One Fractal Every Now and Then (@ComplexFractal) header
One Fractal Every Now and Then (@ComplexFractal) profile picture

Trying to post a new fractal every day. Animations are coded with the help of https://teapaw.org/t.co/7pKEKo1gN7 by @ensl0

Joined March 2021
Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)Photo by One Fractal Every Now and Then (@ComplexFractal)
This gives a little more context to yesterday's shader. I watched @3blue1brown's video on constructing a parameter space for the Newton fractal, and made this. The two roots are fixed at 0 and 1, and the remaining root is the parameter.
2
5
13
Another cool detail is that the larger basins meet at that triangle-ish shape I wrote about yesterday. I had a notion of why we chose that shape, but viewing the parameter space adds another layer of understanding to that decision, which seemed arbitrary at first.
1
2
A closer look at what happens near the edge of those weird regions in the last animation. In this one, the stationary roots are at i and -i and the third root moves from sqrt(3) to about 1.8364. I also colored the roots white instead of black to avoid confusion.
9
1
25
Here's an example of a region where none of the points converge to a root. The orbits inside these regions "get trapped" and they don't escape, which is why they don't find roots. These are also familiar shapes so it's kind of funny they show up here.
4
17
117
The two stationary roots are at 0 and 1, and the other root traces the dim path. You can find some neat patterns for when the roots are arranged this way, including regions where initial conditions fail to find roots.
1
2
12
This path is (roughly) the boundary of the intersection between the upper half plane and the unit disks at 0 and 1. This is a significant path because basically all cubics are globally conjugate by a linear map to the cubic where the third root is inside or on the boundary.
1
2
We spent the last couple weeks or so looking at the family of Newton fractals in my complex dynamics class and @3blue1brown also talked about them over on youtube yesterday, so I wanted to share some of the stuff I wrote on smoothstep.io
1
7
This fractal is actually the banner I use for this account. I coded it a while back on Shadertoy when @3blue1brown's Lockdown math series was coming out. If you're interested, here's the video that sparked my curiosity: youtube.com/watch?v=elQVZLLi…
8
z->c*sin(z+k) k = [0,2π] Notice how changing k alternates between the sin and cosine fractal. Also, here's the parameterization for k: desmos.com/calculator/xotbku… It gives k some time to stabilize at intervals of π/2 to show the sin and cos fractals.
2
9
You can extend the definition of sin and cosine to the complex plane in a meaningful way. Pictured here is the fractal generated by the iterative map z->sin(z)*c.
3
An exploration of some of the the stable regions in the Burning Ship.
2
The orbits for the Julia sets are a little less interesting because the function is constant for all points in the complex plane. This implies that if some point z is in the Julia set, then all points in its sequence must also be in the Julia set. This is not true for Mandelbrot.
4
I'm back!!!!! The big red dot is the initial point and the smaller red dots are the points after each iteration of the map. You can see what the mapping is actually doing by following the sequence of lines. You can check it out for yourself here: smoothstep.io/anim/adee1df5a…
2
7
In the last sprint of my semester, so I'm going to take a break from this account to focus on school for a week. I'll be back after my term ends👍
5
39