*Keywords: camera ray, primary ray, camera, ray-tracing, ray.*Want to report bugs, errors or send feedback? Write at feedback@scratchapixel.com. You can also contact us on Twitter and Discord. We do this for free, please be kind. Thank you.

**News (August, 31)**: We are working on Scratchapixel 3.0 at the moment (current version of 2). The idea is to make the project open source by storing the content of the website on GitHub as Markdown files. In practice, that means you and the rest of the community will be able to edit the content of the pages if you want to contribute (typos and bug fixes, rewording sentences). You will also be able to contribute by translating pages to different languages if you want to. Then when we publish the site we will translate the Markdown files to HTML. That means new design as well.

That's what we are busy with right now and why there won't be a lot of updates in the weeks to come. More news about SaP 3.0 soon.

We are looking for native Engxish (yes we know there's a typo here) speakers that will be willing to readproof a few lessons. If you are interested please get in touch on Discord, in the #scratchapixel3-0 channel. Also looking for at least one experienced full dev stack dev that would be willing to give us a hand with the next design.

Feel free to send us your requests, suggestions, etc. (on Discord) to help us improve the website.

And you can also donate). Donations go directly back into the development of the project. The more donation we get the more content you will get and the quicker we will be able to deliver it to you.

## Prerequisites

Most of the techniques we will be studying from that lesson onwards, will use what we learned about points, vectors, matrices, cameras, and trigonometry in the lesson on Geometry. We will reuse a lot of the things we learned in the lesson Computing the Pixel of a 3D Point about the different coordinates systems vertices and vectors can be transformed into. You should also be familiar with the concepts studied in the lesson 3D Viewing: the Pinhole Camera Model. Be sure you have covered these grounds first before you start reading this lesson.

## What Is This Lesson About?

In the introductory lesson on Ray-Tracing, we already quickly mentioned how ray-tracing can be used to solve the visibility problem. Let's recall that visible surface determination in the context of 3D rendering is the process used to determine which parts of the scene geometry is visible through the camera. We can use ray-tracing to compute visibility (this process was already explained in the previous lesson) by casting a ray through each pixel in the image and looking for the nearest object that this ray intersects (if any). Let's also recall that ray-tracing is a technique for computing intersections between rays and surfaces. Using ray-tracing to compute visibility is also called **ray-casting**.

Producing an image using ray-tracing to solve visibility requires to loop over all the pixels in the image, generate a ray for each pixel, cast this ray into the scene and look for a possible intersection between this ray and any surface in the scene. These rays are called **primary rays** (or camera or eye rays) because they are the first rays cast into the scene (secondary rays are used to compute things like shadows, reflections, refractions, etc.). To find if the ray intersects a surface, we need to test each object in the scene for a possible intersection against this ray. A ray might intersect more than one surface. The visible surface is the surface with the closest intersection distance. By distance we mean the distance from the origin of the ray (which in the case of a primary ray, is the position of the camera) and the intersection point.

In this lesson, we will formalize the concept of ray and more importantly how primary rays are generated to simulate pinhole cameras. Once we understand how primary rays are generated, the next step will consist of learning a few techniques to compute the intersection between rays and geometry. These are the minimum requirements to produce an image of 3D objects using ray-tracing.

In this chapter, we will learn a few useful things about rays. In the next chapter, we will study how primary rays are computed.

## Defining Rays

As stated in the previous lesson,

The part we will explain in this lesson (and in this chapter more specifically) is the notion of ray and the way we can define it both in theory and in programming. The only two variables we need to define a ray are a point and a vector. The point (which programmatically we will simply define as a vec3f) represents the origin of the ray and the vector, its direction. Keep in mind that direction should generally be normalized.

What this point and this direction represent when combined together is a half-line. Mathematically, any point on this half line can be defined as:

$$P = orig + t * dir.$$Where \(t\) is the distance from the origin of the point to the point on the half-line. This variable can either be negative or positive. It \(t\) is negative, the point on the ray is behind the ray origin and if \(t\) is positive, the point P is in "front" of the ray's origin. Practically, when we use ray-tracing, we are generally only interested in finding the intersection of the ray with surfaces located in front of the ray's origin. This means that we will only consider an intersection between a ray and a surface valid if \(t\) is positive.

From a mathematical standpoint, the equation above is called a parametric equation. The half-line is described by an equation which is a function of the parametric variable \(t\) (or parameter).

Ray-geometry intersection routines always return the intersection (if one was found) in terms of the parameter \(t\). In other words, if an intersection was found then the ray-geometry intersection routine would compute the distance from the ray origin to that intersection and return this information to you. From there you can easily compute the position of the intersection or hit point in 3D space using the ray parametric equation we introduced above.

This is really all there is to rays. From a programming point of view, rays can also be defined as a C++ class:

Some programers like to add more member variables to this class, such as for instance a \(t_{min}\) and a \(t_{max}\) distance. They define a range of valid values for \(t\). In other words, if the ray-geometry routine returns a value for \(t\) that is not contained within the range [\(t_{min}\), \(t_{max}\)], then there would actually be no intersection (even if \(t\) is greater than 0).

You can add as many parameters as you want to the Ray class. You can decide to use a Ray class or not. This is entirely left to personal preferences and requirements. There is no rule about the way you represent this data in your program. Some programmers like to add information to the Ray class such as the distance \(t\) to the closest visible surface (which you set in the ray-geometry routine when an intersection is found), a pointer to the hit object, etc. Some other programmers prefer to separate this information from the ray variables and store them instead in a structure or a class called Intersection for instance. Again, apart from having to at least define the ray origin and ray direction, everything else can be done in just the way you like (and is optional).

## What Are Rays Used For?

Rays can be used pretty much everywhere. They are used to solve the visibility problem, gather information about the colors of objects, compute shadows, etc. The type of rays we will be learning about in this lesson are called **camera rays** or **primary rays**.

For each pixel in the frame we will need to construct one camera ray which we will cast into the scene. If rays intersect objects we will compute the color of the objects at these intersection points and assign these colors to the corresponding pixels. This is in short, how a CG image is created with ray-tracing. We can make the distinction between primary rays (the first rays cast into the scene which have for origin the origin of the camera) and **secondary rays** (shadow, diffuse, specular, transmission, etc. rays) which we use in shading. Secondary rays are spawned from primary rays at the ray's intersection point. The direction of these secondary rays depends on their type: shadow (we cast a ray in the direction of the light), reflection (we cast a ray in the reflection direction), refraction (we cast a ray in the refraction direction which can be computed using Snell's law), etc.

## What's Next ?

In the next chapter we will learn how to initialize camera rays for each pixel of the frame, at which point all we need to do to get an image is to implement a simple ray-object intersection routine and store the result to an image file format.