Genre: Arcade      Creation Date: 1997      Language: 6809 Assembly Language       System: Tandy 512K Color Computer 3

After leaving the Tandy Color Computer scene in 1992, I spent most of my computer time playing with  Commodore Amigas. I had an Amiga 1000 and then moved on to an Amiga 1200. I liked the Amigas, great hardware and an efficient, reliable and responsive OS but I never programmed anything for it. When Commodore folded and the resurrection of the Amiga by another company was looking more like a myth, I jumped onto the PC bandwagon with a Celeron 300 Mhz system. I hated Windows! It was the complete opposite of what the Amiga OS was except that it had a huge software base and was supported by so many companies with their products. It was clear that this was the future of home computing (over 90% market share!).

What was also clear to me was that the days of creating commercial software as a one man show operating from a bedroom were over. It was hard to make a dent in this market unless you joined up with a software company. I had also fallen far behind with current popular programming languages like "C" and was unfamiliar with game programming on the PC. Yep! I was an old dog that couldn't learn any new tricks!

So, I did what all "old dogs" do and that is search around for other "old dogs" to talk about the "good-old-days" and I found it in the form of the CoCo newsgroup on the internet.

I was surprised to see that there was still some life in the Color Computer. I discovered new programs that other Color Computer enthusiasts had written during my years of absence. I made some good friends like Jim Davis who at the time was running a disk based magazine called "CoCo Friends Disk Magazine" (CFDM). John Kowalski who in my opinion is the greatest Color Computer 3 "hacker" (The old definition, not the new 'bastardized' version!) who has managed to make the machine do things that it wasn't suppose to do. Steve Bjork, the most famous of Color Computer game programmers. And many others whom if I am to name would fill the rest of this page and constituted most of the Color Computer community at the time.

I was starting to get the bug again and I decided to create just one more game.

Original Story Pretext

"You are PAC-MAN, a friendly-looking yellow circle with a little wedge missing. You travel through the maze, eating dots in your path and are pursued by four ravenous monsters. If any of them catch you, you're a goner.

There are four 'energizers', one in each corner of the maze. Hit the energizers and the monsters are rendered helpless and you can eat them, making them disappear for a few seconds.

If you eat all the dots on the board, you are rewarded with another board, in which the monsters have become smarter and stronger."

I think everyone knows this story and for those who don't.... you missed a great era in the history of video games!  :)

Game Development

I set myself an important design goal for this project. It had to look, sound and play as close to the original Namco classic as possible. I didn't want to do Namco the injustice of another poor clone and I wanted to do the Color Computer 3 justice by showing what could be achieved. The game had to run at 30 frames per second, the maximum frame rate of a standard TV or CM-8  RGB monitor (the monitor sold by Tandy specifically for that computer).

I should point out that at the time, I didn't have the MAME arcade emulator for PC/MAC. Instead, I re-created all the game graphics by redrawing them on my Amiga based on observing a real Pac-man machine and what I could find in magazines. The sounds were sampled on the Color Computer off tape recordings made from a real Pac-man machine.

I improved on my 2 channel sound routines from Cosmic Ambush to the point where sound generation took up minimal CPU time. I improved on my sprite graphics routines to achieve more speed. I used the tallest vertical resolution I could muster on the Color Computer 3 of 320 x 225 pixels. I allowed the maze to take up this full height by moving the score displays to the left and right sides of the maze. This allowed me to create a screen aspect ratio that wasn't too far off from the original. I included all the animated interludes between mazes just as the original and also added the feature from the original where the Pac-man moves slightly slower than the ghosts when eating dots but gains a slight distance when it is taking corners. This is an important characteristic used by expert players of the original.


Information Page

Game Screen

One area I could not work out from the original was the ghost character's intelligence algorithm so I had to develop my own. I spent some time trying to think up of an algorithm that would give similar results to the original and this is what I came up with...

I created 4 "target" points on the screen positioned in the 4 corner areas of the maze. Each of the 4 ghosts is given one of these targets points and they work their way to each of these points. I set a timer on these target points and when the timer count reached zero, they would roll their positions into the next target point in the list. So, a target point pointing in the top left corner of the maze would shift to the bottom right after a time, then to the top right, then bottom left before finally returning to the top left. This kept each of the ghosts "patrolling" the maze without them bunching together for too long.

As each ghosts navigated the maze to reach its target point, it would check to see if the Pac-man was on the same row or column position on the screen. If it was, the target point is altered instantly to point to the position of the Pac-man. The ghost would then begin to close in on this point until the timer timeout reset the target point back to its normal cycle.

With this algorithm, the ghost movement is very similar to the original when viewed at a glance. I was happy with the results of the gameplay and that is how it stayed.

Marketing and Sales

I kept my expectations low for sales of this game. There was no Tandy to sell through, no local clubs, no printed magazines (except CFDM the disk magazine) and not a large user base. I had set a top end projection of 20 copies. I sold 60 which is pretty good considering the conditions.

I sold almost all my copies to US customers. The local Australian Color Computer scene seemed non-existent. Thanks to my good friend Jim Davis who acted as my US distributor and who also helped me to attend 2 Color Computer fests in the US - more in later pages. It gave me the confidence to continue with more Color Computer product development.