Digital Lith Case Studies 5

Digital Lith Case Studies 5
Published: Tue, March 10th 2020

Originally published Sun, March 10th 2019

Grain again you think? Yes indeed. DigitalLith is about creating that grainy look of a lith print. Color? You can add later. Tone curve? You might manipulate before or after. All those possibilities like coloring, highlight contrast, mid-tone gradient, whatever are built for convenience around a grain machine. It is built so that you can feed nice b&w images and get nice digital lith images without the need of any other image editing program.

So yes, this episode is about grain again. And the next one will be so too. This time we are after a look of those papers which do lith, but do so in a more chaotic way. This leads to images where all the tones between shadows and highlights are built by a grain structure. OK, gray tones in b&w photography have all the times been build by grains, but we are going for a very visible and prominent grain.

IMAGE

To get there I wanted to use the grain gradient parameter and explain that a bit more to you. But then it happened that I understood my own program better - yes, it surprises me sometimes. I was about to show you how you can not get there without using the grain gradient parameter, but then I realized - ok, I can get there easily and what is key is not the grain gradient but the lookup table. In fact we do not even touch the grain gradient in this episode.

First off, here is the preparation for this episode. I chose an image which I took during a short visit to Düsseldorf. It is not too much fine detail and a lot of possibilities to get real black. Here you can download it in case you want to follow this episode with the same image: Click!

And for our experiments we start with a lookup table that you can download here bound into a preset: Click! But we will use it only for a moment and then go over to a different lookup table which we generate from the same data. It can be downloaded here: Click!


Side note: You can download the LUT definition here: Click! and then create the two LUTs that we use in this example yourself using the LUT generator. See the episode about the LUT generator if you want to know more: Click!

With the LUT definition in CaseStudies5Lab.data you can generate the following two LUTs:

ruediger> java -jar lutgenerator.jar -lab -distribute=LAB -lutdef=CaseStudies5Lab.data -lut=CaseStudies5Lab.lut
Read data from file: CaseStudies5Lab.data ...
Compute LUT ...
Write LUT to file: CaseStudies5Lab.lut ...
Done.
ruediger> java -jar lutgenerator.jar -lab -gradient=0.7 -distribute=MULT -lutdef=CaseStudies5Lab.data -lut=CaseStudies5Mult.lut
Read data from file: CaseStudies5Lab.data ...
Compute LUT ...
Write LUT to file: CaseStudies5Mult.lut ...
Done.
ruediger>

Please note above, the LUT definition itself is already in lab mode which is the reason we use the -lab parameter. When we create with distribution method MULT it will compress shadows so much that two of the given values in the definition would fall on the same lightness value. For that reason we spread the values a bit using gradient 0.7.

Needless to say that you need to store the lookup tables in the directory where you store the lookup tables so that DigitalLith can access them.


OK, let us start with resetting the development parameters, selecting the CaseStudies5Lab lookup table and put a border around the image without exposing it. You will see that on the highlights side we do not get any differentiation to paper white and that the blacks block up a lot. For that reason let us increase exposure to 1 and decrease tray movement to 75. The result of the development you can see here:

IMAGE

Now let us trim the parameters to the look we want to get in this episode. That is visible grain to build the tonal scale.

If we go up with the grain value a good amount - and this time it is more on the extreme side, then slow down the highlights through the lookup table and also slow down the shadows with less tray movement ... but hang on, let us do this step by step. And we start with a huge amount of grain by setting the grain parameter to 1. That will as you expect get us to a very dark result. To adjust for that we need to increase dilution by a fair amount and set it to 60 and we also decrease tray movement again and set it to 25. Now here is what you get:

IMAGE

OK so far, next we will use the lookup table CaseStudies5Mult which pushes the dark tones to the left. Keeping all other parameters the same you will get this image:

IMAGE

You can see we loose a fair amount of differentiation in the highlights by this. To get it back let us increase highlight contrast to about 1.3.

IMAGE

This brings differentiation into the highlights, but the differentiation against paper white is still weak. For that reason let us increase exposure a bit which will add density to the highlights. Here is where we end up with exposure being set to 1.7:

IMAGE

Looking at it I find there is still too much blocked shadows. Let us increase dilution another time and set it to 65. This will get us to the final image of this episode:

IMAGE

Instead of looking at those images as a whole let us have a look at a direct comparison of two crops. One from the initial image (on the left) and one from the last image (on the right):

IMAGE

And here we are at the end of this episode. Do some experiments with your own images and see how you can work through it to get a grainy result. Up to this point you should know how to address the different issues that might show up. Have fun.