I would like to define game concept as follows:
A game's concept is that part of the game that is left after doing away with everything that does not influence the structure of the game's tasks.
As an example of separation of concept and theme, consider the game Reaktor (see also below). Its theme is a nuclear reactor, with the player and the opponents starring as elementary particles. Despite the difference in themes, this game's concept is much like pinball's. You control a particle, which can move, and perform a `kick' action, pushing away all particles close to it. The `kick' is analogous to hitting the ball in pinball. The goal is to kick the other particles into the wall, and (this is very much like pinball indeed) to make the particles hit special bonuses, arrayed against the wall. Note that pinball in general is also an example of the idea of having one concept with various themes, though the reason for having the same concept for the different games is mostly technical. At an abstract level, the concept of both Reaktor and pinball is to `bounce away' moving balls using one or more special objects that can generate a `kick' impulses. Learning how to kick effectively is one of the most important skills in these games. The games differ in how many balls there are, which way they accelerate, to where they should be kicked, and whether your kicking tool(s) can be moved or perhaps replaced by the balls.
I like to think about games this way: don't just clone a game, but strip it naked and take a really good look at it. Then you can perhaps see more clearly what its good and bad points are, in what ways it could be varied, and its differences and similarities with other games. Those are not spaceships that have to be shot, but they are square blocks that have to be touched with the pixels moving upwards from a moveable cursor. Who said action games contain gratuitous violence?
For example, consider limitations on ship movement speed in Galaxian-type shoot 'em ups. Because your ship moves slowly, it is hard to avoid and shoot enemies. The ship movement task involves a great deal of keeping your ship in a safe position so as to prevent getting closed in by future attacks. We may also view the ship as a cursor, which has to be aligned and de-aligned with moving objects. Now, restricting ship speed is suddenly less meaningful, as restricting movement of a cursor in a text editor is not meaningful. Is it really the right thing to do? If we place no limitations on the ship's speed (i.e. move ship movement from the task to the interface), we would be able to make other tasks more difficult. For example, we could increase the speed of enemy attacks.
Perhaps it is this lack of novel concepts that puts classic gamers off modern games. One Doom clone is OK (though perhaps a bit lacking in real action), but after about ten you get bored with the concept. On the other hand, some people say that the newest Doom clones are really not like the old Doom at all, but are really highly-evolved and novel games, having Doom only as a common ancestor. When looking at the tasks however, one might say that Doom clones are not new. For example, in many of these games, the technique of incremental advance works best: take a few steps, shoot at every moving pixel till it stops moving, and advance further. Another common task is recall: getting anywhere usually requires memorising the rooms, routes and secret spots. For a die-hard action gamer, it is the nature of these tasks that really make the difference between a good and a bad game. Incremental advance and recall will hardly be considered action.
The usual attempts at making the game more complex only results in the addition of:
So, from an action gamer's point of view, the `highly evolvedness' of current games is not necessarily a good thing. Like some other software products, the games are essentially old frameworks, with more and more inessential gimmicks (at least to action gamers) added on without redesign. In the same way that sightseers like to see new stories, levels, and graphics, action gamers like to see new concepts. However, action gamers have been denied the thrill of a really new action experience lately. No wonder they are not happy. Furthermore, many game concepts already exist in older games, and could be elaborated upon. Many of these concepts have never been perfected, and are really asking to be worked out further. However, except for the occasional unsuccesful remake, this is simply not done in any meaningful manner.
Here's a personal pick of some of my not-so-everyday favourites (descriptions will hopefully be filled in gradually over time):
