Digital cameras are everywhere now, and it’s time for the casual photographer to “catch up” and understand the basic terms and definitions.  This document strives to do that, in a simplified manner.

Digital

What, exactly, does ‘digital’ mean?   Computers (and digital cameras) work in digital format, meaning that everything boils to “ones and zeros.”   Humans work in a “tens” system (because we have ten fingers and toes) but computers work in a “binary” system, meaning everything is either “on” or “off.”  There are just two values in a binary system, and all information must be represented in a binary code.  How the heck do they do that?  By using the basic unit of a “bit”, and putting bits together as “bytes” you can represent a huge amount of data.

Think of a “bit” as a light switch, either ON or OFF.

If you only had ONE bit (or light switch) to convey your message, you could only send TWO messages:  either ON or OFF;  YES or NO;  GO or NO GO.

If you had TWO bits, you send the following FOUR different messages:

ON, ON or ON, OFF or OFF, OFF or OFF, ON

If you had THREE bits, you could send EIGHT different messages or patterns:

ON, ON, ON

ON, ON, OFF

ON, OFF, OFF

OFF, OFF, OFF

OFF, OFF, ON

OFF, ON, ON

OFF, ON, OFF

ON, OFF, ON

With four bits you can send 16 different “patterns” (2 raised to the 4^th power = 16).  With five bits you can send 32 patterns.   Six bits allows 64 patterns, etc.     Computer scientists decided to group EIGHT bits together to form a “byte”, and to use the 256 different patterns possible to represent information that humans can understand.    So when you talk about “megabytes” and “gigabytes”, what you’re really referring to is large groups of ON and OFF switches, representing numbers, colors, and other things that humans understand.

When it comes to digital photos, the “depth” of an image refers to how many colors are shown.   A black and white image is relatively easy to convert to digital, since black and white are only two choices and digital is built on blocks of two choices.  So it takes fewer bytes to represent the image.  Each byte can represent multiple dots or either black or white (plus some other information).  If you have a 256-color photo, it takes one byte just to represent the color of one dot in that image.  A “full-color” image is usually 2 million different colors (the human eye can’t really distinguish that many, but that’s besides the point), and so requires many more bytes of digital information to represent the image.

All these discussions about bits, bytes and depth refer to the digital camera FILE SIZE.   This is the amount of hard drive space required to store pictures.  Certain types of images store smaller than others (JPG’s, GIF’s, etc.) but regardless, when a photo is described as 2 megaBYTES in size, that refers to the PC real estate it takes up, not the size on someone’s screen.

PIXELS

The “resolution” of a digital photograph is measured in pixels.  Basically, resolution describes a large grid of dots (pixels), with each pixel holding a certain color.   The bigger the grid, the more color variation you can show, but at a higher cost in file size.  Digital cameras usually take photos that are about 1600 pixels wide by 1200 pixels high.  Some higher end cameras can be much larger.     If you multiply 1600 times 1200, you get 1,920,000, or almost 2 million.  Mega means “million”, so a camera that takes pictures at 1600x1200 is a 2 MegaPIXEL camera.    MegaPIXELS do not equal megaBYTES.    Because of compression in the camera and in the file type, a 2 megaPIXEL image is approximately 150,000 BYTES.

The resolutions of computer devices vary.   Digital photos can run from 1280x960 pixels to 3280x3280 pixels and beyond.    Computer screens, however, use pixels-per-inch (ppi) and their range is usually from 800x600 to 1280x960.    Printers measure their resolution in dots-per-inch (dpi).   Computer screens are usually 96 ppi and home printers often print at 1440 dpi.  So, it all gets very confusing when moving between camera and screen and printer.

A handy approximation for digital photos is:

· 1 Megapixel is 1280 x 960 = 4" x 6"

· 2 Megapixel is 1600 x 1200 = 5" x 7"

· 3 Megapixel is 2048 x 1536 = 8" x 10"

· 4 Megapixel is 2272 x 1704 = 11" x 14"

When you open up a large digital image on your computer screen, it will likely have a larger resolution that your screen can handle   (images are usually 1280+ wide, a PC screen 1024 wide).  So, the software that you use to display the image will normally zoom out or “shrink” the photo to fit within your screen size.   It does this by throwing away pixels, and only showing you every other or every third pixel.  The human eye ignores the “gaps” and rebuilds the picture in your brain.   The PC hasn’t actually altered the image on the hard drive, it has only altered the electronic copy that it is transmitting to the screen for you to see.

JPEG and Compression

When a camera or PC software makes a digital image, it saves it as a certain file TYPE.  One type is a BITMAP, which is essential a huge grid, laid out just like the image itself:  the width and height are equal to the pixel width and height, and the depth is stored in bytes as data.  Bitmaps are HUGE files and take up lots of hard drive space.  The original bitmap format was by Microsoft, and had a BMP extension.  Tagged Image Format (TIF) is also another bitmap type.  It is known as a LOSSLESS file type, since ALL the data is in the file.  You should avoid using bitmaps were possible.

The Joint Photographic Experts Group (JPEG) created a standard of photo compression and file type that most digital cameras use today.  In a nutshell, the JPG format “throws away” extra data from the original pixels, making the filesize smaller.  It uses an algorithm to represent data that repeats, so for instance, instead of placing 1000 bytes in a row of “sky blue,” it might place the single byte that represents “sky blue” followed by a formula that says “repeat 1000 times.”

Every time you open and re-save a file in the JPG format, the compression algorithm throws away a little bit more data.  This reduces the quality each time.  This is known as a LOSSY file type.

Resizing Your Digital Photos

Don’t.  At least, don’t alter the original.

If you think your digital photos are “too big” because they scroll off your screen, then you need to adjust the ZOOM of your software.  If you resize the photo so that it looks good on your screen, it may loose too much information from the JPG compression to make a good paper print.   If you size your photos down so that they “fit” into an email, you will loose detail that you want later for other purposes.   Always keep the original JPG or RAW (camera format) file in a safe place, and only work with copies.  That way you can scale them up or down or web-sized or email size, but still have quality images if you decide to print them out.

Here are some guidelines, both in resolution and filesize:

ORIGINAL FILE:

Leave at original resolution and filesize.  If you edit the original, always do a “Save As…” and place the new file in a different location.

PRINTING:

You can crop out unwanted portions of a picture and zoom in for greater detail before printing.  Set you JPG compression quality to “High” and use the crop to these sizes:

EMAIL:

Try not to email pictures.  It chokes down the internet.  Use an online service like Picasa or Shutterfly, then email the link.  If you have to email a picture, filesize if the most important criteria.   Less than 100kilobytes per picture, and certainly no more than 5 pictures in an email.  Otherwise the other person’s email server may reject your message.

WEB:

Most people run their browsers at no more than 800 pixels wide.  So any web picture on a site will be too big if it’s wider than 800.  I always aim to keep my web images no wider than 600 pixels, and even 400-500.  Filesize is also important in a web page, because if it takes too long to load, people will leave.  I try to keep my web pics under 50kb.  30kb is best.