Creating a new Fractal piece

[The Rev]

I'm interested in how each of you goes about the process of creating a new piece.  I ask because I tend to just start picking stuff, tinkering with parameters, trying this or that, and allowing the "where I'm going with this" to just reveal itself as I go.  Only once have I started a fractal with some idea of what I wanted the end result to be.

Is this the case with the rest of you?  Or do you work towards a goal you envision before you begin?

The Rev

[Sockratease]

Hope you don't mind me moving this.  It's better off here  (Meet & Greet is mostly for intro's and such).

As for the question - The closest I come to planning ahead is a color scheme I'd like to use.

The rest is just Zen.

I often try an idea, or try to make something similar to something I saw elsewhere, but get so sidetracked that I wind up doing something wholly other than what I intended anyhow!

Or I'll try to see what happens by plugging favorite numbers into odd places, just because I like the numbers!  (primes, pi, e, etc...)

I can sometimes make what I want, but rarely.

[Thunderwave]

I get a concept in my mind about a program I know, and then I try it out.
Apophysis: I usually think about which formula I want to explore and then I test its possibilities.
3D is very new to me so I just explore all it can do.

I have to have an idea in mind in order for me to work it through, otherwise I feel I am wasting time.  This doesn't mean I always get what I want, but mostly I am surprised by what I get.

[lkmitch]

I generally have an idea in mind--I very rarely start from "scratch."  Interestingly, the images for which I had a fully-formed plan are generally not as good as my more "free form" images.  I guess I tend to overthink things.   

[Wel lEnTaoed]

I tend to "go with the flow"  Which BTW tends to distract me from original idea.

[pfrancke]

always such a thing is limited by what one knows....  the more you know the more options there are.  When you know nothing, then start with what you like and play with it.

[ker2x]

i write code, i patch code, i explore and improve algorithms, i test stuff, i try to understand things.
The fractals rendering i post here is an incidental product derived from the coding process, a by-product 

[heavenriver]

I go by instinct most of the time. I have some guidelines I can follow (like starting with formulas I know well and whose effects are more or less familiar to me), but I simply can't sit still, so I keep exploring all the time. That's a rather scrambled way, but I've learned a lot with this fractal empiricism!

[DarkBeam]

Me = randomness

Anyway lately I posted images made while testing my formulas if they came up good

[Xazo-Tak]

I tend to experiment with different settings to work out what makes awesome and what makes beiber.

[_db_]

I start with a greyscale gradient and usually with an image that has previously caught my eye.  Then I play with parameters, colouring formulas, and sometimes the fractal formula looking for interesting shapes. Then I crop, rotate, zoom, etc. to try and get an interesting composition. This is the most time consuming part for me, I may many hours and sometimes days before finding something worthwhile.

Once the basic composition is there, I try to embellish and strengthen it. I may add layers with different colouring formulas to add texture to the shapes or visual interest to the areas around the basic shapes, to add visual flow and emphasis. Then I experiment with colours until I arrive at a workable palette. Finally I refine the colouring by adjusting individual colours.

I often save quite a few interim results, some will have certain attractive elements that I try to combine in the final piece.

Usually most of my time is spent trying to find the basic shapes and composition. When I find nice shapes I sometimes do a series of variations.