One thing I’ve found myself doing repeatedly whilst building this ‘ere Atari 8-bit game is trying to wedge colours into it – that’s fine and dandy of course, as long as those colours make sense in context but I’ve really had to restrain myself for the last couple of days to prevent the entire thing devolving into a mass of rainbows! Some Atari fans would actually like it if that happened, but I pretend to be more subtle and it won’t be happening on my watch!

The other issue of the project not having a title has been slowly eroding me internally (okay, that’s an overstatement but it’s still a bit annoying) so something had to be found and the game now has at least a working title of Energy Arc. This was chosen partly because it survives the traditional “sci-fi voiceover” test (where the name is said repeatedly in the gravely tones required for science fiction film trailers) but mostly because I’m pretty sure that there isn’t a game beginning with E in the list of titles I’ve written.

Yes, things have been quiet… partly because I was putting words into what I’m hoping was an interesting order for Retro Gamer over the last week or thereabouts. That’s done now, so back to the programming!

Nothing has actually been coded since the last post apart from a temporary titles page at the start of the current A8 project (and a name for that is still not presented itself, this is the longest time I’ve known a project to remain nameless) but one of those Atari 8-bit versus C64 threads in the programming section at Atari Age has seen me mulling over some theory – for example, I’ve been wondering if it’d be possible to produce a working 2D action game (and we all know that’s a euphemism for “scrolling shoot ‘em up” since this is me pontificating) using the A8′s 80×192 16 luminances mode.

I’ve already done a few tests previously to get hardware sprites over a vertically scrolling sixteen luma background (using single colour objects for the action and splitting colours on every scanline like Warhawk does for the player, that’s it in the screenshot above without a real background decoder running) but these more recent thoughts revolve around using a character-based screen and software sprites. The problem is resolution, of course, and a quick attempt at doodling a player sprite or some tiles at a 4:1 aspect ratio will rapidly demonstrate that to just about anyone who isn’t a legend at pixel art (so not me, then)…

But it isn’t impossible to produce something workable as such, so if this goes any further the target will probably be more proof of concept than complete game, with simple enemy movements and so forth. And it’s a “back burner” project for the moment so it may go nowhere for quite a while, because “Nameless” is staying on the front ring.

Righto, the ongoing Atari 8-bit project still doesn’t have a name as such (it has a couple of nicknames right now but they’re all expletives) but progress is being made. The sprite rendering is all in and working for one joystick-controlled player object, a bullet for the player, up to six enemies (the ones in the screenshot running on a single enemy “command” are heading upwards for 32 frames, then left and up for another 32 before cycling back to the first move) and collisions between objects are being detected and reacted to accordingly apart from player death which causes a deliberate stall because there’s no lives counter yet!

The Atari 8-bit sprite data conversion tool I wrote a while back in BlitzMax spent some of Saturday receiving a serious re-write that saw almost everything apart from the file handling replaced! The original was slung together over a couple of hours at the end of last year and wasn’t scaleable to allow for differently-sized objects or even source images – that was major job that has really been nagging at me and it felt pretty good to get it sorted!

I’ve also started using Altirra for second line testing and this screenshot was taken from there instead of Atari 800Win. Although it hasn’t made a difference with this project, Altirra reckons that some of the more… erm, “ambitious” routines in my work folders won’t actually work correctly on a real machine without losing most of the colour saturation. I’ve had to delay it for a week or so, but hopefully I’ll be ordering an SIO2SD for my recently-acquired 800XL soon and that’ll hopefully say definitively what the situation is.