Digital Outback Photo
- Photography using Digital SLRs

The Art of Digital B/W #011

 

The Challenge Of Digital Black And White
Part 2: Editing Your Image

a personal view by George Barr

 
 
 
Part 2: Editing Your Image
 

Perhaps what you’d really like to know is just how good can black and white inkjet prints be, and do you have to feel guilty for not spending hours in the dark with smelly chemicals. Do you need pyro stained hands to be a ‘real’ photographer?

Of course, I can simply tell you, no you don’t need to, that black and white inkjet prints can hold their own, that they do compare to silver prints and do deserve to hang in galleries, be sold, collected and generally admired. Problem is, it’s only my word, and there’s lots of respected traditional black and white printers who regularly announce they have never seen an inkjet print to equal a silver print. Mind you, they have a huge investment in time and reputation based on silver printing, so perhaps they have a viewpoint no less biased than my own. Those of us who have invested thousands in fast computers, Photoshop, colour profiling, and high quality printers clearly do have a vested interest in inkjet being top drawer.

So how do you the reader interpret this? In an ideal world you could look at prints made both ways, each by an expert in his field, all from the same negative or file. Regrettably, for many that isn’t practical. You don’t have ready access to top quality prints of either medium for comparison. You can rely on experts, but which ones - they don’t all agree. You can look at images on the web, but what’s that got do do with print quality?

If what you are wanting to do is produce quality prints and you don’t have a darkroom, why not give digital black and white a try. Listen to advice from the experts, perhaps follow some of the tips I am discussing here - and see if you like the results. If you are already an accomplished wet darkroom printer and you are wondering about switching, the good news is your skills in the wet darkroom translate quite easily to Photoshop and with a little help you can be up and printing digitally in no time. Either the inkjet prints are going to be better than your wet darkroom prints or they are not. If they are better, then your problem is solved. If they aren’t, and you have got past technical issues, then you either try harder or you abandon inkjet and go back to your wet darkroom in the confident knowledge that you are producing the best prints you can. I think you may be surprised to find that if you have good printing skills already, it won’t be that difficult to produce inkjet prints that you are happy to hang next to your silver prints. Once behind glass and hanging on the wall, it’s darn hard to spot which prints are which. In hand the prints currently look different because of the differing surfaces but where inkjet may be weak in the depth of the shadows, it is strong in the complete lack of glare from the matte print surface, image colour can be controlled better than in the darkroom (warm, cool, selenium, somewhere in between - just let me know) . It’s all very well having really deep shadows, but if you can’t see them for reflections 99% of the time, and if behind glass you can’t see them at all (compared to matte inkjet paper), then what’s the big deal? You think metamerism was invented with inkjet - then you haven’t seen the sometimes subtle yellow highlights of silver images (which goes just lovely with the purple shadows after selenium toning).

Anyway, on with the show - part two of a discussion on how I produce what other people think are very nice prints (and pay for them, display them, and publish them). In this episode, I’m going to show you how I get the image looking good on the screen - 95% perfect. Next time I will discuss separately how to get that last five percent, and the final installment will discuss printing.

I’m not going to get into computers except to say that I currently use a Mac G5 dual 2.3 with 4.5 gig of memory and 750 of drive. I use Photoshop for my editing and have absolutely no experience using other software so I can’t compare, I can only tell you what works for me, from the school of hard knocks.

I have only ever used Camera Raw for converting my raw images. Since version 3, I really haven’t felt the need to use any other but it’s possible I’m missing something - still, life is tough enough becoming an expert with one set of software (kind of like learning to use only one film developer combination). Most of what I talk about from here on will have an equivalent with other raw developers. Uwe tells me I maybe missing something by not using the other raw processors, but I’m 56 and i don’t have that much time..., besides, I’d rather be out shooting.

Having selected an image in Bridge upon which to work, I open it in Camera Raw and the first thing I do is double click on the magnifying glass to view the image at 100%. No point on spending hours working on a terminally fuzzy image. Camera Raw has a setting at the bottom of the screen which allows me to set the output file size as normal (ie. the same pixel dimension as the sensor), or 1/3 bigger or smaller, or even more change. Through experimentation I have found that I get the most detail and the best sharpening if I use one size up from normal then use a bit more aggressive sharpening later in the processing. Is this uprezing (which I have slammed elsewhere - I guess I’m guilty but would argue that doing it right from raw is somehow better, and the test is in the print - my 1Ds2 produces 16X20 prints that can be inspected from 8 inches away, without sharpening artifact visible).

I do not let Camera Raw do any of the sharpening for me (sharpen preview only). I almost never use the saturation setting in camera raw to convert the image to black and white, preferring to save the ability to ‘filter’ the colours later.


The image as it comes out of camera raw.

Most images can be adjusted in Camera Raw using the exposure, shadow, brightness and contrast settings to get a workable image. I do not however try to produce the best looking image on screen at this point. In working in 16 bits, I am more concerned that the image file sent to Photoshop has all the detail in the highlights and shadows that I will need - if it ain’t there to start with, there is no way to get it back. If this means a muddy looking image without much in the way of good blacks and whites, well, it’s only the start.

If I find that it’s not possible to show both really detailed shadows and highlights, I will use Russell Brown’s place-a-matic routine to blend two versions of the raw conversion of a single exposure in layers, one for the highlights with exposure slider moved left (darker), the other for the shadows (exposure slider to the right, shadow setting protecting all the shadow detail (even if this means no deep blacks in the image - I can create those later. There are newer Photoshop plug ins which can do something similar and even use 32 bit hdr imaging in Photoshop to blend the entire image into a single layer. So far I haven’t needed to do this but it is something I am looking at, complements of the suggestions of Uwe and his various articles. It may be that tone mapping will be the answer in the future (I’m now experimenting with it) but I have a suspicion that raw holds more information than can be sent to photoshop in one go if there are a lot of details at both ends of the curve, shadows and highlights. Now if tone mapping worked directly from the raw file...

From here on in, you have to know where you are going - that you know what a good image looks like, that you can recognize when things aren’t right - you can be a good driver but if you don’t know the destination... It’s not possible to show that in a short article such as this. Probably workshops are the best way if in fact the instructor is known for his or her printing skills. I don’t think it really matters whether the instructor is working in silver or silicon, the pathways are similar. That said, I wouldn’t go to a workshop which has darkroom work as part of the curriculum. I have been very happy attending Bruce Barnbaum workshops even though all his work is wet (well, you know what I mean). If you can’t afford a workshop with a good printer, then try to see good prints and failing that buy books of photographs. Reproduction these days is so far ahead of 20 years ago that you can learn a lot about the qualities of rich highlights and shadows from books. Let me strongly recommend subscribing to Lenswork magazine - it’s printing is absolutely superb. I had occasion to reprint one of my images for a customer after she had seen the image in Lenswork - I had to work darn hard - and I had the original file I’d sent Brooks Jenson, the editor.

I sell my photographs at a local Farmers Market. I have noticed that often someone will come in to the market with an slr around their neck. They seldom stop to look at my prints, yet I’d bet I’m selling more than they are. Even if they were to look at my images only to find fault, that would be a useful experience. They don’t even do that, they just walk by. I don’t understand. Photographers. even if they do stop to talk (and lots do); very rarely buy images. I find my market elsewhere. You could do a lot worse than buy a few good prints, but don’t buy them based on the image on the net.

Right, back to the topic at hand. My routine once the image is in Photoshop is to blend layers if need be, do basic contrast adjustment with a curves adjustment layer, convert to black and white, then do local adjustments using more curve adjustment layers with black masks into which I paint white at varying opacities to get the effects I want in the locales I want.

The very last thing I do is dodge and burn, but that’s a topic for the next time.

In making those first general contrast adjustments with a curve layer, there are five things I am looking at - the white point (how close do the darkest pixels in the image come to pure black), the white point (ditto), the shadows, mid point and highlight parts of the curve.At this point I’m going to digress and discuss why I don’t simply use burning and dodging on the image as my only manipulation . Well, I quickly found that if I set the undos to 20 (seemed like plenty), I frequently couldn’t go back to the beginning of working on one part of the image - remember that 20 undos is 20 strokes of your pen or mouse and often in burning and dodging I’d use hundreds of light strokes - clearly I couldn’t rely on undo. I could simply save multiple copies of the image - but boy that sure eats up disk space quickly and besides, remembering which version had which adjustments becomes a pain.

The second problem with dodging and burning is that once a given pixel is driven to pure white or pure black, it no longer contains any relevant information about the image. Gone is gone. With a curve adjustment layer, I can lighten or darken without pushing the most extreme pixels over the edge (unless I adjust the black or white points).

So, the answer is adjustment layers. They can be saved within the single image. They can be thrown out at any point, they can be toned down using the layer slider, they can be masked and adjusted at any future point before finally flattening the image. they don’t eat away at the image.


A curve layer to balance the image before converting to black and white.

There are various ways to use layers to adjust your image, curve adjustment layers just happens to be the one that works best for me. I don’t think I’d use it if I were a commercial photographer having to edit dozens of images a day - I frequently take several hours to get a single image right - just as much time as it took in the wet darkroom (but at least it doesn’t have to be continuous time). I use a mouse for my brush strokes (it’s all I have) and Uwe tells me I’m doing it the hard way - that an overlay layer is much easier. He mabe right, but I wasn’t aiming for easier. That said, I have reasonable drawing skills and have LOTS of practice with a mouse. Were I a commercial photographer preparing dozens to hundreds of images for client selection and perhaps a dozen images the client actually pays for, then I’d want a much easier way to do things. Uwe has a number of tools which help in this direction and if you are all thumbs with a mouse, you might well find a better way than mine, but this is my article about how I make my prints - take it for what it’s worth.

Photoshop offers several different ways to convert to black and white, from converting to grey scale to desaturating the colour, to working with channels. Early on I learned about another Russell Brown trick - he creates an action which creates two new hue/saturation adjustment layers. The first layer is set to colour instead of normal (upper left in the layers palette). The second is left as normal but the saturation slider is set to the far left to remove all colour for the image. You then use the hue slider of the first (lower) layer to effectively alter the colour of the image before it is desaturated and this has the effect of lightening some colours and darkening others and you can simply play with the slider till you get the effect you want. You can even adjust the saturation of the lower layer to modify the intensity of the effect.


image without any ‘filtering’


image with ‘filtering’ by adjusting hue slider

I highly recommend spending some time at russellbrown.com and view his quicktime tutorials.

I next work on parts of the image, using adjustment curves and black masks to limit the effect to where I want. I use a round brush with it’s fuzziest setting (so I don’t leave telltale signs of my work. I use the 1 to 0 keys to set the opacity of the effect. I can use black or white to paint with depending on whether I want to undo some of the effect or correct for going over edges with the brush. Some of my images have as many as 25 adjustment layers. Sometimes I have to flatten the image part way though working on it just to reduce it’s file size. I don’t generally save the unflattened version and occasionally regret not doing so when I find I need to go back before the flattening - that’s life - I could be better organized, I could be an accountant, I‘m not, I’m a photographer.

If I don’t think an effect is strong enough after painting into the mask, I double click on the left of the layer and up comes the curve again and I can adjust the curve and watch the image to see the effect. Other times I just add another layer and curve on top.


this is the painting into the black mask - lighter shows more of curve effect

In general, the curves I create tend to have some basic shapes. S shape increases contrast. A sagging curve darkens the image, increasing the contrast and separation of the highlights while dumping the shadows together into the dark. A simple curve that is high in the middle lightens the image, separates the shadows and compacts the highlights. You can of course create compound curves but the odder the shape and the more extreme the variation from a simple smooth curve, the more likely you are to muddy some part of the image. Very occasionally I will adjust the black or white point of the curve - usually dropping the white point so that in the areas revealed by the mask there won’t be anything too bright - typically something in the background of an image that is distracting. Large areas of this are almost certain to be distracting, but used carefully...


the result of the first masked curve

I’m aiming for an image that has strength, yet shows good detail in the shadows. It should look rich, creamy, bold and subtle, all at the same time. Even Ansel Adams went though phases of printing lighter and darker but this was done between highlight and shadow, not messing with either end. The amount of ambient light in the room can affect what kind of image you like on screen - we often tend to sit in darkened rooms with only the screen lit - not generally a good idea - avoid glare on the screen by all means, but don’t work in a cave. Don’t even think of working on a laptop screen. They are a lot better than they used to be and Apple’s are the best, but a laptop LCD screen is not the same thing as an Apple desktop LCD screen (which I find is fine for editing photographs). I know photographers who use PC’s yet use an Apple display screen.


Not all images have to have a pure black or pure white anywhere - but you’’d better have a good reason for doing so - fog perhaps?

At this point I have the image looking pretty good on screen, it might even be exactly what I want, but usually I have some highlights that I want to sparkle, some shadows that are a little weak, and I will finally use dodging and burning to polish the image, then sharpen, adjust for printing and voila - 4 hours later I have a print, the first one, of several, before I get it right, before I am satisfied. See you next time.


The real, three image stitch, after all adjustments


A note on RAW Developer by George Barr

I have been writing a four part article on black and white in the digital world, and wrote that I have only used Camera Raw. Uwe Steinmuller, for whom I was writing this article (above) had recommended I check out
Raw Developer (available for Mac only from).

I had argued that fussing with more software takes me away from my camera, but it's been pouring rain all day so this evening I decided I had to have a look - boy, the images sure do look different from Camera Raw which has a distinct painterly look to it when making big prints. Raw Developer images do appear to hold low contrast detail
better. If you only look at high contrast edges there is no advantage to Raw Developer, but if you study the lower contrast areas, there seem to be significant advantages to Raw Developer.

That said, the smoothness of the Camera Raw image is in some ways more appealing and although painterly, doesn't look as 'grainy' as the Raw Developer processed image.

Frankly, this is a hassle, I don't want to deal with two raw processors. I know that for normal size prints (300 dpi of real pixels (not upsized) Camera Raw is great, for images larger than this, everything is a compromise and I think it comes down to what kind of compromise do you want - do you want the smoothness of the Camera Raw image or the fine low contrast detail of Raw Developer but which leaves the image looking 'grainy'. Sigh...


Camera Raw


RAW Developer

 

You can find George Barr's website here.

 


 
 
 
 
   

For Comments post in our News Group

© 2000-2004 Digital Outback Photo