Looking for disassembly sources...

[Staalwart]

Hello everybody! It's been a while! I'm about to start a project in Atari 2600 that requires from me a study of the "Human Cannonball" (CX-2627) disassembly source code. Anyone knows where to find it? So far my search has come dry, and I would like to have a commented version of the code instead of an uncommented, very-heavy-to-understand disassembled code generated by myself with Distella... Please help!

 

BTW, I own the original cart...

[Staalwart]

Anyone, guys? I disassembled the original game, but as I was saying it's pretty hard to separate the code that I need. Gosh, what I would do for ASM-to-bBasic translator!

[Wickeycolumbus]

I don't think anyone has done a commented disassembly of that one. What part do you need?

[Staalwart]

Hi, Wickey! I'm interested specifically in the part that controls the physics of the human cannonball himself. That is, the ballistic trajectory function generator. A pity no one has disassembled and commented this game, being one of the few (if not the only one) of the original 2600 carts that relies mainly on this particular aspect - the ballistic trajectory of a projectile, in this case a human cannonball.

[LS_Dracon]

Search for "subpixel movement". The code consist in add a value plus the carry flag wich not aways are set.

 

Here is one thread about it :

http://www.atariage.com/forums/topic/93452-subpixel-movement-and-friction/page__p__1135115__hl__subpixel%20movement__fromsearch__1?do=findComment&comment=1135115

[Staalwart]

Hmmm... But, Dracon, isn't this principle an example of linear movement instead of parabolic movement? Although I could use a combination of horizontal (constant) and vertical (non-constant) speed to achieve the effect, using the "speed" of the projectile as the constant X variable and the "angle" as the Y variable factor...

[LS_Dracon]

Yes but for the X be constant in Atari hardware must to be at least 1 pixel shift per frame. Sometimes you need half of pixels per frame. This is why it's called subpixel movement.

With fractional speed,you can set "for X, shift 1 pixel left every 3 frames" as constant and for Y "start shifting 2 pixels up and decrease half pixel every 3 frames", simulating gravity.

All you need to do is increase or decrease x and y variable. Keeping the value fixed the result is fixed speed, but in this type of game there is the wind factor that changes the X speed also.

I think it's what you're looking for.

 

 

I use this technique in my game

[Staalwart]

Yeah, that's precisely what I have in mind! I mean, the Xj=Xi+n and Yj=Yi-m, where j is i-frame+1, n is the initial horizontal speed (to be estimated by the actual initial speed and the elevation angle, I know I can't do trigonometrics in 6502) and n the "gravity" constant. For To (initial state), Xo=0 and Yo=an initial value, again to be estimated by the initial given speed and the angle. At least I think that's how it's made in Human Cannonball - and hence my initial request - but naturally without the benefits of subpixelling. Am I correct in my reasoning?

 

By the way, Dracon, I can't see your video... :-(

[Staalwart]

Oh, another thing... There is another game based on ballistic trajectories, although much less known than Human Cannonball: Xonox's "Artillery Duel".

[LS_Dracon]

I don't know how de Human Cannonball code works, but I think you're correct.

 

There is Incoming!, a homebrew game much more fun than Artillery Duel

 

On the previous link for the sub pixel movement thread, Thomas Jentzsch posted a zip file with source code, give a look to learn about the code.

[Staalwart]

Sure thing! Thanks! I'll be in touch when I have questions!