It’s really easy to get tripped up over image file size, but once you get a few key concepts, you’ll be off and running in no time.
WHY DOES FILE SIZE MATTER?
Imagine going to the supermarket and picking up a can of Coke and a can of seltzer water. Both come in 12 oz cans, but since I’m trying to lose a few pounds, the most important thing to me is the number of calories in each drink. It turns out that Coke has 110 calories per can, whereas the seltzer has zero. So the volume of the can (12 oz) has nothing to do with the amount of calories.
Photography has a similar ambiguity. When dealing with stock photography, we’re concerned with the amount of picture data, and the uncompressed file size is the only accurate way to determine the amount of picture data in a given image.
MEGAPIXELS – Your Camera’s Sensor
Your camera has a sensor that is used to capture an image. The sensor is composed of a whole bunch of photosites that collect light to create the larger image. Each one of these photosites represents a pixel in a picture. The number of total photosites on the sensor determine the camera’s resolution, which is expressed in megapixels.
For example, the Nikon D300 has a sensor size of 4288 (width) x 2848 (height). If you multiply these two numbers together, you get the total pixel count.
4288 x 2848 = 12,212,224
If you divide this number by 1,000,000, you’ll get the camera’s resolution expressed in megapixels (mega from the Greek for “great”, which is commonly used as a prefix for million).
12,212,224 / 1000 = 12.21 megapixels
MEGAPIXELS AREN’T MEGABYTES – Pixels vs. File Size
A byte is a unit of storage in computing, and unfortunately, a byte isn’t big enough to hold a pixel’s worth of information. It actually takes three bytes to store one pixel of a color image.
So the pixels in your image store a color at a given point in the image, but it takes 3 bytes of storage to record this value. Think about it this way, if you store “mom” in your phone’s speed dial, the phone isn’t actually dialing “mom,” — that’s just the representation of your mom’s phone number. Three letters actually represents an area code and telephone number (10 digits in the USA).
So the file size of a color image is:
sensor width * sensor height * 3 = 36,636,672
which gives us the file size in bytes.
But this is a big number, so we want to convert it to megabytes.
There are 1,024 bytes in a kilobyte.
There are 1,024 kilobytes in a megabyte.
So the file size of a color image in megabytes is:
sensor width * sensor height * 3 / (1024 * 1024) = 34.9MB
OH YEAH? THEN WHY IS MY FILE ONLY 5MB?
34.9MB is a big file. Fortunately for you, those crafty computer scientists created file compression algorithms to reduce the file size to make working with the file easier. The most popular of these algorithms is JPEG. And the amount of compression is dependent on the image, so you might see files sizes ranging from 2 – 8MB when you’re shooting as JPEG.
Another file format that most D-SLRs offer is TIFF, which is an uncompressed file format. TIFF will actually give you the uncompressed file size.
SO WHY DO I NEED TO UNDERSTAND THIS?
PhotoShelter has a minimum 11MB uncompressed (and maximum 125MB) file size requirement for your photos, which roughly translates into a 4-megapixel camera. As long as you don’t downsize your camera’s images, you’re probabaly ok.
If you shoot film and are dealing with scanning, then you’ll want to understand file size because scanning services charge more money for the higher resolution scans (which are expressed in uncompressed file size).