Wow. A great discussion.
I've got some catching up to do.
Awesome, Suli, that's a great write-up.
Everybody's version of perfect is different. Something as little as a gun animation not functioning in the way that the player thinks it should, may ruin it for that particular player.
For example, I'm into consistency. I don't know how many of saw the Fantastic 4, but the only part I remember from the whole movie was the part that ruined it for me. There was a big fight with a dude at the end of the movie, and everyone was involved, and it showed a close-up of Jessica Alba's face. She had a cut above one of her eyes, and blood was streaking down her face. It went away from her for maybe five seconds, and then went back. And... huh? There was no cut, and no blood. That ruined the whole movie for me, and I walked out right then.
So, theoretically, a perfect game for me would be one with absolute consistency. Everything is cohesive; there are no holes in the way the game world is put together. Some games come close to this, but the fact of the matter is, there will probably NEVER be a day when a game meets that criteria.
A very puzzling issue, indeed.
And I'm with Sup on this one, guys.
Our studio goes through a long, and quite tedious, process to sort out ideas.
We start with the idea bank, which is as several people have already stated; the huge pile of every idea imaginable. And then we whittle it down. We start with the things that just don't make sense. Like, for example, putting a gun in (this is just for example, I'm not allowed to actually talk about what I'm working on) that with a single shot killed every single enemy in the whole level, while healing all of your allies to 1000% health and spawning a purple hippo with laser beams and an invincible sports car with the power to open wormholes and suck enemies into oblivion (heh). Although this may sound super-cool and badass on paper, in a gaming sense, it completely breaks all of the play mechanics and after a while it would just get boring.
From there, we take the ideas that sound really cool, and we know wouldn't break the game, or, if an idea is so good and ingenious, take it and build a whole new play-mechanic around that idea, and set those aside for ones that we would like to see come to fruition.
After we've got all of the really cool ideas together, we start to put the ones we think will be the most fun and work the best together. As more and more ideas see the light of day in a playable form, things start to become clearer as to what's possible and what just won't work. I've had ideas that I thought would be the best thing in the world, and when you put them into motion in an actual scenario, it just isn't fun, or it doesn't make any sense at all. Sure, it sucks, but not everything is going to work together.
People should think about these kinds of things before shooting down developers because their idea didn't come to light. They should also think about this before they submit "OGMGFDG CRAZY AWSUM?!" ideas that they think would make the greatest game on the face of the planet. It just doesn't happen. And to be honest, I don't think it ever will.
While games get better all of the time, they will continue to do just that, get better. They'll get better until games are no longer of interest, or the game industry crashes, but they will NEVER be perfect.
In closing, and getting back on topic, I'm going to be writing up a whole big article just for you guys that answers the questions asked by Colin in the first place. Well, at least from my standpoint.