Assimilation: AI is Irrelevant

Reviewed On
Available For

I hope I’m not hopelessly dating myself, but I remember when the pizza joint down the street from the college campus got in a game called Ataxx (this was back in 1990). It was different from a lot of the other games that were coming out at that time, in that is was a turn-based puzzle game. I spent many hours at this game, trying to beat ever-increasingly-difficult AI opponents in an effort to get a higher ranking. (Take that, Droolman!)

Later, in 1993, 7th Guest came out with a rehash of the Ataxx game as one of its puzzles (you remember, the one in the microscope). While the game had its own problems, at least the AI of the opponent in that puzzle was quite challenging.

Now, almost a decade later, Global Star and Small Rockets come out with a new version of this classic puzzle, with mixed results.

In all of these incarnations, the game works roughly the same. There are two kinds of moves a piece can make: to an adjacent space, which leaves a copy of the piece in the original square; and a two-space jump, which doesn’t leave a copy. In either type of move, any enemy pieces adjacent to the target space are "assimilated" by turning them the color of the moving piece. You can win by either turning all the opposing pieces your color or by occupying more of the board at the end.

Assimilation does add a few new twists to the original. One, instead of a 7×7 board of square spaces, they made up different boards of circles or hexagons in a beehive pattern. While this changes the look considerably, I believe it made the game simpler by reducing the number of possible directions of an attack from eight (diagonal moves were legal in Ataxx) to six.

Also added were the concept of ports, which were spaces (outlined with white dashed lines), scatted on the board, which allowed a jump move from one to another, assimilating enemies in the new space as a normal move. This did add a level of difficulty to the game, causing me to have to consider all the ports as potential attack avenues.

Another new development was a new way to defeat an opponent – locking out. If you lock an opponent in so he has no more moves, he is defeated. In prior versions of the game, this would end the game by filling out the board with your pieces, and then announcing a winner based on area controlled. But here, you can win by lock-out no matter who is ahead. This did make for one more thing to be careful of while playing.

Assimilation also introduces the possibility of more than one simultaneous opponent. Having up to seven opponents on one board does increase the difficulty, by producing the need to be aware of all enemy fronts. Also, when one opponent is locked out, his pieces disappear from the board, leaving a vacuum that the remaining players get to fill.

There is also the capability of finding human opponents online. Although I couldn’t find an opponent the first time I looked, I will keep trying, because I find a human opponent to be a much more pleasurable one than a computer AI. (Sorry, Deep Blue.) This is particularly true in the case of this game.

Where this game falls way short of my expectations is in the level of challenge the computer opponents present. I tried all ten boards at the middle challenge setting, and managed to win practically every game. The only time I lost was the first time I was facing two opponents, but once I got used to dealing with that many variables, I had no problem winning, even in the eight-player games. Even the "Expert" level was not a significant increase in difficulty.

While the AI would jump to make an assimilation play, it could not recognize when it was being locked-out, and would not make any move to prevent it. And once an opponent was sealed into an area, the game did not recognize it’s inevitable conclusion, allowing players to stall the game by making (seemingly random) jump moves, which of course left the same number of spaces for them. There was a "fill" button which would automate moves and end the game, so I guess they did cover for that contingency.

The graphics and music were about average for a bargain title (it retails for about $15) such as this. And for what you pay, you get a pretty fair amount of entertainment with the variety of boards. And the online capability does allow for a vast amount of different opponents, any of whom can be more challenging than the AI ones provided. Assimilation ends up earning a respectable 3 GiN Gems overall.

Now if you will excuse me, I’m off to find some online opponents. Look out, Droolman Jr.!


Leave a Reply

Your email address will not be published. Required fields are marked *