Archive for February, 2015

Essentials: 256 Shades of Grey


Apologies if you were drawn to the post by the similitude a popular book/film but this knowledge isn’t likely to make your weekend any more fun. However, if you’re a novice or self-taught Photoshop user it’ll maybe make your workaday life a bit easier and you’ll probably have a bit more energy and enthusiasm for other things.

Before We Begin…

It isn’t necessary to know how the engine in your car (or on your motorcycle) works to be able to drive (or ride) but knowing little things like how it should behave or how it should sound, is really useful—and sometimes vital; ask any motorcyclist what is likely to happen going into a corner with the revs too low, although you can probably work that out for yourself. Photoshop can be a bit like that, and knowing at least a bit of what is going on “under the hood” can help you make more informed decisions about your edits. This post is the first of a series covering one of the fundamental aspects of any image you’ll come across in Photoshop, regardless of the colour model you’re working in: Channels.


While you’re most likely creating full-colour work, you may be surprised to discover Photoshop is at its core a greyscale editor; that’s right, greyscale with black at one end (a value of 0) and white at the other (a value of 255) with 254 other varying shades of grey in-between, giving us a total of 256 shades of grey. When you make edits to a colour image in Photoshop, you’re actually individually affecting each channel separately, and then seeing the composite result. There are a handful of special case scenarios also: Bitmap, Multichannel (which kind of includes for our purposes Duotone, Tritone, Quadtone) and Lab—you’ll find out about how those work later on.

Aside: RGB and CMYK Editing

As a side note, this is one of the reasons why RGB workflows will almost always yield better quality images than working in CMYK. Every command-generated operation (adjustments, filters, etc.) will create some image noise; as an RGB image has three channels and a CMYK image has four, you’re potentially introducing about 30% more noise into your final result. If your process demands that you produce CMYK content, then at least make the conversion when you’ve done working on the image, and if you’re working end-to-end outputting your final result from InDesign, then explore PDF/X-4 options.


The bitmap colour mode is the exception to everything else, in that this carries strictly two levels of information whether a pixel is “on” (black) or “off” (white). This mode is most commonly used for “line” art, and typically when used in print has about four times the resolution of colour work, starting at 1200ppi and almost always originated at the dimensions it will be used at.


In the image above, there are two representations of approximately the same area—indicated by the red box:

  • a is greyscale at 300ppi, magnified by 3200%. You can see how the differing shades of grey optically smooth out the curves.
  • b is bitmapped at 1200ppi, magnified by 1200% (the value is smaller as the image is larger) using a diffusion dither to simulate the greys. As the image can only have black or white pixels, a dithered pattern optically creates tones when viewed at size.


Open any RGB image, and go to Image > Duplicate… and click-through the dialog (that step isn’t completely necessary—although it is good practice—it’s just that I don’t want to be blamed for you accidentally damaging anything).

Now make the Channels Panel visible; it should be nested with the Layers Panel but if you can’t find it, go to Window > Channels. You’ll see at least four channels:

  • RGB—this is a composite channel that you usually work in, and is the combined result of the colour-map channels
  • R—this is a brightness (luminance) map of the intensities of red information in the image
  • G—likewise, a map of the intensities of green information in the image
  • B—you guessed it, this is a map of the intensities of blue information in the image.

If someone has been working on the image previously, it’s possible there may also be some Mask or Spot channels present, but you can ignore these for the time being.

Go to Image > Mode > Grayscale to convert your image, and just click “Discard” (or hit RETURN) if the conversion dialog pops up—this isn’t the best way to make a greyscale conversion, but we’ll look at that in a later post—and you’ll have a greyscale image; if you take a look at the channels, you’ll see there is now only one. The channel and the image itself now contain exactly the same information, 256 levels of grey, from 0 to 255 (remember that 0 is a number in computing).

Back to RGB

Undo the conversion, or go back to your original image. Now you can see all three colour-map channels again; click on each one in turn and you’ll see varying greyscale versions of your image; remember that these are maps of luminance in each channel, and at this point you may be a little confused as to what is happening here. Let’s do an exercise to try and make sense of it.

Exercise: Understanding RGB Channels

An exercise file is available for download to work along with this exercise: If you’re interested it’s of one of the great recycled underground trains, used as studios on the roof of Village Underground on Great Eastern Street in Shoreditch.

In this example, we are going to use Layers to stand in for the channels, along with Blending Modes to make the magic happen. Each of the layers here contains exactly the same information that the channels would have in the “real” image, only that here you are seeing them in colour, rather than just as greys (you can view channels in colour by changing your preferences, but it doesn’t make any difference, and in my experience only serves to confuse beginners).

1redChannelTurn the layer visibility on and off for each layer in turn, to see how the information changes. Now make the Red “channel” layer active (click on it) and change the blending mode to Screen, using the drop-down in the Layers Panel (it’ll say “Normal” by default).


You’ll see some yellow and yellow-ish tones appearing in the image along with red and green colours—the next section will explain why.

Now make the Green “channel” layer active and again change the blending mode to Screen. You should now be seeing the full-colour image.


So what is happening here? Let’s take a quick look at the basics of the RGB colour model and hopefully you’ll get a grasp on what Photoshop is doing.

RGB Colour

RGB is an additive model—additive because the maximum intensities of all the primaries combined (added together) equals white—and it is also very similar (but not identical) to the way that our own eyes work. It isn’t the easiest of things to explain but in training sessions I’ve had the most success with scenarios (even if they are quite wacky, as in the following example) so please indulge me, if you will.

Try to imagine that you are stood on a board, floating in space, and there are no stars. What colours would you see? None, right? Just blackness. You have three torches, one emits red light, one emits green light and one emits blue light. Turn on the Red torch and point it at the board you’re standing on, you’ll see red at maximum intensity in the middle of the beam and this will fall-off towards the edges of the beam.


Now turn on your green torch, and lets the beams overlap. Where the beams are uninterrupted, you’ll see red and green, and where the maximum intensities overlap, you’ll see yellow; as the intensities decrease towards their edges you’ll see other reddish/greenish/yellowish tones.


Finally, turn on the blue torch and you’ll see that the maximum intensity of the Red, Green and Blue primary colours, add together to make white.

The pure (maximum intensity) secondary colours created are:

Red + Green = Yellow

Blue + Red = Magenta

Green + Blue = Cyan

This is probably going against everything you learned about colour in primary school, but your teacher at the time would have been referring to paint mixing (we’ll deal with CMYK and ink mixing next time) so that information is worth holding on to, and you’ll find it really useful later on in this series. You should now see why there are three channels, and how those channels use luminance to create the colours you see in your image. Now what you need to know is how to use that information, and that is where we’ll be going in the next post. If you’ve any questions or comments, please post them using the comments box below, or on my Facebook page: