This is a big assignment. Start NOW, or you will probably not finish.
I recommend laying out a plan of attack before coding. Data structures are less of an issue for the basic assignment, but if you are doing more than one primitive type, a common effective strategy is to create a generic object class with spheres, polygons, cones, etc. classes derived from it, each with a specialized virtual method to compute the intersection of a ray with that primitive type. To assist in your planning, here is an outline of the steps your ray tracing program will need to do:
- Read file format into scene data and list of objects
- For each pixel:
- Calculate pixel location in world space
- Calculate ray from eye point through the pixel and into the scene
- Calculate ray-object intersections, choose smallest/closest
- Set the pixel to that color
- Write all pixels out to your PPM file
Creating a C++ 3D vector class with addition, scalar multiplication, and dot product operators will make many operations more compact and more like the vector math equivalent. The
std::vector data structures may also be useful for this assignment for your list of objects.
For a file format as simple as this one, it isn't necessary to get too fancy with the parsing. The easiest method is probably to read a line at a time (e.g. with
fgets), identify the type of line by looking at the first character, then parsing it with
sscanf. There are C++
fstream equivalents for these functions, but even when programming in C++, I still generally prefer to use the stdio-style functions. None the less, you can use whichever you prefer.
If you read a line at a time, it is easy to count lines as you parse. If you ever see unexpected input, you can print the line number and line contents to help identify where the error occurred.
It is very difficult to find the error in one of hundreds or thousands of primitives without a systematic approach and simplified data.
Note that you can easily write your own nff files by hand, which can be very handy for debugging. I recommend that you start with a test scene looking from (0,0,0) at (0,0,2) containing a single sphere centered at (0,0,2) with a radius of 1. With a single sphere, it's easier to tell if your loading is working, and with a simple view it is easier to tell if your ray positions are correct.
Start by trying to find intersections with a 1x1 pixel image, which should give you a ray straight down the Z axis with a closest hit at (0,0,1) and second hit at (0,0,3). Move the sphere around, making sure you get the right answer when you miss it, when you are inside the sphere, or when it is behind you. Then, you can scale up to 2x2 or 3x3 images to make sure your ray position code is correct. Once you have the basics working, move up to a larger window so you can visually tell if your sphere is rendering as a sphere.
Only when you are confident of your ray direction computation and intersection code should you switch to the SPD scenes. If you still run into problems, you can successively eliminate spheres from the SPD file to find the one that is not being computed correctly
It is also worthwhile getting image output working early. Printing debugging values works OK for one pixel/one primitive scenes, but for larger images and scenes, outputting values other than colors at each pixel can be a valuable debugging tool.
Viewing files remotely
I have provided a starter build files for CMake and make, both will build a program from any .cpp files in the directory. It is possible to add additional rules to run it on an nff file, convert that to a png, and even launch a viewer when done. This can shorten your debug cycle just edit / build / edit / build.