Every once in a while an idea will just “pop” into my head. Most of the time, these ideas are either impractical by modern technology or would take more time to implement than I have to give. But a few months ago, one of those ideas not only would take little time on my part, and is easily done with even ten year old technology, but would (I think) be of great benefit to Macintosh users everywhere.
The idea? To publisher previews (entire chapters or just parts) of Macintosh books you can buy today, or a “First Look” of soon to be released books. As you may well know, there are a lot of Macintosh related books out there to choose from, on just about every subject you can think of. And while I do believe that reviews, such as our own authority on Macintosh books John Nemerovski’s Book Bytes, are an invaluable resource, I do think actually being able to preview a book is important as well. Like hearing a song on the radio before buying the entire album.
Looking around what others affectionately call the “Mac Web” I was surprised to find no one is doing this. A few book publishers do so, but the chances of a potential book buyer actually finding said book on said publishers web site is hard to imagine. Why not have those Mac book previews on a Macintosh related web site? Sounded like a no-brainer to me.
I am happy to say that after talking to some great people over at Peach Pit Press, My Mac is doing just that starting right now.
The first book we are presenting here at mymac.com as a preview is “Real World Adobe PhotoShop 6” by David Blatner and Bruce Fraser. I personally read most of this book over the weekend of April 7-8th, and was so impressed, I could not wait to showcase this book. (I should point out, however, that book previews and book reviews here at mymac.com have no bearing on each other. Nor are we being paid to showcase any particular book.)
I do want to thank three special people who helped get the ball rolling on this new feature. They are (in no particular order) our own John Nemerovski who knows everyone in the Mac book-publishing world. Gary-Paul Prince and Kim Lombardi at Peach Pit Press, the publisher of the first previewed book, and two people who take chances and reap rewards for both their open mindedness and creativity. And Adam Karneboge, our intrepid Web Master, for his tireless posting on behalf of the entire staff.
I hope to continue presenting previews of many more Mac books in the years to come, not only from Peach Pit Press, but also from other book publishers as well. There are many Macintosh related books out there, with many more to come with the release of Mac OS X, and all the new software/hardware which is sure to follow. These writers and publishers serve the Macintosh community in a way that they do not get enough praise for. It is my hope that by presenting these previews and special first-looks, we are serving the Macintosh community, as well as helping the authors and publishers of the Mac world.
If you know of a book you would like to see previewed, please contact us using our nifty “Feedback” form on the right side of every page of our web site, and we will gladly see about getting a preview of said book.
Real World Adobe Photoshop 6
by David Blatner and Bruce Fraser
Publisher: Peach Pit Press
Chapter 3: Pages 73-82
(The following page(s) presented here are a preview of Real World Adobe Photoshop 6, written by David Blatner and Bruce Fraser and published by Peach Pit Press. All Rights Reserved. The work presented here is with full authority and permission of Peach Pit Press, and cannot be copied, duplicated, or sold. MyMac.Com is not the originator or owner of the following content.
It’s All Zeros and Ones
Let’s get one thing perfectly clear: this book is not about pictures or work flow or even computers. This book is about zeros and ones. As Laurie Anderson so plainly pointed out, no one wants to be a zero and everyone (at least in America) wants to be “number one.” The digital age is built entirely on the interplay between the two.
To be sure, the digital world (in which zeros and ones, offs and ons, and whites and blacks frolic together in cooperation, not competition) is not as confusing as some people make it out to be. And, as it turns out, you can’t really be efficient with digital imaging without knowing a little bit about that dark underworld. In this chapter we’re going to break it all down for you. To some of you, most of this chapter is going to soundpretty basic, but we urge you to peruse it anyhow. You might be surprised at how many “power users” find themselves stumped by something as small as a misunderstanding of how–and why–bitmapped images work.
Bitmapped versus Vector Graphics
In all the grand canon of computer imaging, there are really only two kinds of graphics: bitmapped and vector.
Bitmapped images. Bitmapped images are simply collections of dots (we call them pixels or sample points) laid out in a big grid. The pixels can be different colors, and the number of dots can vary. No matter what the picture is–whether it’s a modernist painting of a giraffe or a photograph of your mother–it’s always described using lots of dots. This is the only way to represent the fine detail and subtle gradations of photorealistic images.
Just about every bitmapped image comes from one of three sources: capture devices (such as scanners, video cameras, or digital cameras), painting and image-editing programs (such as Photoshop), and screen-capture programs (like Exposure Pro, the System, and a host of others). If you create a graphic with any of these tools, it’s a bitmapped image.
Vector graphics. Vector graphics, often called object-oriented graphics, are both more complex and simpler than bitmapped images. On the one hand, instead of describing a rectangle with thousands (or millions) of dots, vector graphics just say, “Draw a rectangle this big and put it here.” Clearly, this is a much more efficient and space-saving method for describing some images. However, vector graphics can include many different types of objects–lines, boxes, circles, curves, polygons, and text blocks, and all those items can have a variety of attributes–line weight, type formatting, fill color, graduated fills, and so on.
To use an analogy, vector graphics are like directions saying, “Go three blocks down the street, turn left at the 7-11, and go another five blocks,” while bitmapped images are more like saying, “Take a step. Now take another step. And another … .” Photoshop has traditionally worked with bitmapped images, but in Photoshop 6 you can create a variety of vector graphics that retain their object-oriented characteristics, or use them as selections and masks on bitmapped images.
Outside Photoshop, vector graphics come from two primary sources: drawing programs such as FreeHand, Canvas, Illustrator, and so on, and computer-aided design (CAD) programs. You might also get vector graphics from other programs, such as a program that makes graphs.
Bitmaps as objects. The distinction between bitmapped and vector graphics is slightly fuzzy, because vector graphics can include bitmaps as objects in their own right. For instance, you can put a scanned image into an Adobe Illustrator illustration. The scan actually acts like an object on the page, much like a rectangle or oval. If you include a bitmap as an object in an illustration, you can rotate it and scale it, but you can’t go into the image and change the pixels.
Note that a vector graphic file might include a bitmap as its only object. In this situation, the file is a bitmapped image that you can open for editing in a painting or image-processing application. Photoshop’s EPS (Encapsulated PostScript) files are good examples of this. While EPS is typically a vector file format, you can create a bitmap-only EPS in Photoshop.
Vectors in bitmapped graphics. Just to round out the confusion, Photoshop lets you include vector graphics in bitmapped images, either as standalone objects (like text) or as clipping paths. A clipping path in an image is invisible; it simply acts as a cookie cutter, allowing you to produce irregularly shaped images such as the silhouetted product shots you often see in ads (see “Clipping Paths” in Chapter 16, Storing Images).
Photoshop lets you open, create, edit, and save bitmapped images. Bitmapped images are its sine qua non, its raison d’être, its “precious bodily fluid.” So to get the most out of Photoshop, you’ve got to under-stand bitmapped images inside and out.
Every bitmapped graphic has three basic characteristics: dimension, bit depth, and color model (which Photoshop refers to as image mode).
Bitmapped images are always big rectangular grids. Like checkerboards or chessboards or parquet floors in your kitchen, these big grids are made of little squares (see Figure 3-1). The dimensions of the bitmap grid refer to the number of pixels wide and tall it is. A chessboard is always eight squares by eight squares. The grid of pixels that makes up your computer screen might be 640 by 480.
A bitmapped image can be any dimension you like, limited only by the capabilities of your capture device, the amount of storage space you have available, and your patience–the more pixels in the image, the more space it takes up, and the longer it takes to do anything with it.
Note that dimension has nothing to do with physical size in inches or picas. Bitmapped images in their pure digital state have no physical size; they’re just data. They exist as Platonic ideals, waiting to be realized by reproduction in some physical form. No matter how you stretch or shrink a bitmapped image, it still contains the same number of pixels.
When you print a bitmapped image, you print it at a specific size, and the relationship between that size and the number of pixels the image contains is called the resolution of the image. But it’s very important to understand that resolution isn’t innate to the digital image; it’s a rubber measurement that changes depending on the physical size at which you reproduce the image. We’ll discuss resolution and why it’s important in more detail later in this chapter.
Each pixel in a bitmapped image is represented by a particular number of zeros and ones, otherwise known as bit depth (one bit can be either a zero or a one). That number dictates the range of possible values for each pixel, and hence the total number of colors (or shades of gray) that the image can contain. The number of possible values is 2 to the power of the number of bits.
A 1-bit image (one in which each pixel is represented by one bit) can only contain blacks and whites. If you have two bits of information describing a pixel, there are four possible combinations (00, 01, 10, and 11), hence four possible values (22), and four possible colors or gray levels (see Figure 3-2). Eight bits of information give you 256 possible values (28); 24 bits of information result in over 16 million possible colors. (With 24-bit RGB images, each sample actually has three 8-bit values–one each for red, green, and blue; see Figure 3-3.)
We call 1-bit images flat or bilevel bitmaps. A deep bitmap is any image that has more than one bit describing each pixel (see Figure 3-4). Photoshop also allows you to use 16-bit-per-channel images. Sixteen bits of information (216) can describe 65,536 possible values. Forty-eight-bit RGB produces almost 3 billion possible colors–again, each sample is made up of three 16-bit values. These may seem like mind-numbingly large numbers, especially when we consider that though we can display a maximum of 16.7 million possible colors on our monitors, we can see perhaps seven or eight million discrete colors at best, and we can print at most a few tens of thousands of colors on the best printing processes available. Nevertheless, 16-bit channels offer more than massive overkill.
Most of today’s capture devices (scanners and digital cameras) record more than 8 bits of information per channel. A 10-bit capture provides 1024 possible values, a 12-bit capture provides 4096 possible values, anda 14-bit capture produces 16,384 discrete shades. Photoshop treats any-thing more than 8 bits per channel as a 16-bit channel, because it’s much easier to handle bits in groups of 8 (called bytes), and pad out the miss-ing values with zeros, than to create separate routines for handling 10, 12, or 14 bits per channel. We refer to any image that contains more than 8 bits per channel as a “high-bit” image to avoid nitpicking over whether it really contains fourteen bits of information or only ten. In any case, Photoshop treats all high-bit images as 16-bit-per-channel images.
Why capture many more colors than we print, or even see? The simple answer is that the larger number of bits allows us much more editing flex-ibility.
When you start out with only 256 shades per channel, each edit you make has the inevitable result of reducing that number. As you’ll see in “Stretching and Squeezing the Bits” in Chapter 6, Tonal Correction, every edit opens up gaps between some adjacent pixel values, and smooshes others together, reducing the total number of shades.
This discarding of data is a normal and inevitable part of image edit-ing, but we’ve always advocated holding on to as much data as possible for as long as possible. If you start out with a 16-bit-per-channel image, you’ll be able to edit your image with much less risk of losing detail or introducing posterization or banding than you would with an 8-bit-per-channel image. In some cases, you can squeeze some extra editing head-roomfrom an 8-bit image by converting it to 16 bits–it doesn’t increase the amount of information in the image, but it gives the data points more places to land.
Bit depth has an important relationship to the quality of an image, which we’ll cover more fully later in this chapter.
The problem with bit depth is that it doesn’t really tell us (or Photoshop) what each color (numerical value) means. A 1-bit image is easy: each pixel can only be on or off. It doesn’t have to be black or white, though; if you were twisted enough, you could make this orange or blue.
But as we’ve seen, each pixel in an 8-bit image can be described using one of 256 values. Are those 256 levels of gray? Or 256 colors? Or some-thing else? We’ll let you in on a sad, sordid little secret here. Everyone who works with color on computers discovers it eventually anyway: computers don’t understand color at all–they just understand numbers: zeros and ones.
The color model–otherwise known as image mode–is the missing piece of the puzzle, the magic decoder ring that tells how to translate each pixel’s numerical value into a color or a shade of gray. Actually, image mode and color model aren’t exactly the same thing, but they’re so closely related that it makes sense to discuss them as aspects of the same thing.
If the image mode is set to Grayscale under the Mode menu, the value of each pixel is a grayscale value: 0 is black, 255 is white. If the image mode is Indexed Color, then each value is tagged to a specific, arbitrarily chosen color. (An indexed color image can only have 256 different colors in it; see “Indexed Color,” later in this chapter.)
However, if the image mode is set to RGB, Lab, or CMYK, then the color of the pixel is actually made up of multiple 8-bit or 16-bit values. For instance, in a 24-bit RGB image, each pixel is described using three 8-bit values, each of which specifies a level of brightness for the red, green, and blue channels (see Figure 3-3). In a CMYK image, Photoshop looks at and composites four 8-bit images.
You can look at the individual channels and view each one as a grayscale image. The color image is made by colorizing each channel with the appropriate color and stacking them one atop the other.
Note that unless you’re working with esoteric scientific or medical imaging equipment, you won’t have to tell Photoshop which image mode to use: virtually every file format that Photoshop recognizes has the secret decoder ring built in. But you need to understand image modes and their related color models if you want to do much useful work with Photoshop. (For a fuller discussion of color models, see Chapter 4, Color Essentials.)
We’ll look at each mode that Photoshop uses, and why you’d want to use one or another, later in this chapter.
Resolution is one of the most overused and under-understood words in desktop publishing. People use it when talking about scanners and print-ers, images and screens, halftones, and just about anything else they can get their hands on. Then they wonder why they’re confused. Don’t worry; resolution is easy.
As we noted earlier, a bitmapped image in its pure digital state has no physical size–it’s just a bunch of pixels. But you can’t see a pure digital image unless you can decipher zeros and ones in your head. So when-ever you give an image tangible expression, whether it’s as an ephemeral representation on the screen or as a permanent printed form, you con-fer upon it the property of physical size. And with size comes resolution.
The resolution of a bitmapped image is the number of pixels in each unit of measurement. If we’re talking in inches, then we talk about the number of pixels per inch (ppi), which is what most people mean when they say “dots per inch” (dpi).
If your bitmapped image has 72 pixels per inch, and it’s 72 pixels long on each side, then it’s an inch long on each side. If you print it at half the size, you’ll still have the same number of pixels, but they’ll be crammed into half the space, so each inch will contain 144 of them. If we take the same bitmapped image and change it to 36 pixels per inch (changing its resolution), suddenly the image is two inches on each side (same num-ber of pixels, but each one is twice as big as the original; see Figure 3-5).
You can also look at bitmap resolution in another way: if you know the size of an image and its resolution, you can figure out its dimensions. When you scan a picture that is three inches on each side at 100 pixels per inch, you know that the bitmapped image has 300 pixels on each side (100 per inch). If you scan it at 300 pixels per inch, the dimensions shoot up to 900 pixels on each side.
The key to making resolution work for you (rather than against you) isin knowing how many pixels you need for the intended purpose to which you’ll put the image. We discuss how much data you need for different purposes in the next section.
Real World Adobe Photoshop 6 by David Blatner and Bruce Fraser ©2001 published by Peachpit Press. For a list of bookstores who carry Peachpit Press books call (800) 283-9444 or visit http://www.peachpit.com/special/bookstores.html