If you’ve followed the campaign from the start you know I’m a fan of changing strategies if I even catch a whiff that my current strategy isn’t working. I typically don’t wait for mistakes to become ruinous before acting – if you think you’re standing on quicksand, step off.
For the past couple of weeks I’ve been sending out versions of the alpha to testers on a wide range of systems. I’d say the majority of those systems had at least one platform or hardware problem of some kind – shaders bugged out, unsupported libraries, quirky input issues, terrible framerates on low-end machines, etc. In a lot of cases the game was unplayable. (This is all on top of the platform-independent gameplay bugs, of course.) The machines that ran the game properly mostly mirrored my dev machine – mid-to-high-end WIndows machines. I made a decision right off the bat to squash those problems and get people playing on as wide a range of machines as possible. I was happy with this decision at the time.
Now it seems like a bad call. Terrible, even. And not just because it’s tedious (though it is tedious). For the past week in particular I’ve focused on nothing but compatibility – my original schedule lists today and tomorrow as ‘Path Editing System’ and ‘NPC generation’ but instead I spent most of this morning either a) hard restarting my Linux machine after FRONTIERS mysteriously crashed on startup, b) rewriting a ton of classes for performance improvement and c) implementing a patching system, with all the accompanying difficulties you can imagine.
In other words: What the fuck am I doing?
A minute ago I glanced at my calendar and it hit me like a ton of bricks: there is exactly one thing I need to be focusing on at this stage, and that’s gameplay. Hardware and platform compatibility is a black hole – I don’t even have a clear idea of how long it will take to fix these issues, and meanwhile gameplay is getting shoved aside. Hell, I still haven’t implemented chopping firewood. I’m standing knee-deep in quicksand and I only just noticed.
So here’s my strategy moving forward: I’m going to focus on gameplay. With respect to hardware and performance, I’m taking the path of least resistance in the short term and focusing on mid-to-high end Windows machines. Linux and OSX testers – I’ll still be releasing versions for you guys, and I’ll continue to develop with Linux and OSX in mind, but in the short term if you’ve got a hardware issue then you’re going to be shit out of luck. This means I’m going to lose the input of a few good testers, but there’s no point in keeping you guys if all you’re testing is a hardware demo. Meanwhile, any new testers I bring on will be weighted in favor of optimal machines. Once gameplay is going strong and I’ve got people testing for balance and fun factor, then I’ll return to hardware and platform support, and not a minute before.
In the long term nothing has changed. I’m still hell-bent on bringing this thing full-featured to all three platforms, and I’ll be damned if you’re going to need a beast of a machine to run it. But that’s a secondary goal. First comes the fun factor, then comes the rest.
Okay. Time to get these NPCs generating.
Read more here: 11/3/2013