Re: [stella] Collaboration

Subject: Re: [stella] Collaboration
From: Glenn Saunders <cybpunks2@xxxxxxxxxxxxx>
Date: Sat, 18 Jan 2003 09:08:52 -0800
At 02:36 PM 1/18/2003 +0100, you wrote:
Well, in my general kernel (missiles differ by one scanline, P0_Y % 1 !=
P0_Y % 1) there is no tearing problem. Only the special two kernels
where the missiles are on the same scanline (P0_Y % 1 = P0_Y % 1) are a
bit more complicated.

You have verified that there is no tearing for all position combinations?

If that's the case then if you add in the simplified ball code I'd consider the main kernel done (save for the score section of course).

If you want to keep hacking away at it to try to get single-scanline Y on the zombies, that's fine with me, but at least make it so that it will be easy to conditionally assemble for either kernel variation because I have a feeling we're going to have to go back to this kernel in the end to make room for important "added value" game features.

I'm glad to eat the ROM space on the kernel and risk making the gameplay ultra-minimalistic if it was the ONLY way to avoid the tearing problem. Tearing is a showstopper. Graphical glitches are things that just shouldn't happen on the 2600. But I wouldn't want to strip it down too much just for the single-scanline Y zombies. It's nice to have somewhat smoother movement but like I said, I don't think it's going to be that apparent.

Besides, I'm anxious to put the controller code back in and get some user-controlled movement in the game. Speaking of which, how different is the driving controller code you used for Thrust+ DC and Asteroids+ DC vs. what I employed in the last build of DD? I'm guessing yours is more compact.

Also, your current kernel is loading the car sprites the same way I was doing it before so I can just plug the controller/rotation code back in, right?

Archives (includes files) at
Unsub & more at

Current Thread