Processing: Modern Approaches to Get Rid of Banding. Part 3

“Problems are not stop signs, they are guidelines.”
| Robert Schuller

Processing: Modern Approaches to Get Rid of Banding

This is the third part in a series of articles of processing. Especially, I’d like to focus on different methods to cope with quite well known problems and challenges in landscape photography. Most of these problems in processing appear as a result of either the low quality of the raw material or rude editing applied to the raw material. In earlier articles, I explained what banding is (Processing: What is Banding? Part 1) and introduced some fundamental methods on how to get rid of it (Processing: How to Get Rid of Banding? Part 2). In today’s article, I’d like to quickly highlight two modern approaches to cope with the banding problem.

1. Using RAW 32-bit Smart Objects

In Part 1 of my articles, I explained why we should always use the highest image bit depth available. Most of the standard JPEGs are made in 8-bit (JPEG-2000 uses 16-bit as do the RAW files), i.e. we have an available set of 256 (= 2^8) shades of red, green and blue creating an available set of roughly 16.8 million colors. When doing the math, 16-bit (65’536 = 2^16 shades of red, green and blue) corresponds to about 281 trillions different colors. The more shades of tonal differences we have, the less likely banding will appear. So what if we edit in 32-bit raw? This is possible by using Smart Objects.

Smart Objects have been introduced in Photoshop CC 14.2 already (January 2014). Originally, they have been introduced to preserve an image’s source content enabling the photographer or artist to perform nondestructive editing to the layers. From Lightroom or Camera Raw we can open an image file as a Smart Object in Photoshop by clicking the right mouse button and choose EDIT/OPEN AS SMART OBJECT IN PHOTOSHOP.

Image 1: Opening As Smart Object in Photoshop 

When your image opens as a Smart Object in Photoshop, you’ll notice a little icon on the layer thumbnail. By double clicking this icon, Camera Raw re-opens and you can fine-tune the adjustments.

Image 2: A Smart Object layer in Photoshop 

The downside of Smart Objects are that we integrate raw information into a layer which increases the file size. The advantages however are that we have a nondestructive workflow and can always re-adjust parameters. Once we have created a Smart Object layer, we can right click this layer and create a copy of this layer as new smart object. This technique has the advantage that we have two independent Smart Object layers. Both can re-adjust parameters in Camera Raw by double-clicking the layer, but independent of each other. Originally, this technique was widely used with double-processing techniques (for example, without having multiple exposures of an image, one copy of an image is processed for the highlights, another copy is edited for the shadows).

Image 3: New Smart Object via Copy

As a result, we have two different Smart Object layers.

Image 4: New Smart Object Layer via Copy

Now we can start working on the second layer with Camera Raw adjustments and blend it into the base layer with Layer Style Blending options (Blend If), basic masks or advanced color masking techniques. The main difference is that instead of degrading the image file to 8- or 16-bit files through adjustments, we can work nondestructively. This technique can be used for dodging and burning as well, for adding texture and clarity or for creating a dreamy Orton atmosphere. Once we are done with these adjustments, you can rasterize the layers to create actual pixels (of flatten the layers down). This file can be saved as a PSD or Tiff file and be used for further adjustments. Well yes, not everything can be edited in 32-bit, for the possibilities are huge.

You can find a few YouTube videos on related topics. Mark Metternich announced to offer a tutorial on this kind of 32-bit processing soon.

 

2. Using the History Brush and Clone Painting

The History Brush is another Photoshop tools that allows nondestructive processing. This tool uses one or more previous states or so called snapshots of an image, creates a copy of  these states and uses the content of one of these copies to paint with on a working image layer. Thus, the history brush basically uses the information in the image’s own pixels to modify the very same image. A similar idea is being used by Marc Adamus for years. He titles his very artistic technique Clone Painting. Instead of having different history snapshots like with the History Brush, Marc uses different image files that he synchronizes. Afterwards he uses the Clone Stamp tool to clone real pixels from one image into another. The core idea however, remains the same: Using real pixels from original images (in Marc’s approachs sometimes different images, in Allister’s approach the very same image) to paint other images with. In the case of the History Brush, we start with creating a duplicate layer of the original image file which is going to work as our work space layer.

Image 5: Original Image File and Copied Work Space Layers

Next, with the Camera symbol in the lower right corner we can at any time create a snapshot of a current state. We can rename them, in this case it is called Snapshot 1.

Image 6: Creating a Snapshot with the History Brush

The History Brush icon in front of the two snapshots in Image 6 (right now adjusted to _RZ74968) sets the source for the History Brush. It basically means that the History Brush is defined to use this snapshot layer as source in order to paint it onto the work space layer. As we don’t want to use the initial snapshot, we change it to Snapshot 1.

Image 7: Snapshot 1 Acting as the Current Source File for the History Brush

By using different Blending Modes, we can darken an image with Multiply or any darkening Blending Mode, or we can lighten an image with Screen or any lightening Blending Mode. With a low flow we then use the History Brush Source layer to paint onto our work space layer. To do this, we:

  • Select Snapshot 1 as History Brush source file;
  • Select Work Space as our working image layer;
  • Choose the History Brush with a low flow (about 5% for example), potentially at an average Opacity (about 50%) to not overdo the effect;
  • Select a darkening or brightening Blending Mode;
  • Potentially, select a specific luminosity mask and activate it to paint through the mask;
  • Paint from the Snapshot 1 source onto our Work Space image.

The precision can even be improved by working through luminosity masks in order to only darken the darkest darks. Instead, one could also use color or saturation masks to only work through specific color or saturation ranges. If he overdone things, we can just change the Blending Mode to Normal and paint pixel from the original image file back onto the work space image. Thus, it is a very painterly technique.

By using image adjustments, we can create new snapshots, for example to darken/lighten the overall tones, for changing the saturation or adding overall contrast. Using these snapshots as source files in the History brush starting with the Snapshot 1 start layer, we can paint these effect locally in. For a demonstration of this procedure, have a look at reference 2. The advantage of this process is that we always paint in original pixels and do not need to apply any 8- or 16-bit mask that might create banding effects. The disadvantage is that we don’t have a pile of layers we could go back stepwise. Instead, we have to navigate through history states, but we also save space by that.

With this method, we have very precise control of what we want to modify in our work space image with a nondestructive workflow and without creating halos or any other artefacts through bad masking techniques.

While there are a number of YouTube tutorials on the History Brush, Allister Benn has a nice set of three videos showing its application. They are listed in the references.

Conclusion

In this article, I talked about two modern methods to avoid banding problems in image editing. In fact, it seems highly advisable to use the combination of both technique mentioned above in each of your editing workflow. The former allows to bake in some fundamental tonal contrast, color and color contrast as well as micro contrast adjustments. As a consequence, also Dodge & Burn or Orton effects can be implemented this way. The latter method offers the possibility to paint the image with itself. Therefore, it is a wonderful solution to locally dodge and burn with lots of fine grained emphasis. Both support a nondestructive workflow and will lead to a base image file of highest quality possible these days. Saving the resulting image to a PSD or Tiff file allows further adjustments that are not possible to use in 32-bit and preparing the image for specific target outlets, such as print or web.

 

References

During these difficult times for artists and because I do believe we should cite original ideas, I am happy to connect with some great material to dive into these topics further:

Benn, Allister: How to Use the History Brush in Photoshop – Part 1, 2020.

Benn, Allister: How to Use the History Brush in Photoshop – Part 2, 2020.

Benn, Allister: The Best Dodge and Burn Method Ever, 2020.

Metternich, Mark: Video Tutorial (32-bit workflow coming soon), 2020.

Adamus, Marc: Workflow Example and Clone Paint and History Select Techniques, Video Tutorial, 2020.