8/12/2012 - 7:22 PM

Unity 2D framework comparison

Unity 2D framework comparison

"-" negative point
"+" positive point
"+-" could go either way depending on your opinion or if it functions or not

Background Info: the last project we worked on needed basic 2D functionality (sprites and animations) and SpriteKit was born to fill this need ( It has no fancy editors and was really made to fit a specific need and as of yet it has not been extended much further than the video shows. Something a bit more full featured is required for a new prototype and instead of spending a bunch of time extending SpriteKit we went on a search for alternatives. None of the below frameworks are perfect and they all have pluses and minuses. One thing SpriteKit had that none of the others do is automatic texture selection based on screen resolution. That is one thing I would like to see incorporated in every 2D framework and it kind of surprises me that it isn't one of the first features added when making a 2D framework.

ex2D (

+ top notch animation editor that is leaps and bounds beyond the rest
+ includes a handy clipping system
+ font editor is excellent
+- excellent texture atlas editor but it doesn't always work
- documentation is lacking
- video tutorials are outdated, most lack voice over explanation and there aren't many
- demo scenes are limited and all rely heavily on Unity's animation system (which is fine but I want ex2D demos NOT Unity animation clip demos)
- lots and lots of components for each GameObject
- while testing my sprite atlas broke extremely often and then it would magically work after switching scenes a few times
- no source code without expensive pro license

2D Toolkit (

+ everything "just worked"
+ good documentation and the video tutorials available are good (though there arent many)
+ includes source code with standard package (I can't resist combing over source and making my own changes)
+ does non inundate GameObjects with piles of scripts
- sprite animation editor needs some love

Uni2D (

+- it pretty much does only one thing: creates a collision mesh from an image
+ it's still young so there is plenty of time for it to gain some features (or get renamed to properly label its purpose)
- should be renamed to "Image to Collision Mesh"
- sprite batching is totally broken

Futile (

+ uses GoKit for tweening (self love time!)
+ open source accepting pull requests
+ it's only in alpha and it ALREADY has more video tutorial time than all the rest
+ works extremely well with regard to handling different platforms
+- very, VERY code centric approach (good if you like it)
+- very Flash-like naming conventions (personally makes me have nitemares of the Flash development days)
- no editors available
- it really has no editors