Do you really need to know any of this stuff to get decent prints from your digital camera?
Not really, but if you ever run into trouble getting the print you want it's nice to know what might have gone wrong and what you can do to get it right.

The very first thing you'll want to do is make sure your camera is saving the highest quality image files it is capable of.
These settings will vary from one camera to the next. They may refer to highest quality, largest file size, or lowest compression. Bigger is definitely better here.
Some cameras today are also capable of saving uncompressed formats such as TIF/TIFF or RAW. If yours is one of these this would be the best option.
Many digital cameras default to a middle image quality setting, so get into the habit of checking these settings every time you recharge/reinstall it's batteries.
While these settings will likely require you to buy extra memory, when that real keeper comes along you'll be glad you have it saved in the highest quality available to you. With prices of camera memory sticks and cards so low I wouldn't even consider using anything less than 512MB and 1GB would be better still. Especially if you are shooting with a 6 or 8 MP digital SLR.

It is much easier to reduce the size of a digital image than it is to enlarge it. This is because there is a fixed number of pixels* available in a digital image. Increasing it's size without adding pixels reduces resolution. Reduce resolution too far and you lose image quality.
In a digital camera this will usually be determined by the number of pixels it's sensor can capture.
Resizing a digital image has no effect on it's pixel dimensions unless you crop or resample it in the process.

A 2400x3000 pixel image will be 8"x10" at a resolution of 300 pixels per inch (ppi). At 16"x20" this image will have a resolution of 150 ppi. It is still 2400x3000 pixels
A good image resolution to send to most inkjet printers is between 150 and 300 ppi. Much higher is a waste of storage space, much lower and image quality may be reduced.

Obviously a 24"x36" image with a resolution of 300 ppi (7200x10800 pixels - nearly 78 megapixels) would be difficult to achieve for most digital cameras by themselves.This is where resampling becomes necessesary.
Resampling is best used sparingly if at all. Resampling or interpolation, adds or subtracts pixels in a digital image file using a mathematical algorithm. When adding pixels the software is essentially making an educated guess as to what the colors of the new pixels should be based on the adjacent existing pixels. Regardless of the algorithm used (some are better than others) this results in errors being introduced. The more it is used the more errors that are introduced resulting in a blurred or distorted image. When reducing image size or downsampling, image quality is not lost, but this is irreversable. It is always advisable to retain the larger original for future use.
To some extent as print size is increased the normal viewing distance increases as well. So that 24"x36" print need not be printed at quite as high a resolution as a 5x7 snapshot to still please the eye. Some great looking very large prints can be made from even 3 or 4 MP digital cameras through balancing the resolution requirements of the printer with the judicious use of resampling.
Terribly disappointing results can occur when an overly compressed image is resampled. When an image is saved in a compressed format such as JPG/JPEG, information is thrown away during compression. When the image is viewed it is then uncompressed using a proprietary algorithm. This can significantly reduce the amount of storage space required and speed the transfer of these files. Lower levels of compression will result in retaining very good image quality while saving a great deal of storage space. If higher levels of compression are used, image quality suffers irreparably. Typical compression artifacts are the result of errors introduced and can look very much like excessive use of resampling. When these errors are interpolated during enlargement they are compounded exponentially.
Excessive compression can also occur if a JPG is repeatedly opened in an image editor and saved since it will be recompressed each time it is saved. This is the reason many photographers will save their images in an uncompressed format such as TIF as soon as they transfer them from their camera. While very large, uncompressed digital image files are ideal for those who wish to edit their images as the information they contain remains stable through repated edit/save operations.
Recompression does not occur when simply opening, copying or moving the file to a new location.

Digital noise is also made much worse by resampling. Digital cameras have problems with extreme lighting conditions. Much more so than their film counterparts due to limitations of the sensor. When enlarging a noisy image using resampling the noise is enlarged and interpolated as well. What may not have even been visible on a monitor or when printed at 5x7 becomes splotchy skies and speckled shadows in a larger print. Some digital noise is inevitable, but it can be reduced by using manual exposure settings and selective editing. Noise reduction software can be helpful in some circumstances, but any overall noise reduction will soften or blurr the detail of the image..

Avoid resampling your images when sizing them for printing whenever possible. You do this by starting with the largest image file your camera can produce. Resampling is a necessary evil when making very large prints, but even here it should be used judiciously. It is often better to send a lower resolution file to the printer than to excessively resample the image.

Avoid excessively cropping your images since you are reducing the available image size. Learn to "Crop" the shot with your camera rather than relying on an image editor whenever possible.

Be aware of the resolution of your image at your chosen print size. Too low and it may produce a poor quality print. Too high and the file size may make it difficult to work with. Sending very high resolutions to some printers will actually reduce image quality, however most inkjets will simply discard any excess in it's own software before printing begins.
Pay no attention to the marketing hype regarding how many thousands of dots per inch* (dpi) your inkjet printer claims to offer. This has nothing to do with the resolution of the file you can send to it and very little to do with what it actually puts on paper.
Upsampling an image to 1200 or 2400 ppi is counterproductive in the extreme. Even if you could get it to look good with all those software created pixels, it will just take longer to transfer and print.

Avoid excessively compressing your digital image files. A good practice is to save your originals in an uncompressed format, always saving any changes to a new file to preserve the original. When saving a JPG using an image editor always use best quality/lowest compression level/largest file size unless you have a specific reason to do otherwise.

Avoid using digital zoom. Digital cameras generally offer both optical zoom (produced by the lens) and digital zoom (produced by software within the camera). Digital zoom can cause distortions which will become visible if the image is appreciably enlarged.

If you are editing your images resize them first to the largest print size you will likely need, especially if resampling is necessary. Many common edits will look awful when later enlarged.
Viewing your image in the editor at print size should reveal any trouble spots which you should correct before doing other edits. For instance, color shifts frequently occur when manually correcting noise which becomes visible after enlargement. Any careful color correction done earlier would be wasted. Never add special effects like gradient borders or fancy text to an image until after it has been enlarged.

It is often best to simply use the original camera file if you are sending your digital images out for prints.The original file will most likely be small enough to send electronically yet still be of high enough quality to provide a good starting point. When sending digital images to an online snapshot printer, it is preferable to size your images within their posted guidelines. Too big and it may be rejected, too small and print quality will suffer.

Crop your image to the aspect of the print you are interested in to avoid surprises when sending digital files out for printing. This is especially true when sending it to an automated snapshot printer where subjects have been known to lose their heads. In most image editors you can specify a given aspect like H 20 x W 30. Fit the crop box to the image as you want it to look and crop it. Your image may not yet be 20"x30", but it will retain this aspect if it is enlarged to this size.

Never use photo sharing programs when sending images you intend to be printable. The requirements of images for viewing on a computer monitor are entirely different than those needed for printing. To speed transfer of images over the internet the images produced by these programs are usually very small, low resolution files. To further reduce file size they are also frequently highly compressed. If you view these images at a useful resolution for printing in an image editor you may be surprised at just how small they really are.
Photo sharing programs simply take your higher quality digital images and automatically down sample and compress them for speedy transfer over dial up connections. Information lost in this process cannot be replaced.

Color Calibration - At very least you'll need to be sure the colors you see on your monitor actually reflect the colors in the image you are viewing. This adjustment is best done with a special tool called a colorimeter and special software designed for this purpose. Most monitors are properly calibrated this way at the factory, but any serious photographer should consider buying one of these tools. This information is then stored as a profile which can be accessed by your monitor to tell it how to properly display images.
One must avoid the urge to jump into the monitor's settings and adjust them like it were a TV set, setting it to please the eye rather than acurately displaying what is actually contained in the image file being viewed. Changing the monitor's settings will not effect what is being sent to a printer or editing software, only the way you see it.
Your image editor must then also be calibrated to work with your monitor settings and your printer may require calibration. Printer software also stores media profiles which are used to attempt to accurately duplicate the results when printing on a variety of papers or other materials.
If you are not seeing your image acurately you cannot expect to see anything similar on paper as printers are designed to very acurately print the information as it is specified by the digital file sent to it.
The only way to be absolutely sure of what you will get on paper is through proofing.
Whenever large quantities of printed materials are to be produced, proofs are always ordered and approved in advance. Any necessary adjustments are then made until a satisfactory proof is obtained. All of the calibration and media profiles available today cannot match the usefulness of viewing the actual print produced. Once one satisfactory print is obtained, the results are easily duplicated with very little variation from one print to the next.

What's in a pixel?
Color digital images are made up of multi colored picture elements or pixels. Normally in photography these colors are specified as red, green, and blue (RGB). Each pixel contains the information needed to display the correct combination of these colors to provide the correct color in this small piece of the image. The amount of information each pixel contains is expressed in bits referred to as the bit depth of a digital image file. JPG images always contain 8 bits per color channel or 24 bit color. Many uncompressed image formats can contain 16 bits per color channel or more.

The actual size on disk of an uncompressed RGB image is the number of pixels it contains multiplied by the number of bits of information each pixel contains. This is expressed as bytes, kilobytes, megabytes, etc. ( 8 bits = 1 byte, 1 kilobyte = 1024 bytes)
An 8x10 digital photo at a resolution of 300ppi (2400x3000 pixels) stored in an uncompressed image format with a color bit depth of 8 bits per RGB channel (24bit) contains 172,800,000 bits of information. That is 21,600,000 bytes or 21,094 kilobytes or about 21 megabytes of disk space.

2400 x 3000 = 7,200,000 x 24 = 172,800,000 / 8 = 21,600,000 / 1024 = 21,093.75 / 1000 = 21.09

Digital image files can also contain other information about the image as well which will add a "bit" to the actual size on disk. Working with layers or a higher color bit depth in Photoshop can increase file size on disk quite dramatically.

In the printing industry CMYK (cyan, magenta, yellow, black) is the more common color specification used.
RGB images can easily be converted to this specification, but colors may shift quite dramatically. Significant editing to match the original may be required. This must be carefully considered before sending photographic work out for volume printing on a press.

Pixels can also store greyscale information.

*ppi = pixels per inch.
Is the term used to describe the resolution of most digital images. When resolution falls too low and the pixels themselves become visible it is referred to as "pixelation".
Pixelation is a term frequently and erroneously attributed to compression artifacts and other digital distortions seen in poor quality digital images that have been enlarged beyond their capacity but have yet to show any true pixelation.

*dpi = dots per inch. A term used to describe image resolution in the print industry for images made up of small dots on paper. Scanners and inkjet printers still refer to resolution in this way even though they are actually saving and working with pixels rather than dots. Because of this, dots and pixels have become somewhat interchangeable when casually discussing image resolution.


Chris is the owner of Postersize-it (www.postersize-it.com) an online digital photo enlargement and custom art printing service.
Postersize-it also offers a gallery for online print sales (www.postersize-it.us) and a large arts and photography link exchange (www.postersize-it.net).