In my opinion there is no right or wrong Engine. My biggest deciding factor to go UE4 was the visual scripting system. It looks pretty, it causes sleepless nights, it sometimes gives me medical conditions that i cannot disclose in public... what more could you possible want?
Yeah it's all personal preference, what someone's good at may be another's biggest nightmare, so you gotta factor all that in. If you know UE4 and you can use it proficiently then there's no reason to learn another engine, unless it's giving you troubles.
I still use gamemaker, because I know it better. But I'm learning unity because gamemaker's giving me troubles on the web deployment front and i'm not good enough a developer to be able to write my own dlls and whatever to solve those things. So I hopped onto unity.
Wouldn't even dream of bothering with another engine at the moment. Don't do 3D modeling enough to care about Unreal.
For 3D I think it might have been a matter of Unity being there first in terms of editor UI and cross platform support. The editor UI for Unity is really good for programmers and non-programmers alike and thus is/has been quite popular with a wider audience I guess. Unreal has more recently upped it's game to provide a better editing workflow (I've not used Unreal so this is just what I've gathered from light conversation and a video tutorial I just watched online). They both look pretty similar nowadays, but seems like Unity still has the slightly better UI (though that's probably my bias, I use Unity a lot).
So the point is I think many studios have invested time and effort into Unity because it was the better choice in the past, so they built up skills and tangible frameworks to speed up their development in that engine. Now that all that time is invested, switching to a new platform is not quite a personal preference question anymore. It's a "stick with what we know well and can develop fast in" or "switch over to a new editor and relearn how to do things that way". Both engines seem on par for what we are doing ... also we have limited time and budget fuck it there's no good reason to switch let's dev.
If you were starting fresh (again, if we are talking 3D here) then I guess it comes down to preference, but also perhaps what style you want to implement easiest (I'd still say go with Unity but that's an unfair assessment given I've not used Unreal before :p). Either Unreal and Unity can emulate any number of aesthetic styles but for instance Unreal seems the simpler to create the lovely 'realistic' looking game world that might require more effort to create in Unity (though, as you say the differences are probably slight nowdays).
If we are talking 2D then there are tons of options. Not even going to try compare, there are so many I've not kept up with them all
DISCLAIMER: not saying companies will stubbornly stay with what engine they are used to, most are intelligent enough to explore new options when new problems arise. Also most of my research is "that chat I had to that one person that one time (while drinking)" so correct me if I'm wrong about anything. Love you all xoxo :p
So I've used Unity FAR more than I've used anything else, and Unreal generally has much better tools for empowering artists, with an accessible node-based shader editor, kismet, a game engine that expects you to be working using modular structures and grids, BSP geometry, and other cool stuff. There's more of a feeling of trying to make things easier there, whereas in Unity it's more a case of the feature merely "existing". And I feel that this comes from Epic actually using their game engine to make their own games, and having their artists and designers bitch at them about the tools' shortcomings and fixing it with a production environment in mind, and they themselves having to ship games using their own game engine.
Unity seems to rely massively on 3rd-party support, error reports, and voting for features, rather than actually making games themselves. And they've made it work, with the plethora of plugins and tools available on the Asset Store to fix Unity's shortcomings, or making the editor pretty extensible to allow for you to add these things yourself. I dunno, maybe I'm just more used to Unity's shortcomings having had to ship games with it more often. It's entirely possible (nay, probable) that UE/UDK has its own share of gotchas that I'm simply unaware of.
Personally, 3D art-wise I'd likely have a better time working with UDK/UE4, but noob-programmer-wise, I'd rather deal with Unity's C# than fiddle with UnrealScript or C++. Unity being so cross-platform and mobile-friendly is a big win though.
I think what @raxter says is true, but only now that there is a good enough engine that allows it. When I started at Luma, we migrated frequently, because each engine was so bad* (Blender, Torque, XNA). When Unity came (I was gone from Luma then, but we had the effect where I was working at the time), it was good enough to stick to it, and for a long time by far the best engine in that price range. And indeed, many people moved to Unity, and started to develop frameworks etc. What is also now true is that you can get developers with deep experience in Unity much easier locally than for other engines.
(I still love** Unity, but I also suffered from some undisclosable medical conditions when I checked out UE4 for the first time this weekend. I find it strange/amusing that @Elyaradine says it empowers artists, because I thought it really empowers programmers. Although C++ is of course ... not my favorite language, and also, messing about for a few hours is not the same as shipping a title.)
*(XNA was not really bad, nor is it an engine). ** (coloured by the usual dashes of anger and frustration that I'm sure all tools bring about).
Unity really grabbed me this year. In a big way. It was like a revelation...
Why? the workflow, the hierarchy of objects. the predictability of using its patterns, the simple way of dropping in scripts. The inspector thing that you add things to (shuriken!) and it just does those difficult thingies for you.
As a C# developer who loved XNA and visual studio and using solutions and best practices, classes and interfaces and iOC / dependency injection and all those fancy things that go with coding in general. Darn, it was a lot of work to just get basic things in XNA games working. An awesome API none-the-less, but still, lots and lots of code. To do boiler-plate stuff. grrrr
Unity just went, "hey dude, check this". Code a simple little script. Drop it in this object here and it does cool things - quickly. Oh and switch targeted platform at the press of a button - boom Android, boom Web, boom Windows PC...
Nah, I am blown away by it. In corporate when a client asks for a proposal for a mobile app that interfaces with a web service and database, i am almost shy when they ask me, so what is this Unity platform thing? And I explain to them it is a game development platform firstly, but it does cross-platform corporate mobile apps just really really well too. Cannot wait for the new GUI...
I am not surprised then that Unity is such a popular platform. I am kicking myself for NOT switching at least 3 years ago...
I'm not liking the fact that I am tied to Unity at the moment. I need it to finish current projects, and to maintain them in future.
I ideally want to switch to something that lets me have ownership of my code. This would likely mean using a bunch of open-source libraries that don't have the same ease of use/community support as Unity/UDK etc. but my games in general are not particularly complex, so that doesn't bother me too much.
I did part of a project before using Irrlicht and that wasn't too difficult to jump into.
Unity seems to rely massively on 3rd-party support, error reports, and voting for features, rather than actually making games themselves. And they've made it work, with the plethora of plugins and tools available on the Asset Store to fix Unity's shortcomings, or making the editor pretty extensible to allow for you to add these things yourself. I dunno, maybe I'm just more used to Unity's shortcomings having had to ship games with it more often. It's entirely possible (nay, probable) that UE/UDK has its own share of gotchas that I'm simply unaware of.
Personally, 3D art-wise I'd likely have a better time working with UDK/UE4, but noob-programmer-wise, I'd rather deal with Unity's C# than fiddle with UnrealScript or C++. Unity being so cross-platform and mobile-friendly is a big win though.
This was my biggest deciding factor. UE4 was my choice due to the Epic staff responding the same day regarding issues i experience. Also blueprints enables me to do stuff without having to code or script.
My biggest concern was 3rd party requirements / support.
I'm not liking the fact that I am tied to Unity at the moment. I need it to finish current projects, and to maintain them in future.
I'm generally a massive fan of Unity, but this is my biggest worry using it. Unity's component based system encourages some pretty bad practices that leave you very tied into the tech if you're not careful. It doesn't really encourage layering of your code so your rendering is independent of your game logic.
This isn't a massive problem for a game where you're looking at development plus say a year of maintenance. But for applications where you expect to have a longer maintenance and development cycle, not great. And really, do you want to be in a situation where a couple years down the line you want to re-release your game to some new device, but it's in an ancient version of Unity and the port just wouldn't be practical because everything's too tangled together. I'm looking at ways to structure code more sensibly for that now, but if anyone has any good tips in that regard, I'm all ears :)
A more immediate side effect of components is that debugging can be more of a pain than it needs to be, because you don't have an explicit game loop that runs through all updates in an easily predictable and follow-able way.
konman: As an XNA dev that shifted over to 3D JavaScript (three.js) for Web Dev stuff (People are FAR more willing to visit a Web-Page than download an exe to check out something), I may have to take a second look at Unity. I last took a look at it an age (5+ years) ago, and at the time it had its fair share of problems :)
Chippit: As do I :< You can still hack Visual Studio to work with it though, and people have done just that for several Steam games. I looked into Mono, and it still has SEVERE bugs (Don't even think about converting a 1k+ LoC program with any ease).
Time to look back into Unity Dev with my next bout of free time :)
@mattbenic Would it be fair to say that Unity's component system offers some advantages in a jamming environment? (it encourages re-usability of code certainly)
(Just asking what your perspective is on this. I love the Unity component system, so I'm trying to figure out why your experience of it is so different to mine)
Comments
I know its a bit off topic but who are the big studios in SA?
By that i mean studios who employ local talent.
I was curious as to why dev's would go Unity.
I did find a great non biased article on which engine to chose which comes down to one thing. Its all personal preference and the goals you wish to achieve.
http://blog.digitaltutors.com/unity-udk-cryengine-game-engine-choose/
In my opinion there is no right or wrong Engine. My biggest deciding factor to go UE4 was the visual scripting system.
It looks pretty, it causes sleepless nights, it sometimes gives me medical conditions that i cannot disclose in public... what more could you possible want?
I still use gamemaker, because I know it better. But I'm learning unity because gamemaker's giving me troubles on the web deployment front and i'm not good enough a developer to be able to write my own dlls and whatever to solve those things. So I hopped onto unity.
Wouldn't even dream of bothering with another engine at the moment. Don't do 3D modeling enough to care about Unreal.
So the point is I think many studios have invested time and effort into Unity because it was the better choice in the past, so they built up skills and tangible frameworks to speed up their development in that engine. Now that all that time is invested, switching to a new platform is not quite a personal preference question anymore. It's a "stick with what we know well and can develop fast in" or "switch over to a new editor and relearn how to do things that way". Both engines seem on par for what we are doing ... also we have limited time and budget fuck it there's no good reason to switch let's dev.
If you were starting fresh (again, if we are talking 3D here) then I guess it comes down to preference, but also perhaps what style you want to implement easiest (I'd still say go with Unity but that's an unfair assessment given I've not used Unreal before :p). Either Unreal and Unity can emulate any number of aesthetic styles but for instance Unreal seems the simpler to create the lovely 'realistic' looking game world that might require more effort to create in Unity (though, as you say the differences are probably slight nowdays).
If we are talking 2D then there are tons of options. Not even going to try compare, there are so many I've not kept up with them all
DISCLAIMER: not saying companies will stubbornly stay with what engine they are used to, most are intelligent enough to explore new options when new problems arise. Also most of my research is "that chat I had to that one person that one time (while drinking)" so correct me if I'm wrong about anything. Love you all xoxo :p
Unity seems to rely massively on 3rd-party support, error reports, and voting for features, rather than actually making games themselves. And they've made it work, with the plethora of plugins and tools available on the Asset Store to fix Unity's shortcomings, or making the editor pretty extensible to allow for you to add these things yourself. I dunno, maybe I'm just more used to Unity's shortcomings having had to ship games with it more often. It's entirely possible (nay, probable) that UE/UDK has its own share of gotchas that I'm simply unaware of.
Personally, 3D art-wise I'd likely have a better time working with UDK/UE4, but noob-programmer-wise, I'd rather deal with Unity's C# than fiddle with UnrealScript or C++. Unity being so cross-platform and mobile-friendly is a big win though.
(I still love** Unity, but I also suffered from some undisclosable medical conditions when I checked out UE4 for the first time this weekend. I find it strange/amusing that @Elyaradine says it empowers artists, because I thought it really empowers programmers. Although C++ is of course ... not my favorite language, and also, messing about for a few hours is not the same as shipping a title.)
*(XNA was not really bad, nor is it an engine).
** (coloured by the usual dashes of anger and frustration that I'm sure all tools bring about).
Why? the workflow, the hierarchy of objects. the predictability of using its patterns, the simple way of dropping in scripts.
The inspector thing that you add things to (shuriken!) and it just does those difficult thingies for you.
As a C# developer who loved XNA and visual studio and using solutions and best practices, classes and interfaces and iOC / dependency injection and all those fancy things that go with coding in general. Darn, it was a lot of work to just get basic things in XNA games working. An awesome API none-the-less, but still, lots and lots of code. To do boiler-plate stuff. grrrr
Unity just went, "hey dude, check this". Code a simple little script. Drop it in this object here and it does cool things - quickly.
Oh and switch targeted platform at the press of a button - boom Android, boom Web, boom Windows PC...
Nah, I am blown away by it. In corporate when a client asks for a proposal for a mobile app that interfaces with a web service and database, i am almost shy when they ask me, so what is this Unity platform thing? And I explain to them it is a game development platform firstly, but it does cross-platform corporate mobile apps just really really well too. Cannot wait for the new GUI...
I am not surprised then that Unity is such a popular platform. I am kicking myself for NOT switching at least 3 years ago...
I ideally want to switch to something that lets me have ownership of my code. This would likely mean using a bunch of open-source libraries that don't have the same ease of use/community support as Unity/UDK etc. but my games in general are not particularly complex, so that doesn't bother me too much.
I did part of a project before using Irrlicht and that wasn't too difficult to jump into.
My biggest concern was 3rd party requirements / support.
This isn't a massive problem for a game where you're looking at development plus say a year of maintenance. But for applications where you expect to have a longer maintenance and development cycle, not great. And really, do you want to be in a situation where a couple years down the line you want to re-release your game to some new device, but it's in an ancient version of Unity and the port just wouldn't be practical because everything's too tangled together. I'm looking at ways to structure code more sensibly for that now, but if anyone has any good tips in that regard, I'm all ears :)
A more immediate side effect of components is that debugging can be more of a pain than it needs to be, because you don't have an explicit game loop that runs through all updates in an easily predictable and follow-able way.
Chippit: As do I :< You can still hack Visual Studio to work with it though, and people have done just that for several Steam games. I looked into Mono, and it still has SEVERE bugs (Don't even think about converting a 1k+ LoC program with any ease).
Time to look back into Unity Dev with my next bout of free time :)
(Just asking what your perspective is on this. I love the Unity component system, so I'm trying to figure out why your experience of it is so different to mine)