Fractal Tree Library with Bezier Curves by txoof

Last crawled date: 3 years, 1 month ago

2D Fractal Tree Library
This library creates 2D objects that can be exported as SVG objects
This does NOT produce a valid three dimensional object that can be printed.
tl;dr:
trunk();
F6 (render)
Export as an SVG in OpenSCAD: File > Export > Export as SVG
Thanks
All Bezier Functions based on the excellent work of @caterpillar Justin SDKhttps://openhome.cc/eGossip/OpenSCAD/BezierCurve.html
Thanks to SteveWeber314 for his fractal tree tutorialshttps://www.thingiverse.com/steveweber314/about
The latest version of this can be found at github:https://github.com/txoof/fractal_tree
Licenses
Releasted under GPL v3 - Share Alike
please see the included LICENSE file for the complete license text
README
This library uses psuedo-random bezier curves to generate a 2D vector drawing of a tree.
There are many paramaters that can be adjusted to produce different trees.
To use this within another OpenSCAD design try:
use

trunk();
Here are three examples all grown from the same "seed" but with tweaks to the parameters
to produce three very different trees.
To produce a tree simliar to a birch tree (thin, tall, quickly diminishing,
and straight) try the following settings:
trunk(size = 900, bend = 25, seed = 40, depth = 12, widthBottom = 100, widthTop = 90,
maxAngle = 45, minAngle = 35, branchProb = [40, 20, 10],
maxGrowth = .9, minGrowth = .9, decay = .85);
To produce a tree similar to a gnarled old oak (thick, branching horizontally, with
a bushy crown) try the following settings:
trunk(size = 600, bend = 150, seed = 40, widthBottom = 300, widthTop = 280,
maxAngle = 75, minAngle = 55, branchProb = [20, 50],
maxGrowth = 1.0, decay = .93, depth = 6);
A reasonable poppler (tall, stright, fast growing and vertical) can be prodced with
these settings:
trunk(size = 1500, bend = 100, seed = 40, widthBottom = 300, widthTop = 280,
maxAngle = 15, minAngle = 15, branchProb = [10, 50],
maxGrowth = 1.0, decay = .93, depth = 6);
To produce a tree of a particular dimension use:
resize([0, 0, 500], auto = true) //resize to a height of 500
trunk(size = 1500, bend = 100, seed = 40, widthBottom = 300, widthTop = 280,
maxAngle = 15, minAngle = 15, branchProb = [10, 50, 30],
maxGrowth = 1.0, decay = .93, depth = 6);
Design Considerations
To create more bent branches increase the bend/size ratio. size = 500, bend = 250 will
make a VERY twisted tree
Setting depth above 9 can result in very long excution times as the recursion grows
exponentially with depth
Paramaters
module trunk()
paramaters:
Denotes parameter that is used internally by recursion and is not intended to be
used from the inital module call
(suggested values in parentheses)
size [real] size of first segment (linear from origin)
seed [real] seed with which to generate a psuedo-random tree
depth [integer] recusion level (1 to 8)
widthBottom [real] maximum width at base of trunk
widthTop [real] maximum width at top of first trunk segment
minGrowth [real] minimum amount to grow the new branch (0.1 to 1.2)
maxGrowth [real] maximum amount to grow the new branch (0.1 to 1.2)
decay [real] base amount to diminish each branch by (0.5 to 1.2)
minAngle [real] minimum angle to rotate each branch (0 to 180)
maxAngle [real] maximum angle to rotate each branch (0 to 180)
branchProb [vector] % chance of one, two or three branches occuring
[%one, %two] ([10, 40]) - %three is calculated as the
remainder of 100-%one-%two. In this case, %50.
step [real] step size to use when generating bezier curves
values approaching 0 are smoother, but take much longer
to render (0.05)
*depthMax [integer] records maximum depth on first call
*distance [integer] records distance from "trunk" - can be used to diminish
branches
*start [vector] records [x, y, z] vector at which to start
*first [boolean] first run sets persistent variables for recursion
growing the branch
debug [boolean] turn on debugging including control points