I received this message the other day:
"I don't know how to come up with an idea for a game. My problem is that I can't come up with an original idea. I had many ideas for games, but it turns out those have already been created by other people."
This is a common source of paralysis that I hear variations on pretty often from people trying to get started. Let's look at some ways of bypass these barriers and get back to game making.
You Don't Need the Idea to Start
Contrary to popular belief, it often works out pretty well to come up with the idea after starting it. This is not the chicken and egg problem that it appears to be.
Start building something. Pretty much anything! Anything simple enough that you realistically could finish in a straightforward, predictable way. Make "just a racing game," "just a puzzle game," "just a strategy game," "just a sh'mup," "just an overhead action game," "just a platforming game," etc.
Ask yourself what's the bare minimum basics needed to be coherent as one of those recognizable game types.
Just start piecing it together.
The different ideas, including what's going to make it special, will often develop midway from the details, process, and personal interests, strengths, and constraints.
How the Ideas Emerge
While dealing with those parts and watching it take shape in stages, you'll begin bumping into all kinds of ideas of different things to try. Different ways to approach common problems. Different directions to take the game in. Different things to let the players do.
Some ideas arise from incomplete implementation, a discovery in the cracks of work that you can latch onto and grow into something deliberate. Others may take form while trying to figure out a simpler or more time efficient way to accomplish some immediate goal, involving some approximations or accepting certain limitations. Ideas can also take shape as just needing to fill in for something in a way that you're able to do well (or well enough!).
Doing this on the fly helps find the workable intersection between current capabilities and current curiosity. Often the key "idea" of the game that really sets it apart isn't even something easily described at first, but is instead something subtle in the gameplay that only arises from back and forth tinkering with the machine.
What to Do
Step one: if you're not already actively working on a project, pick an old or otherwise relatively simple type of game or two and start trying to implementing parts of it to get some momentum. Get yourself further along in the process. Create a situation in which you can have ideas that you'll be able to promptly put into action, in a context that's at least partly functioning.
Only with the foundation of some code and functionality in place can the imagination begin to orient itself in concrete possibilities rather than random dreaming. Plenty of concepts that are interesting to think about might not work particularly well in a videogame, or might need a huge budget and large team to get it done. Using this approach you'll always be building in a way that necessarily fits in a game and - just as importantly - within (or just at the edge of) your present skills.
You're not someone else. You don't know the same things, care about the same things, or work the same way. Your work simply isn't going to wind up the same, unless as a deliberate practice and learning exercise you go out of your way to specifically copy some particular example in as much detail as possible (a fine way to just practice when new, though of course that's more questionable when business gets involved).
How Well It's Done Matters More
Concern over originality at the idea level, rather than in the details and execution, stems from overemphasizing the importance of the idea. To put it bluntly that's just not really how media, whether entertainment or art, works. Of course the topic matters to an extent, it can drum up certain kinds of excitement or imply connection to certain audiences. However much of its impact resides in how well it's done in the eyes of the audience that it reaches.
Star Trek didn't invent space ships and science fiction. Quake didn't invent zombies, grenade launchers, or gothic(-ish) architecture. Godfather didn't invent mob films. Super Mario Bros. didn't invent platforming or saving captured princesses. Back to the Future didn't invent time travel. These are so well done that they captured people's imaginations.
This isn't just about TV shows, games, or movies. Moby Dick didn't become a classic because the idea of whale hunting is cool. Moby Dick is so well put together that it creates interest in what it's about.
The idea of Mona Lisa is... well, you get the idea.
Pick something to do, practice to do it better, and you'll discover your own tricks along the way to make them more personalized and unique.
The other approach to come up with gameplay ideas that haven't yet become a common pattern is to rapid prototype, which requires a high degree of development fluency and often returns a relatively low yield. However it's one source for original and decent things that aren't evolved from things that began as clones. I took precisely this approach for InteractionArtist, when I built an experimental interaction project daily for seven months. Those 219 prototypes yielded... just a few iPhone games.
That's still thinking by building. It's just not starting from as established a foundation, accepting a higher chance of producing something uninteresting in exchange for a shot at coming out of the process with something more unusual.
Don't Wait for Inspiration
Waiting for an original idea to just happen is an unproductive trap.
Ideas grow out of action, iterative building, and through collaboration with others that have different tastes and strengths than your own.
Want to come up with more ideas? Start making something.
Even if a project starts out as totally unoriginal, with persistence and some practice it'll soon lead to directions you wouldn't have thought up otherwise.