Blender Foundations: The Essential Guide to Learning Blender 2.6 / Edition 1

by Roland Hess

Overview

Blender Foundations is the definitive resource for getting started with 3D art in Blender, one of the most popular 3D/Animation tools on the market . With the expert insight and experience of Roland Hess, noted Blender expert and author, animators and artists will learn the basics starting with the revised 2.6 interface, modeling tools, sculpting, lighting and materials through rendering, compositing and video editing. Some of the new features covered include the completely re-thought interface, the character animation and keying system, and the smoke simulator.

More than just a tutorial guide, "Blender Foundations" covers the philosophy behind this ingenious software that so many 3D artists are turning to today. Start working today with Blender with the accompanying web site which includes all of the projects and support files alongside videos, step-by-step screenshots of the trickier tutorials, as well as a direct links to official resources like the Blender download site and artist forums.

Product Details

ISBN-13:	9780240814308
Publisher:	Taylor & Francis
Publication date:	07/17/2010
Pages:	416
Product dimensions:	9.50(w) x 11.08(h) x 0.95(d)

About the Author

This book comes to us directly from Roland Hess, noted Blender expert and authority, and is perfect for getting up to speed with Blender 2.6's new interface and many of its new features. Blender 2.6 promises to be a MAJOR departure in interface and workflow from previous versions; Blender Foundations will be a much needed resource for Blender workflows and the preferred one-stop guide for new users.

Read an Excerpt

Blender Foundations

The Essential Guide to Learning Blender 2.6

---

By Roland Hess

Focal Press

Copyright © 2010 Elsevier, Inc.
All right reserved.
ISBN: 978-0-240-81431-5

---

Chapter One

An Introduction to 3D: Recreating the World Inside Your Computer, or Not

You may be under the impression that working in 3D is an attempt to create the real world inside of your computer. It's a tempting and logical thought—the world itself is 3D. It consists of objects that have a certain form, the surfaces of which have properties that make them appear a certain way, and of light that allows us to see the whole thing. We can create form, surface, and light inside of a computer, so wouldn't the best approach just be to make everything inside the computer as close to the way it is in the real world and be done with it?

It turns out the answer is no—accurately simulating the world is not the best approach. That way madness lies.

We all live in the world. We know what it looks, smells, sounds, and feels like. More importantly though, we know what it looks like when carefully lit and shot through a lens by a skilled photographer. It is through two-dimensional (2D) images, either still or animated, that people will experience our 3D work, and this is the target we should be working toward. It turns out that this makes our job as a 3D artist both easier and harder than the job of someone who is mistakenly attempting to simulate the world. It's easier because, well, the world is stupifyingly large and deceptively complex. It's harder because there are considerations other than concrete sensory input (i.e., "A tree is shaped like this, feels that, and acts thusly"), and considerations like what is and is not seen in the final image, and at what level of detail: how it all goes together; how it demonstrates what you are trying to say; composition; art, even.

To be good at 3D creation, you will need to develop a facility for carefully observing the world around you, extracting those elements that will best contribute to your image and leaving the rest out. You will be creating a simulation of the real world that is specifically targeted at producing a final image—a unique virtual mini-world of which the sole purpose for existence is to create the illusion of reality in 2D that we experience when looking at a picture or watching a movie.

Let's take a look at the different elements of a real scene that we will need to analyze and rebuild in order to achieve this.

Form

Everything that we see has a form—a physical structure. We know what these forms are: how a lion is shaped, what it looks like when pudding falls on the floor, the essence of a chair.

The essence of a chair?

Well, what is a chair, anyway? There are thousands of different kinds of chairs, but when we see one, we know it's a chair. That's because although the details differ, the form generally remains the same: a place to rest your back side, some means of support (usually legs, but it doesn't have to be), and a back. If there's no back, it's a stool. And so when we see an overstuffed recliner, a swivel chair attached to a table in a fast food place, or just a basic dining room Queen Anne–style chair, our brain identifies the form and tells us "chair."

It is form that lets us know what things are. In the real world, form is made from matter. Yes, that's "matter," as in solid/liquid/gas from science. I know that no one said there would be science here, but there is, as well as math—get used to it.

A lion's form is made up of organs, bone, muscle, skin, and fur. Pudding is made from milk, gelatin, and, hopefully, chocolate. A chair's form is constructed of any number of things, including wood, metal, and plastic. However, none of the insides really matter to our perception of the form. In general, we only see the outer surface, and that is enough for us to properly identify things.

This is the first place that we decide as 3D artists that we will not simulate the world as it is, but as we would see it through a lens. With a few exceptions, a camera only sees the outside surfaces of objects, so that is all that we need to worry about. If we want to create images of forms ("pictures of stuff" for the layman) we can temporarily forget about what things are made of and just focus on the shape of their surface.

In 3D, surfaces are built from polygons, specifically triangles and quadrangles in Blender's case (Figure 1.1). Usually, these polygons are built from vertices, edges, and faces (Figure 1.2).

These polygons are created and linked together in clever (or not-so-clever) ways until the whole surface of a form is constructed. This construction is called a model. When making models in 3D, it is important to keep in mind how their form will be shown in the final image. If the image of the object will be very small, perhaps because it is far in the distance or just a tiny detail like a flea or a grain of sugar, the model can be very simple (Figure 1.3). There would be no need to create a model of a building in exhaustive detail if it appeared only one-quarter inch high on a distant hillside in the final image.

We know what the form of such a building is in the real world. It has a roof, a chimney, windows with trim, maybe a spigot for a hose, and many other details. However, when we look at the 2D image with a critical eye, we can see that for our purposes, the building is little more than a box with a triangular top. If we were trying to reproduce this image in 3D, we would save ourselves a lot of trouble by modeling a simple box with a triangular top (Figure 1.4).

Obviously, if this same building were the main subject of the image, shown close up, its form in the image would be drastically different, and we would model it differently (Figure 1.5). In fact, depending on the image we are trying to achieve, we might only create a model of a portion of the building. If the image consisted of a close-up shot of the exterior of the building, we might choose to only model that part of the building that shows on camera.

In 3D, models that are made of polygons are our forms. They tell us what we are looking at. As a final example, take a look at the scene project for this book, without any texturing or lighting. It is only the forms, yet we know immediately what everything is. Clearly, though, none of this stuff is real, could be mistaken for real, or is even believable. That's because believability comes not from form, but from surfacing and lighting.

Surfacing

Surfacing refers to the way that surfaces look. Once again, we can examine the real world to learn a few things. We know what wood looks like. However, it looks different when it's part of a tree, freshly cut, or stained and finished in a piece of furniture. Each of these surfaces have different visual properties. A tree is rough—the bark is generally a brownish gray, rough, occasionally covered with moss or lichen. Cut wood is often light (excepting things like walnut, of course) with a pattern of concentric circles. It is generally smooth, although if you look closely you can see a pattern of grain. Finished wood that is part of a piece of furniture can be many colors; for example, it can be extremely smooth and highly reflective, in the case of a grand piano.

So, while models tell us what forms we are looking at, surfacing gives us the additional clues we need to understand what the forms are made of (Figure 1.6).

Once surfacing has been added, the image becomes much more believable. In fact, surfacing gives us so much information that if we mix it up, the scene remains believable in a general sense, but becomes strange. It plays with our expectations of what we should be seeing.

In Blender, an object's surfacing is described by two sets of properties: materials and textures. Materials involve the basic visual properties of the surface, without texturing (Figure 1.7). How does light react when it strikes the surface? Does the surface emit its own light? Is it rough or smooth? Is it reflective? Is it transparent? Does it act like milk or skin, taking a little bit of light inside itself, scattering it around, then letting it back out?

With the basic materials in place, the objects in the scene take on a certain aspect of believability. Even though they don't look real, they at least look like physical objects.

Material properties deal with a number of areas:

• Color: The basic, overall color of the surface in white light.

• Shading: The way that light affects a surface. How much light does a surface absorb or reflect? How do the angle of the incoming light and the viewing angle affect what it looks like? Different shading models are available to give you a better chance at mimicking certain effects seen in the real world.

• Transparency: Whether or not an object's surfaces are transparent.

Simple transparency, based on something called Alpha (Z-transparency in Blender) is quick to render. Another method of calculating transparency, called ray tracing, takes longer to render but is more realistic and able to simulate effects like refraction.

• Reflection: Is the surface mirrorlike?

• Subsurface scattering: Certain real-world materials like skin, jade, or the flesh of a potato exhibit this property, often abbreviated as SSS. Light enters the material, scatters around, possibly changes color a bit, then exits. When you hold a flashlight against your fingers in the dark, you see subsurface scattering.

While these material properties affect an object's overall reaction to light, textures help to define the way that those properties vary across the surface. Textures can be photo mapped onto a surface, generate bumps on a surface, or cause transparency to fade in and out, among other things. A careful observation of objects in the real world will help you to determine which combinations of material settings and textures will produce the most believable results.

Light

The final ingredient in your still images is light. In the real world, light appears to be easy. You have the sun and the sky. If you need more light than that, you flip a switch. That's it!

Of course, good photographers know that it's not that simple. Even if they are shooting outside, they carefully monitor light sources, highlights, and shadows. They might use a reflector to bounce some natural light into an area that is too deeply in shadow. Indoors, things get even trickier. The great indoor photography you are used to seeing in magazines (not to mention television and movies) is usually the result of careful planning and expensive equipment.

It should be no surprise then that lighting in 3D isn't easy. In the real world a lot of lighting, even in complex commercial situations, is accomplished by the simple fact that light bounces off anything it hits, and scatters in the air. Shine a focused bright light into a completely dark room and you will be able to see quite a bit. You will see that objects that aren't even near the beam of light are illuminated, either from the beam bouncing off the opposite wall or simple atmospheric scattering. In 3D though, putting a spotlight into a dark room gets you next to nothing (Figure 1.8).

Some 3D systems allow you to define lights in a fairly natural way: add a sun, add a lamp, done. Blender, however, does not work this way. If you want light to "bounce" off a far wall, you will have to add another light source on the wall, shining in the direction of the bounce. It can become complicated.

In 3D, there are a number of different light styles, each with their own set of properties and effects. Blender's Spot Lamp, for example, can cast the kind of "light cone" that many people expect to see in concert or film noir situations. While real-world shadows are simply a consequence of the physics of light, 3D shadows can be turned on or off and have several calculation methods, each of which is appropriate to a different situation.

More than any other aspect of the 3D creation process, effective lighting will require that you carefully consider the 2D images you want to generate. Beyond just showing the forms in a picture and their surfacing, lighting sets a mood, directs the eye, becomes a part of the composition, and will either make or break your final image.

Motion

In the real world, things move. We're accustomed to the way that things move, and when we see unnatural motion we know it right away. It looks bad. As 3D artists we are not trying to recreate the world. What artistry or skill would there be in a painstakingly exact reproduction of real motion, other than technical curiosity? What purpose would it serve?

Fortunately, this is ground on which lies a well-worn path. Since its beginning in the 1920s, artists and entertainers have a developed a well-known language for animation. Things that would look bizarre in the real world—bulging eyes, rubbery arms, and the ability to hover for a moment before falling off a cliff—are perfectly acceptable and actually expected by the viewer in the realm of animation.

Whether for film, television, or even content produced exclusively for viewing on the Internet, animation works the same way. It is a series of still images shown rapidly one after the other, producing the illusion of actual motion. In traditional (hand-drawn) animation, each of those frames must be drawn individually. At 30 frames per second (the frame rate of North American television), that's 900 images for just 30 seconds of animation. In other words, it is a lot of drawing.

Computer animation, which is what we're about, is in some ways easier. If you were to follow the traditional model of animation, you would create the entire scene 900 times, each with a slight variation. Obviously, that's silly. Within your single scene you have the ability to record and change objects' positions (e.g., on the floor, in the air), states (e.g., lights on, lights off), and poses (e.g., reaching for a towel, pulling on a shoe) along a timeline, and have the software calculate all of the positions, states, and poses in between. The process of saving this information in time is called keyframing ("key" information is saved on certain "frames" in the timeline; Figure 1.9).

In Blender, keyframes are created for entire objects by transforming them in space and recording their status. When you add several keyframes, Blender interpolates the difference between them along smooth curves. Character models are not animated directly. Instead, they are attached to control rigs called armatures. These rigs act like a skeleton and control panel that can be posed and keyframed to produce beautiful character animation.

Rendering

Even after you have created your forms, carefully surfaced and lit them, and animated your scene, there is more to do. Just as there are many variations in cameras in the real world (Film or digital? Color or black and white? 35 mm lens or 50 mm?), there are an almost infinite number of ways that you can capture your scene into a 2D image.

In 3D art, the process of actually creating this image is called rendering. The renderer analyzes the forms in your scene, and decides which ones are visible and which are not and how they appear in perspective from the camera's viewpoint (Figure 1.10). There are different kinds of renderers, and depending on their type, they proceed differently from here. In our case, Blender's renderer begins to go through each pixel of the final image, calculating which form is visible there, and how it will look based on the surfacing properties and lamp and shadow settings that have been chosen. It proceeds through every pixel until the image is complete. If the scene contains simple forms, basic surfacing, and only a few lamps, this can be accomplished in as little as a few seconds. More complexity costs more time. Scenes that contain large numbers of polygons, math-intensive surfacing like true refraction, and certain kinds of diffuse shadows can take hours to render.

This raw render output can be put through Blender's compositor. Even if you are not a photo editing professional, you have almost certainly put your home photography through some kind of management or enhancement software like Adobe's Photoshop Elements or Google's Picasa. These programs allow you to enhance your raw photography: fixing exposure problems, clearing red eye, or adding special effects like a soft glow, directional blurring, or sepia toning. Blender's compositor is a kind of integrated after-processor for your rendered images.

As you can see from Figure 1.11, the compositing process can add a great degree of believability to an otherwise pedestrian render.

And so, after creating your scene, then rendering and processing it, you arrive at a final 2D image that hopefully expresses what you had been thinking in the first place.

But, how do you get to this final image? Anyone firing up Blender for the first time will find a confusing array of buttons, screens, and controls. Before we get into actually doing anything, it pays to spend a few minutes to familiarize yourself with Blender's interface, both the widgets that make up the control system and the thinking behind it. In Chapter 2, we'll look at how to find your way around in this innovative interface.

Next Up ...

In Chapter 2, we actually dig into Blender's interface, going over the different widgets, screen elements, and the thinking behind the whole thing.

(Continues...)

---

Excerpted from Blender Foundations by Roland Hess Copyright © 2010 by Elsevier, Inc. . Excerpted by permission of Focal Press. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---

Table of Contents

Chapter 1: Intro to 3D
Chapter 2: Getting the Blender Interface
Chapter 3: Movement, Object and Animation Basics
Chapter 4: Modeling Basics
Chapter 5: Character Modeling
Chapter 6: Rigging
Chapter 7: Sculpting
Chapter 8: Lighting
Chapter 9: Surfacing
Chapter 10: Rendering
Chapter 11: Character Animation
Chapter 12: Morphing Animation
Chapter 13: Particles
Chapter 14: Effects and environmental animation
Chapter 15: Video editing