Crappy stabilised kernel demo
First, because I was having "attachment issues" earlier, here again are:
The kernel test image: tiledraw.bin
And a screenshot of it courtesy of Stella: 
So, what does this show?
First, there's a map of tiles: it's an 8x8 grid of 4x8 tiles. Each pixel is a playfield pixel wide and 3 scanlines tall.
Second, there's a little crappy player sprite. He's just animating as though he were walking. He's 8x12px and is one player pixel wide by 24 scanlines tall.
The player is rotating through some different colours, that's actually because I'm incrementing his "defense" points and it's doing a colour lookup to determine what type of armour he's wearing.
What you do not see in the still is that every (n) frames the screen seems to jump, I think the sound code is spiking it over a threshold during the game logic section, I'll have to scatter some WSYNC's around more liberally to get rid of that. I think it's every 16 frames or so... for reference, the walking action changes every 4 frames I think.
There's no music in this demo but I think I left the music playback code in. The movement code is stubbed out mostly because I was rapidly changing the map formats to do this new 24-line-kernel version and I kept breaking the "am I walking into a wall or not" logic.
The TO DO list is much shorter now. I want to try to do the dual-banked version of the kernel to see if I can work colour changes back in. Oh, and the size isn't terrible either. The whole unrolled kernel is "only" about 1k long, with a total code length of about 1.5k. That leaves a "lot" of room (for a VCS game) for data.
Oh, the player's positioning logic is accurate to 2 vertical lines, so he can "slide" between tiles in a fairly natural way.
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
3 Comments
Recommended Comments