Printing 'Castle Eilean Donan'

This week's column by Ctein

I frequently describe dye transfer printing as being a little like having Photoshop in the darkroom. For a complete description of the process, as I use it, please read this article. Because I'm dealing with individual separations, I can control the characteristics of each primary color component of the print—think "channels" in Photoshop. This lets me do things that would be impossible with conventional color printing. It also makes some kinds of controls trickier.

Castle Eilean Donan—1995. Photo and print by Ctein.

My photograph of Castle Eilean Donan, one of the four images Mike and I offered in the final blowout dye transfer sale on Wednesday, has several problems as a "straight" print. First, the light in the foreground and in the sky was considerably brighter than the light on the castle, so the print needs dodging and burning in if it's going to look good. (In my experience, most photographs can be improved by judicious dodging and burning-in.)

Second, since the photograph was made from a long distance away with a telephoto lens, there's some contrast loss and color shifting caused by scattering and haze in the air. If you've ever tried making photographs out an airplane window, you know what I'm talking about. The whites come out looking fine, but the scattering fills in the shadows, so that they're bluish and the scene is lower in contrast overall than you'd like. In the case of my photograph, the scattering is more greenish than blue, probably because of relatively low grazing light and all the foliage, but the overall effect is similar.

The third problem is that my telephoto lens had some uncorrected secondary lateral chromatic aberration. In other words, a bit of color fringing. I find this especially annoying in dye transfer printing, because it looks like I'm a sloppy printer and have misregistered the matrices, when it's the fault of the lens.

Solving the first problem takes a bit more cleverness than with conventional printing. Remember, this is a separation process: there are three separate exposures on three separate sheets of pan matrix film, one for each primary color. Manual dodging and burning in just won't cut it; there's no way I could do that precisely enough to avoid color fringing.

One of the smartest things I ever did, early in my dye transfer printing career, was to make up a bunch of dodging and burning masks out of sheets of 4x5-inch B&W film. I probably have 100 of these, in various densities, abruptness of gradients, and shapes of gradients (lines, arcs, and angles). The masks aren't sandwiched with the negative; they sit on top of the negative carrier, several millimeters away from the negative. That way, any dust or scratches or slight imperfections in the masks don't show up in the print.

When I need to do dodging and burning in of a dye transfer print I look through my assortment of masks and figure out which combination will do what I need. In this case I wanted a fairly sharp-edged burn of about a half stop at the bottom, and a much more diffuse burn of half a stop at the top. Figure 2 shows the two masks I used in combination with my negative for this photograph.

Fig. 2.

Reducing the aerial haze is pretty easy in dye transfer. I can mix up dye baths of slightly different contrasts. It's like making the curve steeper in Photoshop in one or more of the channels. In this case, I kicked up the contrast of the magenta dye bath by about half a grade, to neutralize the excessively greenish shadows.

Getting rid of the lateral chromatic aberration is a tricky thing, but it's doable (sometimes) with sufficient care. How did I do it? I changed the height of the enlarger head slightly between the exposure separations. That alters the relative magnifications for the different separations, and if I do it just right, I can undo chromatic aberration. For this photograph, a height adjustment of 1 mm was what I needed to eliminate the color fringing (figure 3).

Fig. 3.

Sometimes this trick doesn't work, and the separations fail to align properly. I suspect that's due to irregularities in the gearing of the enlarger head that causes it to slip a bit sideways if the gears are in just the wrong position. This time I was lucky. The trick worked perfectly and reliably for this photograph.

Just another ordinary day in the darkroom, when you're a dye transfer printer.

©2013 by Ctein, all rights reserved

Ctein's weekly column usually appears on Wednesdays, but this week we were busy with the sale on Wednesday. (May the admiring editor just add that this is the first time he has ever heard of adjusting enlarger height to correct LCA, and his mind is suitably boggled by the degree of precision in technique this requires.)

Links in this post may be to our affiliates; sales through affiliate links may benefit this site.

TOP links

(To see all the comments, click on the "Comments" link below.)
Featured Comments from:

No featured comments yet—please check back soon!

Comments

[Peter]

The impressive thing is, the simplicity of the explanations. These are the kinds of "know how" that only become simple after lots and lots of practice. Nothing in life is easy...

[Manuel]

Ctein, this article is absolutely fascinating. Thank you. Sad you're closing your darkroom, though...

[Roger Bradbury]

Good grief! It's easy when you know how, isn't it? I was particularly impressed by the cure for colour fringing, which is so obvious once you've been told.

[Peter Rees]

Fig 3 = wow.

[Richard Tugwell]

What's interesting, is that this looks similar to what I suspect a lot of people, including myself, are doing with simple adjustments in Lightroom using multiple gradient tools. Always seems to me more intuitive than Photoshop masks/layers, and of course adjustments can be stored in the DNG file

[Darr Almeda]

Wow, I never knew dye transfer printing was so tedious. Great photo btw!

[Pak-Ming Wan]

The chromatic aberration trick totally floored me. As you explain it, it makes so much sense, but to translate that into a darkroom technique ... Amazing!

Pak

[Geoff Wittig]

That trick to reduce color fringing from the lens aberration is just beautifully elegant. It strikes me as one of those things that works better as a physical intervention on the negative/print than anything Photoshop does. I've got a few prized images taken with less than optically perfect zoom lenses that show color fringing. The de-fringing sliders in ACR are somewhat helpful, but invariably it's like trying to adjust a shirt that's too short in every direction. A move that eliminates red fringing at one edge introduces a bit of green fringing at the other. Zooming in to pixel-level and cloning or desaturating out the fringe sort of works, but it's the opposite of elegant. It just feels...crude.

[misha marinsky]

"Sad you're closing your darkroom, though..."

The relentless march of technology.

Some may decry what is happening, but digital is democracy.

[raizans]

How do you make those dodge and burn masks?

[David Bostedo]

Is the difference between the images in Figure 3 obvious to others? I see a very slight reddish fringe, but I had to stare at it for a long while. Is that all there is? Or is my color slight color blindness coming into play?

(And is that perhaps why I have trouble seeing what all the fuss is about various types of color printing? I.e., maybe I'm not capable of seeing why dye transfer is better?)

[Kenneth Tanaka]

A very interesting article that few (and fewer) people could have written, Ctein. I am now even more eager to see the print (which I ordered).

[Gordon Lewis]

I can see why you won't miss your darkroom or dye transfer printing. You are better off applying your considerable knowledge and skills to tasks that provide a higher ratio of reward to effort.

[hugh crawford]

The first time I saw a demo of photoshop's channel feature the comment was "gee now I don't have to have dye transfers made to change the color of someones shirt" which at the time was done by telling someone over the phone to make said shirt look like pantone chip XXX and the next day a messenger would drop off a print.

Geoff,
The last couple of versions of lightroom can remove lateral chromatic aberration automatically even if the lens is a little decentered which it sounds like is the case with you. It even does a amazing job on aquarium photos

[ctein]

Dear Manual,

No it's not. It's going to make me very happy! And this is entirely about me; it's my darkroom, not yours.

Now, if I was forcing you to close down YOUR darkroom, then you would have ample reason to feel sad.

~~~~

Dear Darr,

Well, there's a good reason why dye transfer prints have been so expensive!

~~~~

Dear Geoff,

I wonder if you're doing something wrong in Photoshop or if you're really correcting lateral chromatic aberration rather than some other optical defect? I'm not trying to second-guess you or your skills; it's just that I have found that correcting for LCA in Photoshop or Adobe Camera RAW works spectacularly well. It's never failed on me, not once, and the results are excellent, all the way down to the single-pixel level.

There are other aberrations that can produce color fringing, and LCA correction can't entirely fix those. Also, if you're trying to apply LCA correction to a off-center-cropped photograph, it may fail, because LCA is symmetric about the optical axis.

Just some wild ass guesses. If you want to e-mail me privately and send some sample images, maybe I can help you diagnose the problem you're having.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[ctein]

Dear Mike and others,

Well, the chromatic aberration correction is a lot easier to describe than to do. The theory is simple; the execution is rather tricky.

I don't know if that trick is unique to me, but it's possible. Mostly because most printers won't obsess about that kind of thing [smile].

The use of dodging and burning-in masks for color separation work is a trick that I don't know of anyone else using. (Okay, it wouldn't surprise me to find out that Joe Holmes did it for decades; I don't think there's anything about darkroom printing that he doesn't know.) It's very common in black and white printing, especially for volume custom printing. I think Bruce Birnbaum and others have even written articles in the magazines we wrote for about drawing such masks in soft pencil on a sheet of frosted acetate and using them with B&W negatives as I use my pre-fab ones.

Now I'm curious. Any readers know of other printers using these kind of masks for color work, especially separation work?

~~~~

Dear Richard,

Yes, exactly! Each of these tricks has a direct analog in Photoshop.

I think one of the reasons I had no trouble adapting to Photoshop is that I had several decades of experience already working with separation “curves” and “channels,” so thinking the way Photoshop thinks came a lot more easily to me than to darkroom printers using more conventional processes.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[Richard Tugwell]

@Ctein

I probably didn't make my point clearly enough, which was that LIghtroom, as opposed to Photoshop, feels to me much more analagous to old darkroom dodging and burning techniques. I concur with your observation that most images benefit from some dodging and burning (esp B+W) and with Lightroom I can achieve this much more quickly than with Photoshop, to the extent that dodging and burning have become standard adjustments when processing RAW files. I recently did a photo shoot for a local bar / restaurant and all 50 or so of the images delivered had some d / b applied. I could achieve this in a fraction of the time that traditional darkroom work would have required.

[ctein]

Dear raizans,

Basically, it goes like this. Buy yourself some boxes of 4 x 5 graphic arts film. I think I used mostly Ortho Type III, 'cause I could work with it under her red safelight, but it doesn't much matter. Work out the development times, using very low dilution developer, to give you the density you want. It's important to find a developer, time, and dilution that produces neutral density; some film and developer combinations, when you severely underdevelop the film, produce a brownish image instead of the neutral one. You don't want that when you're printing color. Don't ask me for starting point on this; I made these decades ago. Who remembers?!

Stick a sheet of the film under your enlarger, hold a piece of cardboard over the film, flip on the enlarger, and do an appropriate burn on the film. If you hold the card stationary above the middle of the film, you'll get a pretty sharp edged gradient. Wiggle the card back and forth and you get a broader one.

To get arcs and curves, you can either cut out a curve in the cardboard or you can use your fist.

The object is to get an assortment of each shape, with different sharpnesses of gradients, and different densities, so that you can mix and match them together to produce the kind of dodging or burning pattern that you need. I've used as many as four at one time with a negative.

You can just batch process in a tray; it's not like you have to be very careful with these. Once you figure out what you're doing, you can make 100 of these in a weekend, and that should cover you for the rest of your life.

It helps if you have a transmission densitometer, because you can then measure the density difference between the exposed and unexposed parts of the film and write that in the corner of each sheet. That comes in really handy when you decide you need, say, a one third stop burn on a part of the picture; you just go sorting through your masks to find the ones with the right density difference and then pick the appropriate shapes and gradients from those.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[Oskar Ojala]

Very interesting article and between the lines it's evident how an all-digital workflow can save a lot of time.

The CA removal trick is neat, but doesn't it reduce sharpness due to a channel getting defocused slightly? I'm assuming the effect is so small to be worth the tradeoff.

[ctein]

Dear Richard,

Ah, no, I didn't pick up on that point, but we're still definitely on the same page. Saying that Photoshop's way of working reminds me of dye transfer is NOT a compliment, in terms of photographic intuitiveness. I give Photoshop somewhere between a C- and a D. It's in its genes, unfortunately; all one can say is that it used to be a lot worse. If you can imagine that.

A lot of photographers find Picture Window thinks much more like they do. It's an impressive program, and it costs a fraction of what Photoshop does.

~~~~

Dear Oskar,

1 mm is far less than the depth of focus at the print easel.

pax / Ctein

[Richard Tugwell]

Dear Ctein - I was talking about Lightroom, not Photoshop

[Cmans]

Speaking of control over the image, I very recently took a street picture of a man and his dog which unfortunately resulted in either lens fog or sensor fog over the lower portion of the image and unfortunately the dog was obscured. I would have said Cest la vie, but this pic is almost iconic with the man sitting on a bench with his head down and his elderly Shizh Tsu (sorry SP?) 10 feet away facing him sitting upright on his haunches with an expectant but patient look on his face as if to say "I am ready to continue walking, but I can see that you need to meditate, so I will not bother you". The image software (or at least my grasp of it) could not remove the fog completely and any attempts i made ruined the contrast, and I didn't want to miss the 50 pct off prints sale at the "Big Box" , so I submitted this image as-is with fog and was willing to live with the result. While choosing my images I noticed that the screen seemed to take a little fog off, but when I got the print ALL of the fog was gone! So moral of the story is although we might think we are increasing the odds of making our pictures better via our image software, there appears to be a "Higher Power" and in my case it was a blessing, but i don't think someone like Ctein who travelled thousands of miles to capture the morning fog at Castle Eilian would fully appreciate having the fog removed from his image.

[ctein]

Dear Richard,

Got that. I was agreeing. The horrid UI of Photoshop encourages folks to look elsewhere for more intuitive tools.

pax / Ctein

[Marcelo Guarini]

Ctein, your work is really a work of art, so painstaking. I never thought that you can correct cromatic aberration in such a clever way. Rising that tiny bit the head though sounds so critical to me. I used to do unsharp masks for my 8x10 B&W negative printing in the past, but that was a joke compared to your dye transfer work. Regards

[David Dyer-Bennet]

David Bostedo, to my eye there's a very strong difference between the regular and corrected photo (figure 3) right along where the land meets the sky. In the left one I see a strong red cast for a few pixels just inside the white outline (the white outline exists in the corrected version as well, and is in fact more clearly visible). I can find signs of it in other places, notably the bottom right end of the diagonal light slash about 1/3 of the way up the left side -- and, in fact, the bottom right end of any light area in the print against a dark background.

[Irving Reid]

Somewhat OT, but I wonder if it would be possible to strike a deal with NASA and Ctein to print https://twitter.com/Cmdr_Hadfield/status/325743148166963200/photo/1

[ctein]

Dear Cmans,

Can I have that magic fog-removing tool, please? I live in Daly City!

~~~~~

Dear Irving,

Incidentally, the circular lake on the right of that photo is an ancient impact basin of rather substantial size.

pax / Ctein