It seems like the hardest part of making a game is coming up with the idea. It seems to be the part that takes the longest, for me at least. So, how do you personally do it? I just kind of sit there for a while and when something comes to my mind, I write it down, or if I like it a lot, I making the game around it.
Try sketching (or doodling) a perfect world. Think about putting that on a screen and how you could control it. Also, check this video out http://www.youtube.com/watch?v=MyUC_28HIvA
I come up with ideas randomly. I find it very hard to just brainstorm an idea for a game. Imagine a game that you would want to play. It doesn’t matter how hard to make or how complicated it is, just think of that and write down the specifics, then go from there.
I generally come up with ideas for games while mucking around building little tile engines, or taking various ideas from other games that I find interesting, and mixing them together. Constantly creating imaginary worlds in the back of your mind while you’re out doing other things helps too.
Coming up with ideas is really easy.
Coming up with ideas that other people might actually be interested in is way way way harder.
Then following through with those ideas following each mechanic to its logical conclusion is even harder than that.
Finally implementing those ideas and finding out which ones aren’t actually fun or interesting and refining the process is the very hardest step of all.
It’s like refining dirt to get a nugget of gold.
… takes a LOT of dirt.
Cas 
Pretty much what princec said.
Unfortunately, your idea of the perfect game will never be the same as someone elses.
Successful games have this problem as well, even Minecraft. Someone will always find a reason to not like your idea, even if they manage to find that reason among 1000000 other things they like.
mostly on the toilet
I agree with Cas.
Ideas are cheap. Most of us probably get many “offers” from non-programmers (muggles?) who have this “great idea for a game that combines CoD and WoW and all I need is a programmer to do the actual coding, oh and an artist, and a sound guy… but don’t worry, if you work for free, I’ll give you 5% of the future profit from the game!”
To answer your question, I come up with ideas randomly, all the time. I have a list of “future program ideas” on my phone that I pick from whenever I have a weekend free. I think it would be harder for me not to have these random ideas! I realize that might sound a bit jerkish to somebody currently going through programmer’s block, but I would bet that most people here are the same way.
I’ve heard other people say that they have trouble coming up with ideas, and that always baffles me. My prediction is that perhaps you just don’t have enough practice programming, so you don’t even know what you’re capable of yet. Try working through some basic tutorials, or some not-so-basic tutorials (inevitable shameless self promotion).
Start with Pong, even if it sounds boring. When you have it working, try sprucing it up. Can you add different types of levels, maybe with obstacles? How about powerups? What about a cooperative Pong game where you’re fighting against some enemy together? How about switching the game up and playing as the ball, trying to avoid the paddles? Can you reimagine Pong as an RPG game, or an FPS, or a turn-based game? How about as a card game? Or a board game? Have as much fun as possible- add explosions, pictures of cats, whatever.
Also, I think you’re approaching it from the wrong angle: instead of sitting there, bored, trying to force yourself to think of the next big game idea, why not do something else you enjoy? What are your other hobbies? If nothing else, go for a walk! Really look around at what’s going on, and think about how you might abstract that into a program. See that squirrel digging up acorns? (I guess he’d be burying them now that it’s fall.) How could you create a mechanic out of that (note that I didn’t even say game mechanic)? Could you make some kind of “squirrel sim” where you buried nuts? Would it be top-down, grid based, 2D side-scrolling, or something else entirely? What kinds of programming constructs would you need to create your squirrel sim? How would you do each part?
And notice how small you’re starting, just thinking about a squirrel moving around in a world. What would the controls be? How would you program those controls?
Those are the kinds of thoughts that pop into my head all the time. And if that kind of daydreaming doesn’t sound fun to you, then you might be in the wrong business.
But if it does sound fun to you, then maybe you go home and keep thinking about that squirrel. Maybe you’re taking a shower and you think “a-ha! It would be pretty fun if I used a grid-based top-down world and used the arrow keys to control the squirrel!” and then maybe you get some free time and you write a little SquirrelSim prototype.
How would you make a game out of it? The daydreaming begins again. Maybe it’s a one-player race against time, where you try to find and bury as many acorns as possible in 60 seconds. Maybe it’s a two player game where one squirrel hides acorns and the other squirrel tries to dig them up. Maybe it’s a memory game where you try to match similar colored acorns. Maybe it’s a typing game where you have to spell words quickly to dig up an acorn.
Anyway, that’s how I come up with ideas for games: by not trying to come up with ideas for games and just thinking about the world around me instead, and by using what I know about programming to try to dissect stuff into abstractable chunks.
I agree with Cas, but I’ll throw in my two cents anyway.
I get ideas while I play games and see ways they are flawed, then think of how it could be made better, realize why it’s flawed, and repeat until I don’t realize why it’s flawed, and then implement that.
That’s pretty much how it goes
There’s even a phrase for the sort of typical game design ideas going around right now… “minecraftbuts”
Cas
important to keep it a game concept, not just a wild phantasy:
#1 avoid feature creep
its easy to have a good base idea, and then add and add and add more cool new ideas and
features to it - up to a point where the game seems to be a humongous undertaking that just cant be realized.
Also, a lot of ideas might change during development, so dont nail down everything in the beginning.
Just keep the basic gamemechanic intact. Other things will change
#2 dont plan games with unrealisic workload
If you are not a (paid) artist, dont plan your game on a high poly 3D world with thousands of assets.
Look for an art style that can be realized coherently with less work.
If you dont get fundet (hobby developer) dont plan the scope as if you where a full time employee.
at one point the game moves from the fun prototype stage to the actual full implementation stage.
If the workload is too much, your motivation will fade and the project likely gets stored away and abandoned.
(maybe serving as a nice mine for code snipplets)
#3 look back on older concepts
I usually scribble 1 - 2 page text documents with lots of ideas to a project, then just save it in a concept folder.
If you look at the concept again after 5 months and it still looks cool, its probably cool.
(and not just a random dreamup of ideas after playing the big commercial game XXX)
I usually get ideas when either on the toilet, eating, being really busy with something (almost) totally unrelated to games and the idea that comes up itself, or when asleep. (Seriously, it strikes me, wakes me up and keeps me awake for almost the rest of the night).
Some ideas are completely random, others are actually kinda usable perhaps? However my main problem is that I do not have enough time to implement even one of them, and somehow when I think I have time and start planning on implementing/designing that idea I somehow manage to not have time for it (or others do).
I guess my main reason for having the ideas is because I study information technology and it’s part of my life, hence the time troubles.
I also guess everyone can have some sort of an own method for finding very specific ideas when looking for them. Mine for example is either playing a very complicated game, or going out for a walk, look at the nature, think about patterns and listening to music (lot’s and lot’s of music).
Oh and I would almost forget pizza. Pizza is the key component to ideas 
My inspiration usually comes from - surprise - playing games. I find a game I really like, then see another game I really like and then something clicks inside my head and concepts from one game combine with concepts from another game and presto: an “original” game idea.
Really, how to get ideas for games? Play games.
Hi
Some of my inspiration comes from other existing games too, I don’t try to reinvent absolutely everything, some basic elements of the gameplay are inspired of some games, especially Turok 2 (body shrinking, …), Golden Eye 007 (weapon reloading by moving them down and up, the bloody screen) and Doom. I think that the reality of the world is a huge source of inspiration for me. Actually, I don’t tell myself “I want to make a game, I have to look for ideas”. I had a drawing to do for my homework in 1999, the very last drawing of the school year. I wrote “TUER” (it means “to kill”) in the middle and I drew people killing themselves by several ways. Between 1999 and 2006, I became a student unionist and a political activist, I had to do another homework, a big semester project for the university, TUER became T.U.E.R (=Truly Unusual Experience of Revolution®).
You take a shower.
- Jev
Q_3oOUfpdVY
Cas
Unlike what some people suggested, coming up with ideas is not easy for everyone. Let’s take me, an overlord simpleton, as our prime example: I’m way (way way way) better at doing ‘gruntwork’ (like data structure design, optimisation, etc) than at coming up with creative ideas. Give me an idea, and I may make it happen (provided we work in the same office - I can’t handle working remotely). So there: some people really struggle with ideas, while the coding is a breeze. Then there are people that manage both, and the poor souls that have neither. It takes all kinds of people to cooperate, preferably without each individual considering whatever they are skilled at, to be natural/common.
Reading that back, it almost reads like a rant… hm…
I suppose that’s why there are people whose sole job is “game design” too then, on the other side of the coin Maybe that’s the sort of job I want to do. I’m properly sick of programming.
Cas
Non obvious optimizations usually need huge amount creativity I wouldn’t consider it as a gruntwork.
Exactly what I was thinking, I couldn’t have said it any better.
But this is true. Some people don’t have the ability to find solutions for problems, and that’s because they lack a certain type of creativity that allows you to find those solutions*. Be that in actual coding or making a workaround.
- EDIT: Solutions, not problems.