A single bit (short for binary digit) is a zero or a one, or a true or a false, or for that matter any two mutually exclusive states.
The file size of an image is the digital size of the image file, measured in kilobytes (KB), megabytes (MB), or gigabytes (GB). File size is proportional to the pixel dimensions of the image and the amount of colour in it. Images with more pixels may produce more detail at a given printed size, but they require more disk space to store and may be slower to edit and print. Image resolution thus becomes a compromise between image quality (capturing all the data you need) and file size.
Another factor that affects file size is file format. Due to varying compression methods used by GIF, JPEG, and PNG file formats, file sizes can vary considerably for the same pixel dimensions. Similarly, colour bit-depth and the number of layers and channels in an image affect file size.
Create a new canvas with these settings.
Use the Marquee tool to create 4 quarters in Cyan, Magenta, Yellow & Black as shown. To make this easy, set the Marquee tool to a fixed size of 1 pixel by 1 pixel. Type "CMD zero" to center the canvas and Zoom to maximum magnification by repeatedly typing "CMD +".
Now double-click on the Magnifying Glass tool. What you are looking at is one image pixel for every device pixel. This means that one magenta pixel is displayed by one pixel of your monitor. The proof of this can be diplayed by zooming to maximum view with rulers on and set to Pixels. You'll see what each colour square measures.
To show you how re-sizing images damages them, go "Image >Image Size". Enter "200%" in the "Pixel Dimensions" field. See how the image gets "muddy"?
Now, "Undo" this image size change.
Let's take a moment to take a look at the Channels Palette while we have our image in CMYK colour mode. What does the Channels palette tell us about the image? What does it tell us about how Photoshop understands colours?
Want to see how channels are affected by colour changes? Simply draw a Marquee around one of the pixels. Click on the "Foreground Colour". Enter a 50% value for the process colour you selected. Click "OK". Go back to the Channels Palette. What has happened?
Let's change the image to RGB mode by going "Image >Mode >RGB Color". Take a peek at the Channels Palette. What happened? How many are there? The CMYK colours are being mixed with RGB colours. So it takes certain percentage mixes of each process colour to make up RGB colours.
Sample the Cyan square and click on the "Foreground Colour". The resulting exclamation mark indicates that Cyan is out of the RGB gamut. This means that pure Cyan cannot be displayed with any combination of RGB colours. The colour needs to be 'nudged' into the RGB space by clicking on the exclamation mark.
Now we're going to get rid of all the colour in the image by changing it to Greyscale in the "Image >Mode" menu. Go back to the Channels Palette. What do we notice?
Now go "Image >Image Size". Enter a large value in the "Pixel Dimensions" field. See how the image gets "pixellated"? This is what happens to an image when you scale it larger. Not very nice.
Convert the image to "Bitmap Mode". Again, notice the change in the Channels Palette.
Now this is really cool. Increase the image size by 200%. Increase it yet again. And again. Pretty cool eh? What is going on here?
The effect of bit depth on file size can be great. As you can see in the table below, a 16 bit image has to store 65 536 possible colours for each pixel in the image. Multiply that number by the number of pixels in the whole canvas, and you get a quite large number!
| BITS | INDEX | # COLOURS |
|---|---|---|
| 1 bit colour | 21 | 2 colours |
| 8 bit colour | 28 | 256 colours |
| 16 bit colour | 216 | 65 536 colours |
| 32 bit colour | 232 | 4 294 967 296 colours |