Classic Fractals Source Codes

[Blue Raven]

Hello, I just thought, that for beginners of fractal programming, this website could be useful. There are about 10 well commented fractal source codes (sadly, but only written with C++) and explanations how they are working, but about once a week or two, the new ones are added.
http://www.bio-matic.com/programming-works/science/fractals/

[cKleinhuis]

nice attempt, and hello and welcome to the forums

indeed i do not like the technical description in a c++ program, basic things that you might be familar with, like calculating the length of a vector, are hardcoded in you program, to introduce a technique, i think it is rather more helpful to describe it in what has to be done, and define such things seperately, e.g. in a function, as there are:

length(complexnumber)
add(complexnumber1,complexnumber2)
mul(complexnumber1,complexnumber2)

in such a way the mandelbrot way of calculating is more easily recognisable

e.g.

while(length(z)<bailout)
{
z=mul(z,z);
z=add(z,pixel)
}

in such a way the methods and the inner structure is very easy to see and modify

so, and when you want to drive it further, e.g. calculate higher or non integer powers it would make sense to introduce the polar representation of a complex number, because with that view, the complex exponentiation is as simple as "length^exponent, exponent*angle" which then needs to be
converted back to a polar representation...

[Blue Raven]

Well, I understand what you want to say, but fractal generation using defined functions like pow or etc is several times slower, than using just simple arithmetic operations. For example in C++ pow(x, 2) is about 7x slower than x*x . I have already tested it. Mandelbrot generation with 256 iterations takes less than a second while using arithmetics and about 7 seconds while using default C++ functions like pow. But anyway, thanks for your comments, I will try to perfect the given codes and I am sorry for an anonymic try to promote this website.

[cKleinhuis]

optimizing code is a complete different section

this is a whole lot that can be placed in a "remarks&optimisations" section, as well as the various speed improvements for calculation, like squares of same iteration depth finding and so on

[top-quark]

It's C++. You can declare functions inline and get the best of both worlds - modularity without the call overhead. Just sayin'.

But yes, primitive arithmetic is so much faster than math function calls that it makes sense to do things the hard way.

[Blue Raven]

Good, thanks for your response!