Knowledge is the key. The more you know, the more you have as a basis for innovation.
This is gold^
As for how to actually do it, everyone goes about things different ways. Whenever I make something (both for magic and outside of magic), I follow this general format (Specifically for physical gimmicks and tools here, not just a routine or a trick):
Step 1: Have the idea and write it down
Step 2: Figure out what the actual end goal is.
Step 3: Figure out a few of the ways you can go about doing it.
Step 4: Draw the actual design of it out
Step 5: Pick the most reasonable and begin prototyping with random materials.
Step 6: If it didn't work, pick a new idea and begin prototyping, but if it did work, figure out everything that you can cut from it that isn't needed
Step 7: Using this revised model as a guide, create an early version.
Step 8: Add onto it anything that you think will make it better, but is still reasonable.
Step 9: Experiment with it and keep developing it.
Step 10: Build a renewed better version
As for coming up with ideas, just write down the end effect that you think of, and keep thinking of it, and ways to achieve it. Find things in life that work similarly to it, and use those as a model.
I second everything RealityOne and Maaz have shared on here.
As for me, creating magic happens for numerous different reasons. I created my first release called Doubting Thomas after I was dissatisfied with a coin through hand routine I learned on Howcast and wanted a method that allowed more wiggle room with my movements. No Pressure was created when I was bored in my apartment and playing around with an empty water bottle and seeing how crushing it could conserve space in the trash, I came up with a plot to crush a bottle under my command. Joker's Heist was created when I was inspired by the works of Shin Lim, Justin Miller, and Asi Wind. As for what I'm creating right now, I'm creating stage illusions based off of current events and my favorite super hero of all time.
Basically my point is that inspiration for creating magic can come from almost anywhere. To further develop on the plot, method, and presentation having a good foundation of magic knowledge is very helpful that way you can make the routine the best that it can be. Maaz's method of developing ideas is also a good procedure to follow as well but do what works for you.
As for me, creating magic happens for numerous different reasons. I created my first release called Doubting Thomas after I was dissatisfied with a coin through hand routine I learned on Howcast and wanted a method that allowed more wiggle room with my movements. No Pressure was created when I was bored in my apartment and playing around with an empty water bottle and seeing how crushing it could conserve space in the trash, I came up with a plot to crush a bottle under my command. Joker's Heist was created when I was inspired by the works of Shin Lim, Justin Miller, and Asi Wind. As for what I'm creating right now, I'm creating stage illusions based off of current events and my favorite super hero of all time.
Basically my point is that inspiration for creating magic can come from almost anywhere. To further develop on the plot, method, and presentation having a good foundation of magic knowledge is very helpful that way you can make the routine the best that it can be. Maaz's method of developing ideas is also a good procedure to follow as well but do what works for you.
So true. The beauty of creating something is that there isn't a right or a wrong way of doing it. What works for one person may not work for the other, and where one person sees trash, another sees inspiration.
Oooo who is it?
Superman. I want to have it ready for his 100th anniversary.
Here's a quick outline on one of the ways I create -
First: Figure out what you want to start with. I usually start with either an existing routine that I like, but doesn't quite suit me, or with the experience I want to create for someone.
Two: Identify challenges.
Three: Identify solutions.
Four: Test solutions until one works appropriately.
Five: Script it completely.
Six: Practice/Rehearse, changing things as necessary to make it work better.
Seven: Perform, review, change as needed.
One piece of advice, though - don't create a trick just so you can use a certain move or gimmick. That almost always turns out to be meaningless "stunts" that have no purpose other than to show off that particular move.
First: Figure out what you want to start with. I usually start with either an existing routine that I like, but doesn't quite suit me, or with the experience I want to create for someone.
Two: Identify challenges.
Three: Identify solutions.
Four: Test solutions until one works appropriately.
Five: Script it completely.
Six: Practice/Rehearse, changing things as necessary to make it work better.
Seven: Perform, review, change as needed.
One piece of advice, though - don't create a trick just so you can use a certain move or gimmick. That almost always turns out to be meaningless "stunts" that have no purpose other than to show off that particular move.