Mixing B&W with layers
by eNoBlog on Nov.04, 2009, under Post-processing, Techniques
In yesterday’s Mixing B&W with local contrast write-up we saw how adding tiny S-curves rather than one large one can increase local contrast and help separate detail, especially among shadows and highlights that would otherwise wash together. Following up on comments that the final resultant image was still a bit flat, I decided to use layers to see if I could create a composite with a little extra contrast.
The approach, as detailed yesterday, used Channel mixer for the initial B&W conversion. At that point, we duplicated layers. To the background layer, we applied the same curve shown in yesterday’s write-up, and to the foreground layer, we applied an S-curve, used Opacity of 20 with Luminance blending, as shown below.
The selected settings produce deeper shadows that still retain tone separation to portray detail. I like this a little better than yesterday’s outcome (shown on the right below), but it does feel like it’s a bit of a push as now the image is departing from the original scene by a wider margin. Any more than this, and I would stop feeling comfortable with the conversion, but this feels just about right on the edge.
 Layered version: Click for larger image |
 Previous version: Click for larger image |
As I keep exploring B&W conversion, I need to mention a book by Michael Freeman, Mastering Black & White Digital Photography that has led to my discovery of the techniques I note here. Freeman states that Channel mixer in particular, along with the ability to control local contrast through various techniques, pulls digital B&W processing to a higher level than was ever possible with film development. Of course, this somewhat side-steps the additional dynamic range that B&W (Tri-X in particular) offers. Nonetheless, just having gotten my feet wet with B&W conversion techniques, I am really beginning to appreciate Freeman’s assertion. And of course, we should not doubt that one day in the not too distant future, digital sensors will acquire dynamic range equal to, if not superior to film. In the meantime, we have HDR.
November 4th, 2009 on 9:02 am
did you apply the same s-curve on the to the top and background layers or were they different?
November 4th, 2009 on 10:17 am
Bill – I applied the curve described yesterday (tiny S-curves in shadows, bright midtones and highlights) to the Background layer, and the S-curve shown today to the foreground layer, which used Luminance at Opacity=20%.
November 11th, 2009 on 9:34 pm
Coming from film, the lack of DR still bothers me a little. But you’re are absolutely right, multiple layers, especially when combined with masking, can result in some killer stuff.