Player0 and Player1 reposition kernels have been rewritten, but are not yet being tested so the test pattern goes straight down. Still need to redo the reposition kernels for the missiles and ball.
The score/radar/lives/formation routines are functional. If you change the QUADRANT and/or SECTOR in the menu the correct station layout will appear in the radar. For testing the score will increment, lives decrement, the enemy formation is flying in a circle, and the player's ship moves southeast.
Left Difficulty = A for alphanumeric test:
Additionally alex_79 reported that SECAM consoles have a different clock rate than the others:
- 1.193182 NTSC
- 1.182298 PAL
- 1.187500 SECAM
So I was able to revise the TV-TYPE detection to handle it as well.
NTSC
PAL
SECAM
How that works is I use one of the timers in the Harmony/Melody to see how long it takes a single 262 scanline frame to be drawn. At 76 cycles per scanline that's 19,912 total. The Harmony's clocked at 70 MHz, so it's timer will count much faster:
- NTSC = 19912 * 70 / 1.193182 = 1,168,170 = 0x11D32A
- PAL = 19912 * 70 / 1.182298 = 1,178,924 = 0x11FD2C
- SECAM = 19912 * 70 / 1.187500 = 1,173,760 = 0x11E900
The values are off by one from the screenshots due to how the emulation of the ARM timer works in Stella, which results in round-off errors. The values you see on your Atari will most likely differ slightly due to minor variances in hardware.
draconian_20170318.bin
- 1
5 Comments
Recommended Comments