World Magazine, Issue 2.11, February 1998
A Good Place to Start
by Bill Fleming
What is digital cinematography? Well, it's the art of storytelling on the computer. In today's terms it typically means 3-D animation as described in Ben de Leeuw's Digital Cinematography. I'm pleased finally to see a book that covers this crucial topic. You'll find that my review is rather direct and critical since I like to see 3-D books that go beyond superficial explanation and actually explore the deep roots of 3-D animation. Of course, I'm usually disappointed, but in the case of Digital Cinematography I was pleasantly surprised. With the exception of the occasional light coverage on topics, I found the book to be a very informative resource.
To keep this review from becoming a book itself, I decided to provide you with a `First Impression' overview and then cover the most notable chapters in the book. While every chapter provided useful information, there were several that were especially valuable to digital cinematographers. So, what are we waiting for? Let's take a look at Digital Cinematography.
My first impression of the book was one of disappointment, not due to the content, but rather the visuals. To be specific, the example images were less than motivational. This is a problem with nearly every 3-D book. It seems most authors focus their energy on the text, but treat the images as an afterthought. In a world of imagery such as 3-D graphics, it's difficult to take their discussions to heart when the images are disappointing. If we are to be convinced that the information they are sharing is accurate and useful we need to see practical, real-world examples that support their claims. In the case of the images in Digital Cinematography, well, I don't think anyone will ever have the need to render a stickman worshipping a temple that looks like a teapot. At least, I certainly hope they won't. The images in the book are barren and very unrealistic which tends to make light of the subject matter. It's my sincere hope that 3-D authors will soon realize the importance of providing visually appealing and inspirational images, which will make their information more credible. As viewers, we tend to assume that the author is utilizing the information and techniques described in the book to create the images. Therefore, if the images are poor, we must assume the information is also poor. This may not always be the case, but we will never know since we didn't read the book due to the poor quality of the images!
Okay, now that I have that off my chest, let's take a look at the content of the book. While the book does make several good points, it also appears to be one-sided. For example, it states that having characters recognize the camera's presence by looking at it is undesirable, but there are many cases where having the characters look at the camera is a powerful tool in drawing the viewer into the story. For instance, the movie Cuffs where Christian Slater talks to the viewer directly helps to pull the viewer in. It would have been nice for the book to share more than one view on the topic.
You'll also find that the book seems to focus on more classic cinematography rather than modern techniques. If you are interested in the classic methods it's a nice book, but if you lean towards the modern techniques you'll find yourself wanting more. However, this doesn't mean that the book isn't valuable. I just prefer to see a book that covers all possibilities, rather than the author's personal taste, even if it happens to be good taste as is the case with Digital Cinematography. Now, here's a couple of the most useful chapters.
Chapter 1: Digital Cinematography
This chapter had a fairly good overview of film language and work principles though I felt myself wanting more. It seems that several ideas are touched upon lightly, peaking our interest, but never satisfying it with further elaboration. For example: When using movies as a reference source, it suggests that you watch the movie several times to grasp the film's approach. This makes sense because you typically get caught up in the story the first couple of times you view the movie, but you actually need to view the movie on videotape to be critical. The `theater experience' is too captivating with the big screen and THX Sound. It's almost impossible to be objective, particularly if you enjoy the film. You need to critique the film's approach in an environment that gives you control, such as on video. It's more difficult to be pulled into the story on a small television screen without the thunderous theater sound and giant picture. Besides, you can pause the movie for those all important bathroom and snack breaks.
Chapter 2: Introduction to Lighting
Here you'll find an excellent overview of the different light types and uses. To my knowledge this is the only 3-D book that has actually related 3-D lighting to its real-world counterparts, which was a refreshing change. You'll also find some very useful tips in this chapter. A particularly good tip was to identify the influence of individual lights by turning off all the other lights in the scene and creating a test render to see the influence of the light. Another great tip was the use of a backlight to make the character stand out from the background, which is something cinematographers and photographers have been doing for years, but 3-D artists rarely do.
While there was a great deal of good information in the chapter, I did notice the absence of a discussion on specularity. The chapter failed to mention that specularity must be disabled on the fill and backlights to prevent multiple `hot spots' from appearing on the objects in the scene. Since fill and backlights are used to simulate radiosity, they are considered indirect lighting. Since specularity is only created by direct lights, you'll want to disable specularity on your fill and backlights. If your indirect lights are creating specularity, your scene will appear very artificial. In film production and professional photography, reflectors are used to bounce the light from the fill and backlights to diffuse the specularity. Naturally, some 3-D programs don't allow you to control the specularity values of the lights. In these cases, you'll need to use several lights with very low brightness values to simulate the effect of a singular fill or back light. Using lights with low brightness values will diffuse the specularity and by using multiple lights you'll be able to achieve the desired brightness.
Chapter 5: Motion
I must admit I was very impressed by the attention to detail in this chapter. I found it to be very insightful with some excellent real-world references, which helped drive the concept home. If you are interested in perfecting your camera movement this chapter is a must. While the visuals were less than stimulating, the content was excellent. The most important tip in the chapter is that you need to conform to the viewer's expectations when adding camera movements to your animations. I can't tell you how refreshing it is for someone to point out this critical fact. Nothing will undermine your animation like improper camera movement.
Chapter 6: Working with Characters
I was pleased to see a chapter dedicated to character lighting since most 3-D animations do a very poor job of lighting characters, which then undermines the impact of the animation. I can't count the number of times I've seen characters that were poorly lit or had no dedicated lighting at all. This causes the character to blend in with the background, which neutralizes their impact in the scene.
While the chapter content was informative, I was annoyed at the superficial discussion of the specific light types. For example, the coverage of the use of eye lighting was great, but there was no explanation on how to properly stage the lighting. It mentions that you need to keep the light from being noticeable but then offers no explanation on how to do it. When I purchase a 3-D book I'm hoping it will show me how to save time by showing solutions to problems. While introducing the technique is paramount, it is only half of the equation. I need to be shown how to apply the techniques so I don't waste a lot of time experimenting. For example, it would have been good to point out that the character lights should have the shadow and specularity attributes disabled. Otherwise, the character will have lighting that conflicts with the scene's lighting.
Chapter 7: Exterior Lighting
This discussion on lighting was relatively detailed. It was nice to see the different lighting scenarios such as daylight, nighttime and water lighting discussed individually. I did find that the detail of the discussion was rather superficial, again leaving me wanting more. To be exact, the discussion on nighttime lighting failed to identify the fact that moonlight is heavily diffused. Since it is actually light reflected from the sun, it naturally will be diffused. It's basically a large indirect light source. I also noticed that the discussion on outer space lighting was correct in its assumptions that the light is predominantly direct from the sun, but it failed to address the indirect light that comes from planetary bodies, such as planets and asteroids. If you are rendering a satellite orbiting earth, you need to backlight the satellite to account for the light that is reflected off the earth. The same would apply if the moon was in the scene.
The discussion on water light was very well done. I particularly liked the segment that addressed light that bounces off the surface of the water, and described how it illuminates the environment with organic light patterns caused by the light reflecting off the ripples in the water. I've seen many under-water images that incorporate this effect, but I have rarely seen it used on surface images.
In general I'd say it was a very well presented chapter, missing only a few important details.
Chapter 9: Color
I was pleased to discover a discussion on the color of natural light at different times of the day, but I was greatly disturbed to see that the light colors mentioned were inaccurate. It states that high noon light is white or yellow, when it is actually blue and that firelight is yellow and orange when it's actually red. The color of light is determined by its temperature, which is measured in degrees Kelvin. Basically, cool lights, like candles, are red, and warm lights, like sunlight, are blue. Photographers are very familiar with the color of light since they need to use special filters like polarizers to remove the natural blue color of daylight.
Now, we don't see the actual color of light because our eyes automatically convert the light to white through a process known as chromatic adaptation. This process isn't perfect, so some light coloration will be visible but it's actually only around 5%. The result is that objects will take on a different color tint at different times of the day. Try this test. Step outside at noon and take a look at hills or mountains in the distance. You'll notice that they have a bluish tint. It's much easier to see the impact of light colors the further objects are away. You should definitely do a little exploration on your own to discover the actual colors of light.
I also noticed an omission from the chapter, which was a discussion on the color of indirect lights. You see, light is emitted from the source at a particular color, but when it bounces off objects it takes on the color of that object. For example: An incandescent light in a house lamp leaves the bulb pinkish orange but is changed to beige when it bounces off the off-white walls of the room. A discussion dedicated to the properties of indirect lighting would have made the chapter more complete.
In general, I'd say the chapter was very useful but had a few mistakes in the colors of light and definitely could have explored the issues of light color in more detail.
Well, I'm sure it seems that I have hammered the book but that's only because I want to see the quality of 3-D books improve. I figure if I'm to invest my hard-earned money on a book, I'd like to get the biggest bang for my buck. While the book was light on exploration and had some errors, it was generally full of valuable information. Since I like to quantify things I've rated the book based on specific areas which are listed below. Each area is rated in a value from 1-10, with ten being highest, then I've provided an overall score. This rating will make it easier for you to determine the value of each area.
How it Rates
First Impression: 5
The images were less than compelling and the examples seemed simplistic.
There was an abundance of valuable information but I was left wanting more. There were also a few errors in the information provided.
Image Quality: 3
Well, the images could have used a lot of work.
It was pleasant to read and it kept me interested. The information was presented in a very logical format.
Overall Rating: 6
I would definitely recommend Digital Cinematography to any 3-D animator who is interested in creating quality computer animations. It's a particularly useful book for those who are getting started as 3-D animators but if you are a serious animator, you'll probably want something more advanced.
Digital Cinematography by Ben de Leeuw, AP Professional/Academic Press, 1997. 265 pages, illustrated. ISBN: 0-12-208875-1.
Bill Fleming is President of Komodo Studio, one of the world's leading Photorealistic 3-D studios that specializes in creatures and characters for broadcast and film. He is recognized as a leading authority on Photorealistic 3-D graphics and creature/character creation and is the author of the 3-D Photorealism Toolkit published by John Wiley and Sons, Inc.
Note: Readers may contact any Animation World Magazine contributor by sending an email to firstname.lastname@example.org.
Table of Contents
[about | help | home | email@example.com | mail | register]
© 1998 Animation World Network