Jump to content

  • Log In with Google      Sign In   
  • Create Account

- - - - -

Back Into A Groove

Posted by , 30 May 2010 · 169 views

Back Into A Groove

After being a little bogged down for the past few weeks (or months I guess) it seems like I am finally starting to find my stride again. I am beginning to become productive in my writing again, and Hieroglyph is coming along too. Some of the topics that I am writing about are actually being implemented in Hieroglyph, so I guess there is some 'synergy' going on. In any case, I'm happy to be moving forward again instead of floundering on the same project for a while. Come to think of it, I think the vertex skinning is what was taking so long (exacerbated by the fact that I moved to Germany)...

For fun, I have decided to work with one of my friends on a Unity game project. It will be interesting since he is still back in the U.S. while I'm over in Germany, but I think it will be fun nonetheless! I've decided to do the game with Unity due to the rapid development and its robustness (a couple hundred thousand downloads can't be a bad thing!).

I have admired the simplicity of Unity for quite some time, and now that they have made the entry level engine free (mostly) and they are adding support for so many platforms I really do think they are heading in the right direction. With that said, I had somewhat of an open ended question come to mind earlier today, and I figured I would see what the opinions out there are:

While refreshing my knowledge with some of the basic tutorials in Unity, I found myself thinking "Wow, its really easy to get things moving and on screen". But when comparing some of the things that I do in Hieroglyph to what is done in Unity, I realized that there is a hidden cost to such rapid development: I don't think the engine can handle an extended amount of complexity in a project. Since all of the application level code is written in scripts, how complex can you really make your game subsystems? In Hieroglyph, I use render views for special rendering operations - which are fully re-entrant, recursion safe scene rendering operations, that rely on material and geometry configurations for detailed instructions on how to render everything. When thinking of how to do that in Unity, I just scratched my head and ended up stopping thinking about it...

And I think that is the rub - you give up compiled C++ for a scripting solution, and you are certainly giving up some of the power in the language. I am wondering if anyone else has found a similar limitation, or if using javascript for Unity is more flexible that I think it is... Don't get me wrong - I am a fan of Unity and will be using it for my little side project due to its ease of use, but I also don't think it can stack up to another engine with native code flexibility... What does everyone out there think? Have you brushed up against complexity issues with Unity?