Scintography Home
Previous ChapterTable of ContentsNext Chapter
Title Page

Chapter 4: Page Graphics


The Art of Scintography
Edition 3.0 © 2008 Aurora Isaac
www.scintography.com
    4.1   Web Page Graphics
      One of my favorite things about publishing on the Internet is color. Your page is your canvas, your graphic user interface to the world. It is your opportunity to be boldly creative without worrying about the "cost" of color. On the other hand, if you are coordinating your page design with the design of printed materials, this may be a major concern.
      As you envision your site, keep in mind that your colors and images must do two things for you:
  • Represent you
  • Convey your message
Before you design the graphic elements of your page, get to know the mechanics of graphic presentation in this new medium.
 
    4.2   Digital Colors
      As you operate in the computer world today, you are dealing with two systems of making color that are in some ways quite similar, and in others quite different.
Pigment colors:      
Light colors:      

      Whether you are a painter mixing a particular shade of house paint, a baker tinting frosting for a birthday cake, or an artist selecting inks for a limited edition print, you are using pigment colors. The base paint binds the pigment, and is usually as close to transparent, or white, as possible. As you add pigment to the base, the color gets more and more intense. With light colors, the "base" is at zero intensity of light, or blackness. As the light glows brighter, the color gets more and more intense.
      Your computer monitor uses light color, while your printer uses pigment color. Because color originates as light color on your computer, printers emulate the light color system, so you don't generally need to think about the difference. However, if precise color is important to your work, you should examine the details of color representation in the varius media you work with.
      If your color monitor is a cathode ray tube (CRT), it has a screen coated with substances that posphoresce when activated by light, similar to the dots on your glow-in-the-dark watch. These dots are called pixels, and are the smallest differentiable area on the screen. The resolution of your screen is expressed in dots per inch (dpi) or pixels per inch (px/in). Typical screen resolution is 70-100 px/in.
      Each pixel is a combination of three primary phosphorescing colors. These are red, green, and blue. Within each pixel, your monitor can create distinct levels of light intensity. The video adaptor card in your computer regulates these intensities. How many levels of intensity your screen can produce sets the limit on how many colors your screen can make. These levels are called color depth, and are measured in bits per pixel. Typical color depths are 8, 16, 24, and 32 bits/pixel.
      Because the color intensities are in discrete intervals, each intensity can be represented by a number. Thus, each color you see on the screen has a numerical, or digital, representation. Color numbers are most often specified in hex (hexadecimal - base 16). In the RGB representation, the numbers specify the intensity level of red, green, and blue sequentially, where 00 is minimum intensity and FF is maximum intensity. In this system, this color     is digitally represented as #DB7093, with a red intensity of DB, a green intensity of 70, and a blue intensity of 93. The RGB color black is #000000, and white is #FFFFFF.
      The color cube is a representation of all the light colors available on a screen. On one corner is the complete absence of intensity of all colors, which we see as black. On the opposite corner is the maximum intensity of all three primary colors, which we see as white.

Click the arrows to see the hidden layers. Mouseover a color to see its digital value.
      The color cube above shows all the colors that are available on a monitor with 8 bits/pixel. Out of the 256 possible combinations, 40 were reserved for system use, leaving 216. These are sometimes referred to as the 216 net-safe, or non-dithering, colors. Most systems used to view Internet pages today have at least 16 bits of color depth. As the color depth increases, so does the number of colors. Looking only at one layer of the color cube, depending on the color depth of your monitor, you may see adjacent colors blend smoothly as the number of bits/pixel increases:

8 bits16 bits32 bits

      Even if you have a lower color density on your monitor, you may notice that the colors in the higher density represented above appear to blend. This is because the application attempts to represent the intervening colors through a process called dithering. When you mix paint pigments, red and yellow for example, the pigment particles do not change chemically. The red and yellow particles remain distinct as they are mixed uniformly, but your eye sees orange. Dithering is a similar process of mixing different colored pixels within an area of color to make it appear to be the desired color.

 
Color Palette
 
Color Selector
name  R G B
white ffffff
rosybrown ff6347
thistle 32cd32
skyblue 6a5acd
moccasin 5f9ea0
      Colors can be specified by name or by number. The names are actually nicknames, or aliases, for the color numbers. The names in the Color Palette are a list known as X11, and are implemented by both Microsoft and Netscape. Use the Color Palette and Color Selector tools to explore the range of colors and their numbers.
    4.3   Digital Images
mages play a variety of roles in page presentation. They can be the focus of the entire page or barely visible. They can be large enough to overflow a screen or as small as a single pixel. Some images contain essential information, others are part of the structure or decor of the page.
      A digital image is an ordered collection of pixels and their associated colors. As you work with digital images, you need to consider the following:
  • The image size: its overall width and height in pixels
  • The image density: the number of pixels per inch
  • The color palette: how many and what colors
  • The file format: how to store the image
On the screen, the size and density of an image are interrelated, and the file format you use will depend in part on the palette.
      Image size is generally indicated in pixels. The graphic I at the top of this section is 19 × 41 pixels. This image is 9 × 9 pixels.
      Two densities are associated with images on a page. The intrinsic image density may be expressed in units such as px/inch (pixels per inch), or dpi (dots per inch). Intrinsic density, combined with intrinsic image size, determine the size of the image in pixels. A 1 inch image, with a density of 100 px/inch, will be 100 px wide. This density has no effect on the display density of the image. Your screen has a set density. As you resize a window, the width and height may change, but the density does not. The same is true for images displayed on the screen. They may be resized, but once they are displayed, they appear in the same screen density as everything else on the page.

× 1

× 2

× 1

× ½

× 1

× ½
Image Size

      It is best to create an image in the size it will be displayed. This image of a man rowing was created to be 50px wide. When it is displayed in a larger size, the low density becomes apparent, and the image is degraded. Making images smaller does not always work, either. The man-rowing image still looks good when it is reduced, but the geometric pattern framing the sprite image is dramatically altered in appearance.
      Images for the Internet are generally made at low density to reduce the file size, balancing the time it takes to load a page against the integrity of the images displayed. If you plan to use an image offline, you may want to use higher densities, and reduce the size of the image for screen display. Typical screen densities are 72 and 96 px/inch. Densities for quality printed material, on the other hand, are typically 300-600 dpi. Such high density images can be very large, and significantly affect the time it takes a page to load, regardless of the size the image is displayed. If the only purpose of the large image is to make it available for download, make a low density, icon-sized copy for display, and make the large image available by ftp.
JPG

42700
bytes
JPG

2145
bytes
File Size
      An image format is a specification for how a file of image data is to be interpreted. How this data is stored depends on how the format specification is implemented. Thus, different software programs may use different methods of making the same type of image file. These two man-rowing images are both .jpg, both created from the same source image, both the same size and density, both made by software from the same company. And yet, one file is about 20 times as large as the other. The difference is that one was made in 1998, the other in 2002 -- a newer version of software and a better compression algorithm.
GIF

2371
bytes
JPG

1863
bytes
       
Image Format
      Because of their ability to compress image information into relatively small files, JPG and GIF are the image formats used on the web. An image made in each of these formats may be comparable in size, but their differing features make gif better for some uses, and jpg better for others.
      The Graphics Interchange Format (GIF) can be made transparent, interlaced, or animated. An animated gif is a collection of gif images and a timing sequence presented as a sort of self-contained mini-video. An interlaced gif loads in coarse detail first, gradually becoming finer until the entire image is loaded. This type of gif is nice for large images loading onto slow machines, because the viewer can see that something is happening before the entire file is transmitted. Areas of a transparent gif can be made transparent, allowing the background to show through.
JPG

2004
bytes
JPG

2157
bytes
 
GIF

107
bytes
 
GIF

171
bytes
Image Colors
      The Joint Photographic Experts Group (JPEG) was the committee that wrote the original .jpeg (or .jpg) image compression specification. This format was designed to compress picture images for rapid transmission, which it does very well.
      Not surprisingly, increasing the number of colors adds size to the file. This is true for both gif and jpg formats. However, one of the primary differences between these two formats is in the way they handle color. The great disadvantage of the gif format is that it is limited to 256 colors. Which colors are selected depends on the image. For some images this is a distinct disadvantage. Although the jpg format does not limit the number of colors, it does sacrifice color integrity to provide its ample compression.
      The sprite image below is an example of a jpg-friendly image. This image is a picture with at least 256 distinct colors. Although it would be nice to make the surrounding background transparent, the irregular edges would look shabby against some background colors. It would be better to make this a jpg image with an image background that is the same color as the page background. If you must have a transparent background, use an image background color close to the planned page background color, then make the image background transparent. This will minimize the halo effect at the edges of the oval where the background has blended with the image. This gif was made with a white background made transparent.

gif
3288
bytes
jpg
1826
bytes
       
  JPG-Friendly Image

Mouseover the colors to change the background.
Mouseover the images to zoom.

gif
124
bytes
jpg
1372
bytes
       
  GIF-Friendly Image

      The arrow image above is an example of a gif-friendly image. This image has one color, and was edited pixel-by-pixel to be transparent in some areas, making nicely geometric edges. By mousing over the color swatches, you can see that this gif image looks good against any color background. The jpg image would have to be recreated for every background color it is used with. By mousing over the images to zoom in, you can see that the jpg image color, both the brown and the white, are mottled. The same effect probably occurred in the sprite image, but is undetectable. This loss of color integrity does not happen with the gif arrow image. If these weren't reasons enough, the file size makes it clear that this image should be used in the gif format.
      Because of their drawbacks, the gif and jpg format should be used to transmit and display images, not to store original work for future editing. When the integrity of an output image is altered from the input image, successive edits will cause a rapid degradation of the image.
    4.4   References
1 TheCounter.com
http://www.thecounter.com/stats/

2 : Color Depth Stats
http://www.thecounter.com/stats/2008/February/colors.php

3 The Free Online Dictionary of Computing
http://foldoc.org/

4 : Graphics Interchange Format (GIF)
http://foldoc.org/?GIF

5 : Joint Photographic Experts Group (JPEG)
http://foldoc.org/?Joint+Photographic+Experts+Group

6 Microsoft Developers Network Library
http://msdn2.microsoft.com/en-us/library/default.aspx

7 : Color Scheme Test Page
http://msdn2.microsoft.com/en-us/library/bb263947.aspx

8 Pantone: Color Think Tank
http://www.pantone.com/pages/pantone/pantone.aspx?ca=29

9 The Usenet FAQ Archives: .jpeg
http://www.faqs.org/faqs/jpeg-faq/

10 WebExhibits
http://webexhibits.org/

11 : Color Vision and Art
http://webexhibits.org/colorart/color.html

12 : Pigments Through the Ages: Indigo
http://webexhibits.org/pigments/indiv/overview/indigo.html

13 YALE Web Style Guide
http://www.webstyleguide.com/

14 : Color displays
http://www.webstyleguide.com/graphics/displays.html

All Contents © 2008 Aurora Isaac. All Rights Reserved. Legal Notices.
Previous ChapterTable of ContentsNext Chapter