There are many different file types. Each has its specific justifications. We encourage you to familiarize yourself with the various file types your camera or scanner offers so you can choose the optimum one for the application you have in mind. There are three broad categories of file types:

• Compressed files: These are files in which various techniques are used to make the image file smaller and easier to handle. There are two fundamentally different types of compressed files:
   
  • Lossless compression is a type of compression where all the image information is retained but the information is rearranged. The easiest way to try and understand what is happening is to look at a line made up of 100 blue pixels: An uncompressed file would record that information as follows: "Pixel1: Blue; Pixel2: Blue;...; Pixel100: Blue." The same file compressed with a method called Run Length Encoding (RLE) would state the same as follows: "Blue from Pixel1 to Pixel100." It is easy to see that if dealing with millions of pixels, the second way of arranging the information is much more efficient. Every type of file can be compressed that way either with a separate program or as part of the file definition itself. The most common file type of this kind is the Portable Network Graphics (PNG) file format.
    
    

  • Lossy compression is a type of compression where a certain amount of information is discarded. Imagine a bright blue sky. There are going to be subtle shade differences, but chances are that adjacent pixels are of a very similar color. It might not harm the visual impression at all to simply say that all pixels within a certain shade range are considered equal. In this case, instead of defining the color for each pixel, the color for a range of pixels is considered the same and needs to be defined only once for each group of pixels. It is obvious that the file size can be dramatically reduced that way. The most popular file type of this kind is the JPEG file format that virtually every camera uses.

    Two very important points need to be kept in mind when dealing with JPEG files: Whenever a document is being saved as a JPEG file, the compression algorithm is applied. If a lot of editing is done with an image, the numerous save operations can lead to noticeable quality degradation. It is therefore better to start with an uncompressed file like a TIFF and only do the final save as a JPEG to save space. If a TIFF is not available as a starting point, it is advisable to open the JPEG on your computer, save it as a TIFF, do the editing and only at the very end save again as a JPEG.

    The second point is one that is often overlooked. Every major program offers a selection for the JPEG compression factor in the save dialog box. These settings directly determine what color range is to be considered equal. Experiment with those settings and find one that gives you the right balance between space savings and quality loss.
    
    

• Uncompressed files: These are files that contain the recorded image information, pixel for pixel. These files can be large but at the same time offer the best possible quality. Typical file types include TIFF, GIF, and BMP files. If raw files (see below) are not available or if the extra step of a raw conversion is not desirable, these types of files are the best starting point for manipulating your image.
   
• Raw files: These are files as your camera produces them internally. Not all cameras allow you to save your photos as raw files. If you can do it, however, these files allow you the most opportunities to be creative in a true "digital darkroom."
  

[Top]

It is important to realize that there is a difference between a "dot" and "pixel." Pixel is short for "picture element" and represents the smallest unit that carries image information, e.g. a pixel in the sky is blue and is made up of many dots that all together generate the associated blueness of the pixel. Printer manufacturers usually specify the printer resolution in dpi, or dots per inch. A high dpi number does not mean that the printer is capable of printing that many pixels per inch; it rather indicates how many dots can be used to make up one pixel. The more dots per pixel, the better the printer's capability to reproduce smooth shades of similar colors. Photo printers typically combine the available dots to print around roughly 300 pixels per inch (ppi) depending on the specific machine.

300 ppi is a good starting point to assess the right file size: If you want to make a 4 x 6 print you need (4 x 300) x (6 x 300) = 2.16 megapixels (MP); for an 8.5 x 11 print you would need 8.4 megapixels for full coverage. To date, most cameras do not offer image files of that size. However, the algorithm used in cameras, imaging software, and printer drivers is of such a high quality that the differences are often hardly noticeable. On the other hand, if you are sure that you only want to display an image on a computer screen, which is normally a 72–100 ppi device, you only need those 72–100 pixels per inch for your on-screen photo. There is no disadvantage in this case to economize on file size.

There is one factor that is often overlooked: Most cameras allow you to save images as JPEG files. The JPEG file specification calls for a compression factor that allows you to choose between a file that is very small, highly compressed, and rather unsuitable for printing and somewhat larger files with less compression that can be quite hard to tell apart from uncompressed formats under reasonable magnification.

To summarize, we would encourage you to take photos with the highest resolution and largest file size your camera permits. Additional memory for the camera is comparatively cheap; memories are priceless. You can always discard information that you don't need; information that you did not record is lost forever.

As a rough guideline, we would recommend the following file sizes for certain print sizes:
2 x 3: 0.5 MP
4 x 6: 2 MP
8.5 x 11: 4–5 MP and up depending on output device

[Top]

The resolution of a scanner is usually given in dots per inch (dpi), but please keep in mind that the scanner can "see" the color of a dot and thus in reality scans in pixels per inch (ppi). With that in mind, it becomes clear that in order to scan in the most economical way, you need to know something about the output device and the desired output size. A useful example would be the following: If a 4 x 6 photo is to be scanned and reproduced in the same size on a printer capable of 300 ppi, you would select 300 dpi as the scanner resolution and get enough information for each pixel the printer is capable of generating on a 4 x 6 area. If, however, you want to enlarge the image to, say, 8 x 12, you need to scan at twice the linear resolution to provide enough data for each of the printer's pixels, generating a file approximately four times as large. Looking at it from another angle, you decide to scan the 4 x 6 image at 600 ppi thus generating twice as many pixels per inch as before. Since our printer is still only capable of 300 ppi, you now have enough pixels to reproduce two inches per original inch on the printer. The image area will therefore be four times as large.

[Top]

Many modern photo printers allow the use of an optional photo cartridge. We recommend the use of a photo cartridge to print photos whenever possible. There are several reasons for printer manufacturers to introduce photo cartridges, the most important one being the increased color gamut they tend to offer. Color gamut is a technical term for what could be called color range. Think for a moment of boxes of color pencils. Go into an art store and you will find that companies offer sets of 6, 12, 24, or 36 color pencils. It is pretty evident why; it is much easier to get the color you want when you can pick from 36 different shades than if you had only 6. The same is true for photo ink sets. A printer with four primary colors can reproduce fewer shades than one with 6, 7, or 8 different primary inks. This is especially important when dealing with subtle variations of similar colors or photos with a dominant color. If the subject already has a restricted color range, it is even more important to have fine gradations and subtle shades.

[Top]

Changing the focal length of the lens has, besides the effect of being able to capture a narrower or wider field of view, an effect on the depth of field. In other words, while a photo taken with a wide angle will show the subject(s) in focus over a wide depth range, a photo taken with a telephoto lens at the same f-stop will have a much narrower depth of focus. There will also be a distinct difference in perspective and apparent distortion. A digital zoom works in a different way; part of the image is cropped and the result enlarged to the normal image size electronically. There are neither the optical effects of varying focal lengths nor the full quality, since only part of the pixels have been used.

We do not, for this reason, recommend the use of a digital zoom. If cropping and enlarging are necessary, standard imaging programs (like Photoshop^®, Paint Shop Pro, or Photopaint to name but a few) will generally do a much better job than the processing system in the camera.

[Top]

High humidity is the natural enemy of an ink-jet print. Storing prints under very high humidity conditions can lead to ink migration in the sheet, i.e., colored borders or fringes around edges or a general washed-out appearance. We therefore recommend giving your images at least 24 hours to dry out completely before framing or otherwise encapsulating them. A cool, dry environment away from intense heat or sunlight will also contribute to ensuring a maximum lifetime.

[Top]