May
 2004

Of all the settings available in Adobe Photoshop, black point compensation is possibly the quirkiest. Using black point compensation when converting images using ICC profiles will result in one of three possibilities: the resulting image will be improved, the resulting image will not be changed, or the resulting image will be unacceptable.  And, of course, it is difficult to know up front which result to expect.

John Nate

Before I give you some suggestions for when to use black point compensation, I need to explain exactly what BPC is and how it is supposed to work. Let’s cover a little theory first, then we can move on to the practical application.

Simply put, BPC is a way to make adjustments between the maximum black levels of digital files and the black capabilities of various digital devices.

Figure 1 shows possible black-and-white densities for three commonly encountered colorspaces used in the graphic arts industry.

 

Fig. 1

The three bars represent the L*, or lightness channel, of these three spaces. Adobe RGB colorspace has a possible minimum or white density of L* 100 and a possible maximum or black density of L* 0.

This is effectively as white and black as you can get, and is beyond the capabilities of any of the output devices we use.

As shown, an inkjet printer has a smaller dynamic range with white coming in at L* 93 and black at L* 4.

Obviously the numbers can change greatly depending upon the printer, ink and paper combination used. Finally, we have newsprint, which displays a much smaller dynamic range with a white L* value of 84 and a fairly weak black density of L* 35.

Leave it on?

Adobe and many graphic arts and color management professionals recommend that BPC be turned on and left on.

What happens if we do this?  Take a look at Figure 2. The black level in Adobe RGB is slightly darker than what is available on our inkjet printer and both the Adobe RGB and inkjet blacks are much darker than what we can reach on newsprint.

If you convert an image from Adobe RGB to the inkjet printer’s colorspace with BPC engaged, the maximum black in Adobe RGB will be remapped to the slightly less dense black of the inkjet printer and all of the other tones will scale accordingly.

Similarly, if an image in either Adobe RGB or our inkjet colorspace is converted using our newsprint ICC profile, the maximum black will once again be remapped to the L* value of 35 and all of the other tones will be compressed accordingly, such as shown in Figure 2 (A). This is the desired result as we retain the entire tonal range of our original image although the tones will be somewhat compressed.

Turn it off?

If BPC is turned off the results are decidedly different. Figure 2 (B) shows what would happen in this conversion. The areas of our image in either Adobe RGB or our inkjet space that are at an L* density of 35 remain at that same L* 35 density after conversion and all of the detail from L* 0 to L* 34 is lost.

Fig. 2

This shows that converting from a colorspace with a darker L* value to one with a lighter L* value using BPC will ensure that the entire dynamic range of the destination colorspace is used and the image will retain as much shadow detail as possible.

So, BPC is a good thing, right? Unfortunately this is not always the case.

Take a look at Figure 1 again. In this example, our conversion is going in the other direction. This would be the case when we are proofing our images that have been converted for newsprint on our inkjet printer or soft proofed on our monitor using the Adobe RGB working space in Photoshop.

Won’t be accurate

In this conversion scenario, with BPC turned on, the weak L* density level of our newsprint image will be remapped to the maximum possible density of out inkjet printer or Adobe RGB (see A).

This may make the image look better, but it most certainly will not be an accurate proof of what this image will look like when reproduced on our newsprint stock.

Likewise, the softproof in Photoshop will not look accurate with BPC turned on. If BPC is turned off in either of these conversions, the weak 34 L* density of our newsprint image will remain at an L* level of 34 on both our inkjet proof and our monitor softproof (see B).

It won’t look as good as it could, but it will look more accurate.

(Editor’s Note, this conversion would probably be done using the relative colorimetric rendering intent.)

So once again, the answer to a color management question is “that depends”: Depending upon the results you are after, either accuracy or as large a dynamic range as you can obtain, you may or may not choose to use BPC.

In Photoshop, BPC can be turned on or off in one of two places.

Figure 3 shows the Photoshop color settings panel where you would set the default setting for BPC and rendering intent.  

Fig. 3

More important is the “convert to profile” panel shown in Figure 4. If you use this panel to perform your conversions (Image>Mode>Convert to Profile) you can preview not only whether using BPC will give you the desired results, but you can also experiment with the different rendering intents to ensure that the results you obtain match the results you desire.  

Fig. 4

Like many options in a color-managed workflow, the key is to try different options and determine what works the best for you, your images, your equipment and your workflow.

John Nate is a color specialist for Chromaticity Inc., an integrator of color technologies for the newspaper and other industries. He can be reached at 616.361.7773 or via e-mail at jnate@chromaticity.com.