![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
7800 Game Making
Jun. 4th, 2015 06:11 amSo, I did finally settle into a game to try to code on the 7800 using 7800Basic.
It's taking form, as it's a fairly simple concept. Enemies fly in from the sides, your main goal is to graze against them (tag them essentially) without taking a direct hit from them. Your character has two bounding boxes. One that covers most of your character, the other that just pinpoints roughly the center of the sprite. If the outer box collides with an appropriate enemy without the inner box colliding, points are awarded.
The enemies come in waves, you have to last out the duration. At the end of each wave bonus points are awarded as applicable.
Biggest problem is testing on real hardware - I can't do it right now. Even worse, I can't run MESS/MAME either, as it crashes both my computers apparently. MESS is apparently pretty close to real hardware as shown by people who have been doing some tests/screenshots for me. I can use Prosystem... but prosystem isn't anywhere near as accurate. It serves for some basic testing, but I can't trust what I see in prosystem to make it to real hardware. For one it's far more generous on how much can be plotted. This isn't good since while it looks ok to you, you'll have missing information on the screen when playing on real hardware.
Still, things are progressing.
I'm using a high resolution mode for now, "320B" which should give access to 2 4-color palettes. (3 colors plus background for each really). However I've been finding it seems to glitch out with one of the palette registers. P0C1 (and P4C1) seem to fail to show up when they're sharing a pixel-pair with itself or the background color - both pixels simply end up as background. There are some predictable situations when they show properly but this places limitations on exactly where that color can be used in a sprite image.
That's assuming I'm right in what I've seen via testing...
If it seriously gives trouble, I might just downgrade to 160 modes, but I'd really rather keep this game in 320 if I can...
It's taking form, as it's a fairly simple concept. Enemies fly in from the sides, your main goal is to graze against them (tag them essentially) without taking a direct hit from them. Your character has two bounding boxes. One that covers most of your character, the other that just pinpoints roughly the center of the sprite. If the outer box collides with an appropriate enemy without the inner box colliding, points are awarded.
The enemies come in waves, you have to last out the duration. At the end of each wave bonus points are awarded as applicable.
Biggest problem is testing on real hardware - I can't do it right now. Even worse, I can't run MESS/MAME either, as it crashes both my computers apparently. MESS is apparently pretty close to real hardware as shown by people who have been doing some tests/screenshots for me. I can use Prosystem... but prosystem isn't anywhere near as accurate. It serves for some basic testing, but I can't trust what I see in prosystem to make it to real hardware. For one it's far more generous on how much can be plotted. This isn't good since while it looks ok to you, you'll have missing information on the screen when playing on real hardware.
Still, things are progressing.
I'm using a high resolution mode for now, "320B" which should give access to 2 4-color palettes. (3 colors plus background for each really). However I've been finding it seems to glitch out with one of the palette registers. P0C1 (and P4C1) seem to fail to show up when they're sharing a pixel-pair with itself or the background color - both pixels simply end up as background. There are some predictable situations when they show properly but this places limitations on exactly where that color can be used in a sprite image.
That's assuming I'm right in what I've seen via testing...
If it seriously gives trouble, I might just downgrade to 160 modes, but I'd really rather keep this game in 320 if I can...
