January 15, 2014

XNA is Dead. Boo. Long Live What?

I just got home from a little teeny tiny XNA meetup in Seattle that happened to coincide with Steam Dev Days.

It was kind of a weird experience.  The biggest takeaway for me is that, at least for Microsoft platforms, the likelihood of XNA coming back from the dead is almost nil.  For enterprise software development, .NET is king and has taken the place of Visual Basic 6, but for everything else, it's either JavaScript/HTML5 or C++.  I get the feeling that Microsoft backing Unity is more a "meh, Unity gets the basics right and nobody will know about how badly they fuck up the rest until they become advanced" move than an actual strategy.

The really sad thing is that unless Microsoft steps up, XNA may live long and prosper on almost every non-Microsoft platform in existence due to Monogame.  I overheard some NDA stuff that I won't go into, but I find it really funny that there are people who are chomping at the bit to pick up what Microsoft has thrown away here.

That said, I feel my time with XNA is pretty much gone.  My interest in XNA was always related to how it would democratize game development.  Game players who learn how to make games not only end up being better gamers, they also end up being better game community members.  They know how hard it is to do X, and their brief peek behind the curtain into the inner workings of how games work let them have a better appreciation for game developers and games themselves.

However, for a platform to inspire neophyte game developers, you need support, and XNA has lost it.  It's really hard for me to remain passionate about something that's fading away.  So, where am I going to focus my efforts now?


Lua has been used in a lot of games as a scripting language.  All game and UI logic for both Amped and Amped 2 were written in Lua, and the only reason Lua didn't get any credit in the game or manuals is that their license at the time said it was optional and legal saw no reason to provide the credit.  In addition, two of the major low-cost game engine projects targeted at neophytes are shipping with Lua bindings: Leadwerks 3.1 and FPS Creator Reloaded.

However, a lot of the public scripts that are out there are, quite frankly, shit.  Even the little hacks I've published over the last few days are shit.  If I'm going to do this right, I'm going to have to go back to first principles.

Here is my plan.  Starting this weekend, I'm going to try to write a game from scratch against Leadwerks 3.1 Indie Edition.  My goal is to use absolutely none of the scripts that ship with the engine, though.  I'm going to start from nothing and move on from there.

This will do three things.  First, it will give a ground-up point of view of what it takes to build a game using Lua with lessons that can be applied to multiple game engines.  Second, it will help me better understand what it takes to be effective with Lua.  Third, it will help me get ready to better use an engine that I backed on Kickstarter before the C++ bindings

Hopefully, this works out well for everyone.  I'll still remain involved with .NET and XNA, but even more on the periphery than before, unfortunately.

No comments: