This repository contains my final solution for the three-part Buddhabrot Renderer labs (Parts 1–3) from the Caltech CS11 Advanced C++ track. The code implements a complete Buddhabrot renderer for the Mandelbrot set, as specified in the lab handouts.
- Part 1: Start the Buddhabrot program and implement the basic rendering technique for the Mandelbrot set.
- Part 2: Extend the program so it can render full images using the Buddhabrot technique.
- Part 3: Finish the program, with emphasis on code cleanup (“beautification”) and improving usability using standard C/C++ programming techniques.
This repository reflects the final Lab 3 version, which includes everything required from Parts 1 and 2 plus the refinements from Part 3.
At a high level, the renderer:
- Samples points in the complex plane as candidate values for the Mandelbrot iteration.
- Iterates the standard Mandelbrot recurrence (
$z_{n+1} = z_n^2 + c$ ) for each sample and determines whether it escapes. - For escaping points, records the orbit (the sequence of complex values visited during iteration).
- Maps the orbit back onto the image plane, updating a density/histogram of how often each pixel is visited.
- Converts the final histogram into an image file, producing a Buddhabrot visualization of the Mandelbrot set.
All of these steps correspond directly to the Buddhabrot rendering technique described in the lab handouts.
Challenges addressed:
- Implementing Mandelbrot iteration and escape detection for many random initial points.
- Choosing a region of the complex plane and mapping it to pixel coordinates in the output image.
- Accumulating visit counts for pixels based on which points’ orbits pass through them.
- Producing a first working Buddhabrot-style image from the collected data.
By the end of Part 2, the program can render complete images using the Buddhabrot technique.
Additional challenges include:
- Recording and replaying full escape orbits to update the image histogram.
- Managing the number of samples and iteration limits to get a visible image.
- Normalizing or scaling the histogram values so the image uses the available brightness range.
- Organizing the code so the sampling, iteration, and image output logic are clearly separated.
According to the lab description, Part 3 focuses on:
- Finishing the Buddhabrot rendering program.
- Code beautification: improving clarity, structure, and style.
- Usability improvements using standard C/C++ techniques (for example, clearer program structure, better handling of options or parameters, and more user-friendly behavior as directed by the lab).
The code in this repository is the final version that incorporates these refinements.
The labs use a standard make-based build system.
make
./bbrot -s <image_size> -p <num_samples> - i <max_iterations> -t <num_threads>
Example invocations (from the lab handouts):
./bbrot
./bbrot -s 600 -p 50000000 -i 5000 -t 5 > sample_image.pgm
<image_size>– image dimension in pixels (e.g., 600, 1000)<num_samples>– number of random sample points to trace<max_iterations>– maximum Mandelbrot iterations per sample<num_threads>– number of threads to use