Stumbling Through the Jargon...Part Two

After last months introduction to the jargon involved in desk top publishing, this month I hope to start to demystify it. ESP's, Tiff's, bitmaps, JPEG's, resolutions... the list goes on, so where to start? Well, a good way to start would be to explain the differences between the two main types of computer graphics; vector images and raster images, realizing the differences between the two is extremely useful when learning DTP.

Vector images, such as those created in 'drawing' packages such as Freehand, Illustrator or CorelDraw, are made up from mathematically described lines and curves called vectors. As an example, in a vector based drawing package, if you draw a red rectangle, the software simply remembers the colour and the co-ordinates of the rectangle. To draw the shape the computer stores the points that join up to form the shape. If you think of the top left corner of a page as co-ordinates 0.0 (0 along, 0 down), it would remember something like: move to co-ordinate 10.10, draw to 100.10, draw to 100.60, draw to 10.60, draw to 10.10, then fill with red.

The great advantage with vector based images is, that as they are based around co-ordinates, they are resolution independent. This means they will always print at the maximum resolution of the printing device, and that they can be move, sized, enlarged or shrunk without any loss of quality. Create a vector drawing of a logo, and the same file could be used for a letterhead then scaled up to full billboard size without any loss in quality, the file size will remain the same to. Also, whatever you print to, the image will always output at the maximum resolution available, so with a laser printer the image may be 300 dots per inch, on an imagesetter preparing artwork for publication, the same file would output at, say, 2540 dots per inch. Typically, vector images include text, logos, clip-art, and bold graphics with clearly defined lines, no matter what size they are scaled to.

Raster images, such as those created in 'painting' programs such as Photoshop or Picture Publisher, or those that are created when scanning photographs, are based around a grid of small squares, known as pixels or dots.

As an example, the red rectangle described earlier, if created as an raster image, would be made up by colouring each pixel required to make up the rectangular shape. Imagine a piece of graph paper, each of the smallest squares would represent the pixels, to create the rectangle, you would have to colour each pixel, so that they join up to from the final shape. However, when you create the image, you have to set the number of cells or pixels in the grid, this becomes the resolution of the image, measured in dots or pixels per inch. If you increase the size of the image, the pixels are simply enlarged, and if enlarged too much, the individual pixels will become visible, causing a jagged or stepped appearance. As a comparison, if you enlarged the rectangle drawn on graph paper with a photocopier, the rectangle would increase in size, but each of the cells would increase in size too, creating jagged edges. Raster images are best for photograph images, painting type images, and for photo-realistic images.

Because raster images are resolution dependent, that is, once you set the resolution, they cannot print at a higher resolution, no matter what they are finally printed on, if you created an image that prints well on a 300 dot per inch laser printer, it would print, but could look jagged on a 2540 dot per inch imagesetter. If the same image is to be used at a variety of different sizes, each size would have to be a different version of the file, as scaling the file to far up or down, would result in a loss of quality.

Both vector and raster images are each suited to different image types and generally you don't have to make the decision of which to use, as the choice is often decided by the image required. Scanned photographs, which are then edited and manipulated would be raster, whereas text, coloured boxes, logos would be vector. To roughly compare to traditional techniques, raster files would represent photographs, airbrush work and painting techniques, vector files would represent pen and ink drawing, technical drawing, and line and fill techniques.

As mentioned earlier, which type of image to use is often decided by the kind of image you require. Imagine a magazine cover, any photographs or painted type images would be raster and could be edited or created in Photoshop. Any logos or bold graphics with curves and angles would suit vector, and could be drawn in CorelDraw, any text and headings would be vector too, and all the elements would be finally combined together in a 'page layout' program such as Quark Xpress, PageMaker or MS Publisher. Sometimes, logos may be created in Photoshop, as raster images, if they require a 'photographic look' and photographs can be converted to vector images, using tracing software such as CorelTrace or Adobe Streamline. These allow raster images to be automatically traced around, then edited to suit, this can result in a hand drawn or silhouette look, but in vector format, and it is easier to trace the image rather than having to draw it from scratch.

Once the difference between the two graphic types is realised, any further jargon falls into place. Resolution, colour models, halftoning, and the like are more easily understood and next month it'll be time to move on to them.

If you have any questions or queries regarding DTP, please feel free to email me, and until next month happy DTP'ing.--Andy


Home Andy's Columns

Copyright © 1995-1998, SVCG. All rights reserved.