Oftentimes developers are tempted to simply copy past glories, or worse still steal another's formula for success. But I feel that the study of why a game fails to be a commercial success can be an even more useful tool for avoiding disaster. To illustrate my point I want to examine four games that appeared in the early 1990s, all of which suffered from being of two minds when it came to development.
Battle Bugs is one of Sierra's more obscure titles and a strategy game no less. It has real time game play similar to a traditional RTS, but allows the player to pause at anytime to issue orders to individual units. In this case cartoon looking bugs ranging from ants, cockroaches, rhino beetles and even spiders (which incidentally aren't actually bugs). There are flying units too such as mosquitoes and wasps. All battles are pre-set engagements and usually have objectives involving capturing food or exterminating enemy forces. Despite the lighthearted presentation and lack of base building the game was really hard. Usually there is only one way to win a battle, making the game more akin to the puzzle genre than strategy. I'm not sure who the target audience was for this game, considering that it presents itself like a children's game, yet is far too difficult for anyone but the most skilled and determined tactician.
Robinson's Requiem is open world sandbox set in a sci-fi backdrop. You play the role of "Trepliev 1," a stranded survivor of a crashed spaceship on the planet Zarathrustra. Starting off with nothing but the clothes on your back you are forced to eek out an existence on an alien world. The game tracks your character's vitals and has some curious effects such as a jiggling mouse pointer should your character become ill from harsh elements or food poisoning. It's possible to loose eyes and limbs in this game and you could even be forced to amputate body parts should wounds become gangrenous. Sounds like a hardcore simulation, right? Well, once you get into the story a bit the game takes a turn for the goofy. Amazonian women who speak in broken English want you to fight a T-Rex, and another companion you come across turns out to be a lycanthrope. Did I mention that the acting is incredibly hammy? So in the end your left with a game that feels like it was made by someone with a split personality.
XF5700 Mantis: Experimental Fighter is a space flight simulator with realistic Newtonian physics. That's right, no air drag here and even firing the fighter's nose mounted slug thrower at a dead stop will create a slight backward motion. As you can probably imagine dog fighting the hostile bug-like enemies of F.O.E. (Fists of Earth) was really tricky. For the most part missiles are needed to have any chance of success. Regardless, it could have been uniquely engaging if the game didn't piss all over it's own attempts at realism. For one thing the XF5700 has wings (why do you need airfoils in space?) and a FTL drive. There's also sound in space. Nearly 100 missions in length it occasionally throws in a cut-scene now and then to given a sense of story. Unfortunately, it's so random with dead end plot threads and characters coming and going out of nowhere the entire game ends up feeling like a poor man's Wing Commander.
Amazon: Guardians of Eden is love letter to 1950's pulp serials complete with cliff hangers at the end of each chapter. The story centers around finding a brother who disappeared while on expedition in the Amazon jungle. Betrayal, mystery, a robot security guard and lots of attractive women are just some of the highlights. It also has a cool little program which hits the player with a number of anecdotes and trivia while the game installs off a large number of disks. This was in part because it used limited digitized speech and FMVs (I should note that playing the game in its high-res setting only results in the screen being reduced in size with inventory and other windows filling in all the empty space). A built in hint system is included which is a nice feature, but also brings us to the game's greatest flaw. Its B-movie charm is tarnished by the fact that it's a point-and-click adventure game at heart. Complicating things further is the possibility to miss key items, and timed action sequences that can (and often do) result in gory deaths. I'm not sure why exactly, but I got to play a preview build of this game in a software store and for some reason it featured even more grotesque death screens than what was featured in the final product.
While it doesn't hold true for all situations, when it comes to video game development it's better to do one thing well than do two things poorly, wallowing in mediocrity is the worst possible result. Sticking to a vision though (regardless of the results) will at least earn you fame/infamy. Don't believe me? Two words for you - Tim Schafer. He still has a job as a creative director despite making far more commercial failures than successes. And the reason for that is he chose a creative direction and went as far as he could with it.