Atari Pacman game speed

[ravenxau]

while working on a graphical hack of Pac-Man (Tod Frey version) and comparing game speed matrix from the manual with the actual gameplay (via emulation on android tablet) I noticed that there seems to be little difference between the speed of pacman and the ghosts in games 2 and 6....although the manual lists them as being different...I did a simple benchmark of just seeing how far the ghosts went as i ran a standard pattern of just moving pacman from it starting position and going left and straight up to the top wall of the maze (game 6 does 'feel' more intense) ...has anyone else noticed anything similar or different.... perhaps different rom versions or the emulation platform may have something to do with this????

 

 

Edit:- perhaps it's just that in game 6, pacman is a bit faster than game 2...but not by much...

Edited March 9, 2015 by ravenxau

[+Random Terrain]

Speaking of Pac-Man, have you tried this yet:

 

atariage.com/forums/topic/229152-new-pacman-for-atari-2600/?p=3186651

[ravenxau]

Speaking of Pac-Man, have you tried this yet:

 

atariage.com/forums/topic/229152-new-pacman-for-atari-2600/?p=3186651

 

Awesome work - perhaps the most faithful version compared to the arcade game.....Hack 'Em also comes close, and Nukey's Pacman8K is great (considering it uses the original code base)

[Nukey Shay]

Games 2 and 6 are the same. There's a glitch in the original game...

 

 

       LDA    #11           ;2
       LDX    gameSelection ;3
       CPX    #$08          ;2
       BPL    LF8A9         ;2
       LDA    #9            ;2
LF8A9: STA    ramC0         ;3

 

Put simply, bit1 set in $C0 = fast Pac-Man

 

On powerup, gameSelection is given a value of 6 (game#1/1player = $06), and increases with each press of Select to $12 - where it is rolled back over to $02. This way, the upper slow-moving game selections 7 & 8 are grouped with the initial slow-moving game selections. Except that the result is POSITIVE if X is 8 or higher according to the above snippet (gameSelection 8 = game#2/1player).

 

The CPX should have been #$0A according to the manual.

 

It's possible that gameSelection = 0 was used at some point during development, and that this routine was forgotten about when changing the minimal value to be 2 instead (so it always holds a non-zero value).

[tep392]

It's always interesting to see analysis of old code bugs.

[Nukey Shay]

BTW the $09 or $0B thing are only 2 of the possible speeds when you are hacking the code. $01 is the slowest (because 0 won't move at all), while $0F is the fastest. You can do this for the ghosts, too (ram $C1 - $C4). In all cases, the upper nybble holds direction...so don't change those while the characters are moving.

 

9 or 11, weird.