Unreal is really winning a lot of points in my book. First with the engine pricing and now this. If I ever step up my mobile game making to 3D I'd definitely use them over Unity.
Nevermind that their offering is superior in terms of long-term product maintenance and potential platform offerings since you get complete source access.
Unity's advantage remains in the asset area with its pipeline and arguably easier to approach editor, but source code access is a huge win for any developer that has the skills needed to maintain a product long-term.
Having worked in games for 13 years, at studios large and small and projects of the same variety, source code access has almost never blipped on the radar. The few times when we had issues with various engines for which we didn't have source, our support contracts gave us the changes we needed to ship on time. On occasion we modified the engine, only to be hit by tricky merges with subsequent releases. In general, I've found that modifying an engine should be done by engine programmers (and if a game company does enough engine development to have such specialists, it's probably not licensing any of these options).
The latest Unreal is a wonderful piece of engine software. Their blueprint system as well as their material editor and best-in-class renderer really are something to talk about. They also have a gigantic learning community. And I think their licensing terms are a good approach.
However, Unity has blazing fast compilation time on the order of seconds, a properly built play-in-editor mode, large asset and plugin ecosystem, seamless asset pipeline, and support for a modern programming language in C#. Each of these could arguably be considered a game changer in isolation, but in aggregate they are an efficiency avalanche. Nothing makes a better game, faster, than being able to go from idea to prototype in five minutes rather than two hours. It's possible to try more things, to discard ten or even fifty bad ideas for every good one, and still come out ahead. This is what it all comes down to, in my experience. And when you're done, you can port your game to over a dozen platforms (in some edge cases by simply changing a dropdown value).
That said, I'm happy that both engines are so good, because it means neither will rest on its laurels. Unreal's marketplace and Unity 5's renderer are no accidents.
I understand what you're saying, but as someone that has followed various game developers closely, especially those that do porting work, source access is a huge win in many cases.
Imagine your game shipped on a version of Unity that's no longer supported but you want to continue using it for projects or bring back an old game to a new OS platform. Without the source, your only option is to port the game to a new version of Unity. That's not so great.
But again, I'll readily admit that all of this is only a win if you have the right team with the right skillset. There are definitely tradeoffs involved.
I think at scale, that's true. But what if you're a small team, or a solo developer? Let's face it: there's no way you'll be able to port Unity or Unreal to a new platform by yourself, so that might as well not even factor into it.
I think Ryan Gordon (aka 'icculus') is proof that the right solo person can do such a port :-) Nevermind Casey Muratori or others I could name...
There are many small, talented indie developers that definitely have the right set of skills to bring an engine to a new platform, or to one that's just slightly different.
For example, porting an engine from Linux to FreeBSD or some other UNIX should be trivial, but impossible without source code access.
Now I doubt many games are going to be ported to those platforms, but imagine a 3D walkthrough program for an architectural client or manufacturing facilities where UNIX is still found.
My experience is entirely different, 3rd party middle-ware without source is almost always a pain to work with (even a pain to evaluate, sometimes you get a license-key which locks the evaluation copy to one machine, for a limited time, etc). When bugs show up you cannot simply debug-step into the middleware libraries, instead you have to do a lot of guess-work trying to reproduce the bug, and then follows endless ticket back-and-forths with support. Having the source code is also always better when integrating the middleware into your own build process, since you need binary libs that match exactly your compiler version and build settings (exceptions disabled? link-time-code-generation? etc...) I could rant on forever of how much not having source code access sucks ;)
I disagree with the importance of source code. I've saved weeks of time diagnosing issues when given the 3rd party source code compared to attempting to work with the 3rd party vendor without source access.
Also, Unreal 4 compiles in less than 30 seconds on my laptop and has a play-in-editor mode, seamless assets, blah. Also support for a modern programming language in C++11.
I've worked with Unreal Engine 3 for a few years in various companies and in all of them we modified the source code to better suit our needs and never really had much of a problem merging new releases.
As for Unity, I like what they have done, but they are falling behind recently. They are VERY slow to fix bugs and they are late on releasing needed and promised features (GUI system took them what? 3 years from announcement to release?). Also, retina macs are what, almost 3 years old and there is still no support for that (I don't know how unity works with Windows Hi-Dpi support, anyone wants to chime in?)
Anyone have experience working with UE4 on older iOS devices, such as the iPad 3? Specifically, can you make a game that still runs at 60fps on that device? I'm thinking mostly 2D, but with polygon-based meshes instead of sprites for my graphics.
I'm very excited about this. I have been keeping my project under wraps and hope that I can get the prototype ironed out a bit more before I try for this. Just a few thousand dollars to pay for licensing fees of software and maybe a dev machine or two and some marketing.
