Mandelbars is a fractal rendering program that focuses on features that improve image quality. Examples include: high dynamic range images and anti-aliasing.
The currently implemented anti-aliasing algorithm is an adaptive Monte Carlo algorithm, with multiple variance reduction techniques available to increase speed.
The program is targeted at Linux / Unix / POSIX systems and currently has a command line interface.
The only dependencies are the C and Unix standard libraries as well as the GNU Scientific Library. GSL is used for random number generation and running statistics functionality used for the anti-aliasing functionality. The C compiler must support VLAs and complex numbers.
Type make and copy the resulting mandelbars executable to the desired location. Currently only supports GCC, but I imagine the code will compile on any C99 compliant compiler.
mandelbars [-ach] [-f parameter_filename | -L real_location,imag_location, scene_zoom] [-d exterior_stop_distance] [-e stop_std_err_mean] [-E std_err_mean_output_filenam e] [-k kernel_scale] [-l periodicity_check_length] [-n maximum_iterations] [-p num_initial_poi nts] [-P npoints_output_filename] [-s scattering_distribution] [-v visualization_algorithm] [-x image_x_dimension] [-y image_y_dimension] image_output_filename
-a - Use antithetic sampling. Improves speed in images with large numbers of monotonic sections.
-c - Use control variate technique. Uses information from previously calculated pixels to improve speed. Recommended when the fractal visualization is approximately continuous.
-h - Print the usage statement.
-f - Specifies Fractint parameter filename. Fractint can be used to explore the fractal to find interesting regions. The Fractint location can be exported as a parameter file and this parameter file can be loaded into mandelbars to generate a high-quality image.
-L - Specifies location and zoom for the fractal scene.
-d - Specifies the stop distance for exterior distance visualization method. You likely won't need to change this. Default is 2^64.
-e - Specifies the maximum standard error for the adaptive anti-aliasing algorithm. Lower numbers produce a less noisy image. Default is 0.05.
-E - Output a csv file with the standard error values for the image pixels. Used for diagnostic purposes to evaluate the anti-aliasing algorithm.
-k - Specifies the scale of the anti-aliasing kernel. Value is in pixels. Larger values reduce aliasing more at the expense of image sharpness. Default is 0.3.
-l - Specifies the number of iterations to look back to check for periodicities in escape time visualization algorithms. Default is 20.
-n - Specifies the number of iterations to perform for a given visualization algorithm. Larger numbers give better image quality at the expense of computation time. Default is 1024.
-p - Specifies the number of initial points calculated by the anti-aliasing algorithm to determine the initial variance of the pixel. Default is 2.
-P - Output a csv file with the number of points calculated for the individual image pixels. Used for diagnostic purposes to evaluate the anti-aliasing algorithm.
-s - Specifies the kernel type for the anti-aliasing algorithm. Options are uniform, gaussian, and null. Uniform is a rectangular "boxcar" kernel. Scale specifies the width of the kernel. Gaussian specifies a gaussian kernel. Scale specifies the standard deviation. "null" turns off anti-aliasing. Default is gaussian.
-v - Specifies the visualization algorithm. Options are escape_time and exterior_distance. Default is escape_time.
-x - Specifies the x dimension for the image, in pixels. Default is 128.
-y - Specifies the y dimension for the image, in pixels. Default is 128.
image_output_filename - Specifies the output filename. The output filetype is 16-bit PGM.