every year also IOS, on mobile you have to constantly re release updates for years if you want to get the maximum return for your effort. The engine writes know this and should militate against it.
     
  35. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    Well the design paradigm is wrong then. In this example Unity shouldn't even bother making V3 compatible with V2.
    V2 should be internally abandoned but always left in and functional, V3 should just be an extra new library, users can use either version.

    I know I come from an older software industry but if I had written a "Bulb On" library function then a few years later created a "Colour Bulb On" function, I'd never delete the old version, it would be truely librified, the end user could use either, OR and both if they wanted to.
     
    Ony likes this.
  36. BonneCW

    BonneCW

    Joined:
    Jan 22, 2017
    Posts:
    31
    But that doesn't make much sense. There are always changes, either bugfixes or just some refactoring, that might affect some of the old code. So if they wouldn't remove it there would be a lot to change and it would become more and more. And besides increasing maintenance effort also code size would increase faster when they'll never get rid off old stuff.

    It makes absolutely sense to remove old code and even APIs as the code evolves. And especially in gamedev a lot stuff changes all the time. It's not like eg the Qt framework where they can change some internal stuff or add new APIs to be compatible with most applications. And if they deprecated old APIs and remove them somewhen.
     
  37. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    So things should never change, evolve?
     
  38. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    The developers use their own assets I guess? They didn't buy them on some store making them unable to adapt them to newest hardware.
     
  39. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    If you re read what I said, I'm all for change and new stuff, but there is absolutely no need to break or remove old stuff. If the engine is structured correctly there would be no need to remove things. They just don't get included if the end user is not using that functionality. It would help the end executable size too if this was all built to be more modular and library like.

    How much of my tiny games are full of code that is never used? It must be a buck ton of it judging by the huge size of the executables. The unity app on my machine is 8.2GB that is just mental. it suggests a lack of discipline and structure in the design. A free for all by the coders, no central direction. I've stayed mostly silent on this up till now but I can't be the only old school engineer that understands that more is not always better. Unity should stop in it's tracks and take a hard look at the engineering going on and ask how can we make this engine / tool more reliable, faster, smaller and more modular. Not just chasing the "How many new features we have added this year model"
     
    deliquescator and Ony like this.
  40. APSchmidt

    APSchmidt

    Joined:
    Aug 8, 2016
    Posts:
    1,959
    What gets out of this is simple: do not use assets you didn't make yourself. No deprecation in the future, you'd be able to modify them each time necessary.
     
  41. xVergilx

    xVergilx

    Joined:
    Dec 22, 2014
    Posts:
    1,645
    Two cents:

    - Do not rely on third party assets, if you do not want to support it in the future on your own.
    Or do not have access to the source;

    - Do not listen to those who say - never upgrade to newer version.
    Its easier to fix one thing that broke, than fixing your whole project later because core feature just gone forever;
    Or its impossible to build for newer mobile API etc.

    Newer versions usually come with benefits. And those benefits come with better performance and usability.
    E.g. OnGUI is freaking nightmare that should not be even implemented, not to mention the cost of the UI drawing that way.


    Sure there's a cost of upgrading projects. But its way cheaper to do things along as Unity's development goes on.

    Each stable iteration of Unity is a potential upgrade, so yes. It should be done and done in time.
    (Unless you're not willing to support projects at all in the future)
     
    Last edited: Apr 21, 2019
  42. arkon

    arkon

    Joined:
    Jun 27, 2011
    Posts:
    1,088
    This is what I’m doing now, for new games that fine, but I don’t want to rewrite the old ones.
     
  43. neoshaman

    neoshaman

    Joined:
    Feb 11, 2011
    Posts:
    4,095
    That's exactly what they announces they have been doing for a while, hence ecs DOTS, package manager, unity hub, etc ...
     
  44. BIGTIMEMASTER

    BIGTIMEMASTER

    Joined:
    Jun 1, 2017
    Posts:
    2,084
    Opportunity for "remastered" titles. Easy money.
     
    ikazrima, Ony, Antypodish and 2 others like this.
  45. Ryiah

    Ryiah

    Joined:
    Oct 11, 2012
    Posts:
    13,872
    How would you expect them to be able to provide access to all previous systems if they didn't maintain them? Keep in mind it's not just Unity themselves making alterations to their game engine that can result in bugs. Both hardware and software platforms are constantly evolving and removing old features.

    If one of those hardware or software features was necessary for a system to function then Unity would have no choice but to go back and maintain it if they went down this route. Android abandoning 32-bit support is a great example of this. Unity had to backport support to the current LTS.
     
    angrypenguin likes this.
  46. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    7,630
    As said by others, things change and don't rely on third party assets.

    Platform requirements change, so the engines have to change. Maintaining legacy for old engine support is unreasonable to ask, as it would require massive updates to make them useable on current platforms. It's just not the responsibility of the engine. Say old engines last for 10 years or whatever is irrelevant, things change and evolve at a increasing rate, especially on platforms that are hardware cycled so fast. No one has build a engine or tool or software that that has lasted unchanged for 10 years on mobile. Its not a real expectation.

    As normal solution for maintaining old games is incremental updates. This isn't specific to Unity, its every tool, engine and codebase. It's just (game) development. It's a hard lesson to learn this late in the game, but shouldn't be a surprise to anyone who has developed games. It's like warning people that if they go out into the rain, they might get wet.
     
    angrypenguin, Billy4184 and Ryiah like this.
  47. Eric5h5

    Eric5h5

    Volunteer Moderator Moderator

    Joined:
    Jul 19, 2006
    Posts:
    32,084
    That's not really true though. Also you'd hate Unreal, since there's zero compatibility between versions; Unreal 3 is literally a different engine compared to Unreal 4. No doubt Unreal 5 will be a different engine as well.

    Anyway I'm not so sure about the third-party thing, as long as you have the source code. I mean, you're going to have to fix deprecated stuff anyway, whether you wrote it or not. And if it's something you wrote 5-10 years ago and haven't regularly updated, it will be indistinguishable from third-party code by that point.

    --Eric
     
  48. LaneFox

    LaneFox

    Joined:
    Jun 29, 2011
    Posts:
    6,432
    Warning: Keeping up with modern technology is hard.

    ftfy
     
    dogzerx2 and hippocoder like this.
  49. zombiegorilla

    zombiegorilla

    Moderator

    Joined:
    May 8, 2012
    Posts:
    7,630
    It’s only hard if you put it off too long. If you stay on top of things, it’s only a minor inconvenience. *excepting for third party... that is just a straight up gamble.
     
    LaneFox, hippocoder and dogzerx2 like this.
  50. RecursiveFrog

    RecursiveFrog

    Joined:
    Mar 7, 2011
    Posts:
    127
    I think it's a reasonable complaint. I hope that Unity's recent move to make absolutely every part of the engine distributed as a package in the Unity Package Manager could go a long way to ease the pain. That said, have you seen what's happening in the mobile phone space? Game developers with legacy apps are very low on the list of people that the platform owners care about, particularly if your game isn't driving sales of this year's new hardware model.

    Mobile phones may very well be a poor choice for all but the most well funded game developers in the near future, even with tools like Unity. If the space settles down a bit, it may be safer to jump back in, but these are rough waters right now.