this dispute make no sense.
if someone want a mobile game → Unity3d is the only choice
if someone want a cool AAA-game for PC/console and you doesn’t care about mobile platforms → UE4 is the only choice.
my opinion based on PRACTICAL experience.
a lot of people ask Epic to fix the support of mobile platforms but 4.3 didn’t bring any major improvements for this.
Assuming UE4’s mobile support won’t be in an acceptable state by the time a game is ready to ship.
Assuming the indie team can produce AAA assets and has the funding necessary to license middle-ware such as Enlighten.
Epic touched on this in a forum post. They are working to improve compatibility and claim it’ll be better come 4.4. At the rate they are advancing with each update, I believe them and wouldn’t be surprised if most of the problems are fixed within a few updates.
I think a lot of people think the grass is greener but the issues with UE4 aren’t as easy to deal with as the current issues with Unity. UE4 has a lot of things they need to fix with their lighting and lpv, while sounding like a good solution, is not very good in practice. With LPV there’s a radius of influence and you can see the lighting popping and changing as you approach. You can see divisions in the lighting grid too. For distant lighting you would still probably need light baking or some type of ambient lighting fake. It’s almost certain that these methods would not match the look of lpv lighting.
One of the main reasons I’m using Unity is that it’s much easier to code and implement things. For terrain UE4 sounds promising but it’s easy to forget that each terrain heightmap takes up a lot of memory. Terrain heightmaps are not dynamic or able to be generated at runtime in UE4, and even with that in mind their collisions are not very reliable (at least as of the time I was testing). I knew early on that a procedural method would be better for what I wanted and the detail that I wanted. Unity felt like a better fit for that. There’s no way I’d do such bulky mesh generation in a blueprint construction script and UE4 coding was beginning to get on my nerves.
Unity has something somewhat elegant about it’s workflow and pacing that is hard to appreciate until you’ve tried something else, especially UE4 which can have a very bloated feel (if that makes sense). I like the UE4’s tab system though.
On the UI creation side of things UE4 doesn’t have a good UI system yet either (as far as I’m aware), so if you’re going over to UE4 for the UI tools prepare to be disappointed. They currently do UI in a very similar way to Unity, albeit with blueprints.
Of course we could create our own game engine!
I started making one years ago. It had PhysX, integrated python scripting, import of 3DS files, all sorts. Could only make games in it’s own player though. Then I got bored with it because PhysX couldn’t do Mesh-Mesh collisions (like Unity). I wonder if I still got it somewhere? Probably lost on an old computer. Oh well. Got me my first job in software though.
I remember another time I was making a social network. Then I though. Nah, this will never take off. So I stopped that too. Two years later… FaceBook.
Why don’t I have ideas like that anymore? Maybe I’m getting old. 
Maybe I’m having too many thoughts and they’re cancelling each other out. Maybe I’m just lazy.
1 Like
NOT greater than me. 
I don’t think we need ANOTHER Unity vs Unreal thread. Everyone has an opinion, there are existing threads on this subject already where you can add that opinion, please search before you post so we can keep everything in one place.
[Edit in response to this]
The forum threads where we are asking for your opinions and experiences give us a more in depth perspective of what your pain points are. Brett (our VP of Engineering) typically asks, who you are, what you’re working on & what you need from certain features.