Dev notes: August 14, 2007
Well, the game's been out for a month or so, and I've had some very positive feedback from players of both the full game and the demo alike - and that's been a lot of fun.
However there's a few things to consider.
Add a skirmish mode (3 enemies - red/blue/virus?)
An 'absolute beginner' difficulty mode? The game's really designed to be played with the turret independent of the tankbase and you never stop moving or turning. Some people don't like to play like this. I suspect they manouever, then stop, then aim. Of course they get pummelled by incoming shots. The addition of a sniper rifle + extra easy mode may give those style players a bit of leeway.
Checkpoint saves - particularly on some of the maze levels.
Some players are still not using the powerups - particularly rockets. May give rockets sooner, tweak the gui to give a clearer idea of when/how to use a powerup + add extra training levels.
As an indie developer, I feel I've reached a fork in the road - actually more like a major intersection with lights, several lanes of traffic and workmen digging holes in unhelpful spots - just to make things tricky. I have a number of choices as to how I proceed;
The first option would be to give up indie game dev altogether. And there's some attraction to this. 'Cause make no mistake, developing even a small game is a sackload of work. There's no guarantee of a return, but it's certain you'll be spending hours in front of your monitor while the rest of the world goes out to play. And I've spent a lot of time developing TU already.
Second option is to begin development on the second TU game. Well, I'm keen to tell the rest of George's story, but it's just too soon to start down that road. Better I feel to support what's already on the table rather than to rush back into the kitchen just yet.
So the third option is to update TU1. And this is what I plan to do. TU represents a solid codebase on which to build. It's a bit flaky in spots, but in terms of future titles it's an engine that although tuned to tank combat, could be adapted to any number of scenarios. That shouldn't be overlooked and to abandon it at this point would be to ignore an obvious opportunity.
But first things first. I've begun stripping out the existing GUI code. The existing GUI code was functional for what it was, but once the menu systems and pop-up boxes began to multiply, the GUI code quickly became unwieldy and 'top-heavy'. Of particular concern was the vid-mem used and the number of surfaces (given that each gui object was a separate texture mapped poly).
Instead, I'll be replacing the code with the excellent Spritecandy lib.
Spritecandy has a much fuller command set which should make for easier, more extensive menu animation. It makes use of single surfaces - in other words, a single large texture containing all the interface elements is applied to a collection of polys that the GUI code allows you to individually alter. Nice. May well give a framerate increase as well.
Once the GUI's updated - I can then start looking seriously at the skirmish mode.
Expect another update soon - Dev out.