When using Adobe Photoshop (PS), what color space do you work in…?
The two things I hear all the time are:
- Work in RGB. You can choose more colors. They’ll pop more. And, you’ll have access to more filters.
- Work in CMYK. If you start in RGB, you’ll have to convert to CMYK before you print anyway.
My final goal with most of my work is print, so this explanation is geared to that end. The following charts were created using PS Creative Suite (CS) 5.1 and InDesign (ID) CS 5.5; then exported as RGB *.jpg’s for web posting.
I choose to work in CMYK because I can control the color better.
1. We use K20 as part of a 20% CMYK test pattern at the shop. It’s a neutral color that gives us a good starting point for the following conversion exercise.
These charts illustrate (left to right) a 72 dpi grayscale square filled with 20% black (K20)1 and it’s behavior through color conversion followed by one pass of PS’s color halftone filter (Filter > Pixelate > Color Halftone) using default settings.2
What do those charts have to do with anything…?
I thought it might be a simple way to show what happens during color conversion.
Suppose you’re creating a graphic that’s to be printed and used on the web. Based (loosely) on the charts above, starting off with an RGB document has potential for a more vibrant onscreen image. That’s because RGB has a larger colorspace than CMYK, so you get a wider range of visible color. But, since a large portion of the RGB colorspace falls out of CMYK gamut, you risk losing a lot of that saturation when converting for print.
Starting with a CMYK document does limit your color palette. But, it’s colorspace is smaller than, and falls within, the RGB gamut, so converting for onscreen display doesn’t really change what you see, for the most part.
What does that all mean…?
Basically, If I’m drawing something with intent to print, I don’t want my colors getting dulled out beforehand.
What’s the final breakdown…?
When I draw or design something, I start in CMYK so that what I print out will more closely resemble what I’m looking at onscreen. I just want to avoid an unexpected color shift.
- More accurate color for print.
- Conversion to RGB is less likely to result in color shift.
- Less color choices than RGB.
- Reduced PS filter options.
- More color choices than CMYK.
- More filter options.
- Higher chance of color shift when converting to CMYK.
- Less accurate color for print.
Can you see my source files…?
For this one, sure. Download them with the link on the left. I conducted this exercise on OSX, so the color labeling doesn’t apply if you’re looking at the files on Windows.
Green files are the ID files used to export the web images used in this post.
Typefaces used in the ID files.
Green files are layer exports out of the gray files in /_PSD.
Red files are byproducts of the layer exports.
Gray files are the native files that I began with, complete with named layers.
Purple files are the images used in this post, exported out of ID.
I’m open to hear opinions or ideas.
What do we know about the subject at-a-glance? And, how can we expand on that knowledge…?
Comments are welcome…