It's All About The "Small Details"
by Mark Dubovoy
ALPA SWA camera with PhaseOne IQ 180 back and Rodenstock 60 mm APO HR lens
I took a couple of months off from writing articles about photography and I also slowed down my photography work. I think it is good for the mind and the soul to take some time off every now and then in order to smell the flowers, re-energize and re-evaluate one's approach to all of one's passions, including life itself.
This is my first article of 2012, so first of all I want to wish a very happy new year full of success, health and happiness to all our readers.
This article will be the first in a two part essay. The first part addresses some very key issues regarding, as the title implies, the fact that everything matters in photography. There is simply no detail small enough to safely be ignored. All the small details matter and make a difference. The first part of the essay will concentrate on explaining why this is so. As part of the discussion, I will also address some personal philosophy and observations of what makes an image a compelling artistic statement.
The second article in the Essay will address a related issue:
Photographers and photo enthusiasts often get all wrapped up in product specifications and marketing pitches from manufacturers. They also get wrapped up in the minutiae of technology, even though very few truly understand it. Unfortunately, the web certainly does not help, because there are lots of self proclaimed "experts" that are, in reality, far from being true experts and do not understand either the craft or the science beyond a very cursory lay person superficial level. This situation has two immediate consequences: First, there is a lot of "missing the forest for the trees" taking place; second, it gives rise to lots of misunderstandings and myths that need to be busted, and I hope I can bust some of them in the second part of the essay.
The Underestimation Paradox
For reasons I cannot understand, the vast majority of people have a tendency to underestimate the capabilities of human senses. One often hears about a dog's olfactory system, or an eagles eye, but for reasons I cannot explain, one hardly ever hears anything about how incredibly sophisticated, sensitive, accurate and exquisite human senses are. Make no mistake, human hearing, human taste, the human olfactory system and the human visual system, among others, are nothing short of astonishing. After all, our very existence depended on these systems not so long ago; it was literally a matter of life and death. And not only are these systems extraordinary, but the interaction of these systems with the human brain brings an entire higher level of sensitivity and sophistication into the picture (pun intended!).
I frankly get tired of people saying things like "I cannot tell the difference between a fine wine and a mediocre one", or "I cannot appreciate the difference between a fine audio system and an average one", or "My eyes are not good enough to see the difference between a medium format original and a smaller one"....
Except for people that suffer from a specific disease, all the above statements are pure unadulterated BS. In other words, everyone can tell the difference.
Although there certainly is quite a bit of variability between individuals, pretty much all human beings have exquisite capabilities to discern the most minute differences in taste, sound, vision, etc., and they have incredible acuity to detect the tiniest clues. The real problem is that those who make these kinds of statements either never bothered or are too lazy to simply pay attention. So called "expert" listeners or tasters or visual observers are not freaks of nature. Although the senses of some of them may be well above average, the important thing is that most of them are normal people just like you and me. They have either received formal training or trained themselves to really pay attention to their senses, and they spend lots of time doing it, because like everything else in life, to be good at it takes commitment, training and lots of practice.
Forest and Branch
Ebony 4x5 camera with Rodenstock 210 mm lens and Better Light scanning back
The Audio Example
Other than photography, one of my other passions in life is music. I am particularly fond of acoustic jazz and classical. As part of that, I have been a serious audiophile for many decades. My definition of a true audiophile is very simple: It is someone who wants to own a system that can reproduce music as close to a live performance as possible. The ultimate goal is to be able to reproduce a performance of (unamplified acoustic) music with such realism that it gives you goosebumps and it literally fools you into thinking that the performers are in front of you.
I fondly remember when the first CD player from Sony was announced. They declared in all their ads, that the new CD format with this player delivered "Perfect Sound Forever". To prove it, they had all these lists of specs that showed vanishing noise, vanishing distortion, etc. Well, I purchased one, took it home and it sounded so awful that I immediately returned it to the store as a defective unit. They replaced it with another one that sounded exactly the same. I was confused, was there something wrong with me? To my ears, Sony's definition of "perfect sound" was what I would have called "abominable crud". On top of that, how could something with those stellar measurements possibly sound so bad? As I said before, I was confused.
So, here we are, several decades later, and not only Sony, but everyone else in high end audio agrees that those early CD players sounded awful. CD players have evolved a lot, but to my ears the vast majority of them still sound pretty bad. Newer high resolution formats have been introduced and they are definitely better. I will not dwell on this, but my auditory system still tells me that I have yet to hear a digital system that even comes close to Analog. This might partially explain why decades after the introduction of the CD, and as the market for CDs is in serious decline, the market for LPs, turntables and everything else analog is booming.
Interestingly, if you read the high-end audio magazines, or attend the high-end shows, or you visit high-end dealers that are serious and knowledgeable about music there seems to be a pretty much universal consensus that analog still sounds better than digital. There are many reasons for this, and I will not go into them, but the most important reason is the incredible acuity of the human ear/brain system.
The human ear/brain system is particularly bothered by distortions that do not occur in nature, and unfortunately this includes many types of distortions in digital systems. For example, if you try to reconstruct a sound waveform from a digital system whose clock timing precision varies by something around one tenth of a billionth of a second (100 picoseconds), listeners immediately hear a reduction in spaciousness, hardening of timbre, a glassy unpleasant treble, a softening of the bass and a dramatic reduction in the realism of transients, such as a drum hit or a piano hammer when it first hits the string.
Please take a small pause and think about it, we are talking about differences of one tenth of one billionth of a second!
I would be willing to bet that you if mention this to the average person or to many "experts" they will think you are crazy, yet, you can get them in a room and have them listen and they will all hear the difference.
I also remember well the first time I told my friends that changing a power cable in an amplifier made a significant difference, and not only that, you had to break in the cable for a couple of months and the sound got even better. They were about to take me to see a Psychiatrist. That is, until I gave them a demonstration and they all immediately heard the difference.
To paraphrase the title of this article: Everything matters, and it is all about the small details.
The Wine Example
Many years ago I thought I had a great idea. The idea was so great, that it was going to make me very wealthy, and very famous. And, like all great ideas, it was "simple", and "brilliant". The idea was to make an excellent wine by mixing it from raw chemicals in a lab, as opposed to having to grow grapes and go through the whole process.
The way I was going to implement this was to purchase a number of bottles of the best and most expensive wines, then subject them to the most accurate chemical analysis available at the time (down to a few parts per million) and then simply mix the chemicals, bottle the stuff and market it. I had dreams of producing wines that tasted like the best of the best, that could be produced and sold at a small fraction of the price of the real thing.
Well, business issues aside (and there were plenty of those!), the human sense of taste alone killed the idea. Why? Because even mixing the chemicals with incredible accuracy could not fool the human sense of taste. It turns out that like most "great ideas", I was not the only one who thought of it, and UC Davis already had a study in process that quickly came to the conclusion that humans could immediately tell the difference between the mixed chemicals and the real thing. The mixed chemicals tasted worse than a bad cheap wine with a defective cork.
Again, everything matters, and it is all about the small details.
Leica S2 with 180 mm lens
Back to Photography
The above examples are extremely powerful in terms of showing that simply sticking to a few engineering or chemical measurements or a set of specifications does not even begin to address the complexities of the way humans sense and process information. If one only looked at the total harmonic distortion, signal to noise ratio and dynamic range that Sony claimed for its first CD player, one would assume impeccable music reproduction. Yet, that original box sounds like crap. Similarly, if one only focused (like I did) on chemicals down to a few parts per million, one would assume that the artificially made concoctions would taste exactly like the best wines they emulate.
Dead wrong on both counts.
This brings me back to photography. There are many photography enthusiasts, as well as professional photographers, experts, pundits and what not, who spend countless hours and thousands of pages pontificating about a few numbers and a few specifications regarding things such as lines per millimeter, or dynamic range of a sensor. They seem to believe that the final quality of a print delivered by a complete camera/lens/sensor/computer/software/printer system to the human visual system and the human brain can be determined by a few numbers. Well, I am sorry to say that anyone who believes that is awfully naive and bound to be dead wrong a lot of the time.
Just as our sense of hearing and our sense of taste are extraordinary, so is the human eye. Furthermore, the eye/brain combination gives rise to all kinds of psycho-visual effects that are also extraordinary. Bottom line, our eyes and our brains are so sophisticated and so incredibly sensitive and precise, that simply reading the specifications of a product does not even remotely begin to tell you the full story of what images from that product will look like to the eye, let alone what will happen once the brain kicks in.
This is why I always tell (and will continue to tell) everyone: DO NOT rely on engineering specs and marketing claims. Test a product and look at the final results. Sensors with exactly the same DxO dynamic range measurements can produce completely different visual dynamic range results, lenses with exactly the same resolution specs can look very different in terms of sharpness, and so on. Your eyes and brain should be the ultimate judge, not some numbers on a piece of paper.
Also, when you look at specs, you should think about what they mean. Case in point: Some years back I purchased a Nikon lens that had a fantastic sharpness specification (I do not mean to pick on Nikon specifically, they are a fine company and I could have easily picked a different example from a different brand). How many times have we heard: "Oh, that lens is really sharp" and then everyone automatically assumes that the lens is great? Well, the salesman confirmed that this lens was extremely sharp, so I believed the specs as well as the salesman and I bought the lens. Big mistake!
As soon as I started using the lens, I hated it. Why? Because first of all, we need to think in terms of what a sharpness spec really means: It means that under controlled conditions, on an optical bench, at the plane of focus a lens is sharp (the plane of focus is actually a surface and rarely a flat plane). And this means that in real world conditions, there is only one surface in focus. Everything else is out of focus. So, when we look at most images, they usually consist of more than 99% things that are out of focus and brought back to acceptable sharpness levels by depth of field. So, the problem with this lens was that it had an ugly look in the out of focus areas, even if slightly out of focus, and depth of field preserved this ugliness. Added to that, the color rendition was not good and the color saturation was poor. But the specs did not tell me any of that!
By the way, the lens was indeed very sharp at one surface. In fact, it was so good at this surface that it made things outside of this surface look even worse by comparison. Therefore, the extreme sharpness at the plane of focus actually worked against it! It made it feel even more out of balance.
Had I tested the lens with my own eyes, I would not have purchased it.
This is where the art of lens making comes in. It is about finding that nice balance that gives a lens not just outstanding performance across the board in real world shooting conditions, but also about achieving a beautiful aesthetic look. It is not all about computers and calculations. It is about the lens designer's knowledge and expertise, it is about finding the right glass and the right materials, it is about the quality of the grinding of each element, about the quality of construction, the quality of the coatings, the careful final alignment and assembly, the stability and durability in the field, but more than that, it is about testing not just with machines, but also with human eyes. This is why fine lenses are neither common nor inexpensive.
By now, you obviously know where I am going. As the above examples indicate, everything matters in photography, and it is all about the small details. Dismissing small details as irrelevant is a huge mistake. I hope to make some arguments and show further examples that will lead you to realize that it is precisely these small details that can make the difference between masterpieces and utter mediocrity. In fact, instead of getting bogged down in the details, we should celebrate them. We should enjoy them. As imaging enthusiasts and as passionate photographers, it is the small details that make us thrive.
But, before I continue with this train of thought, I think it is pertinent to spend a little time on some philosophical thoughts, as well as my personal approach to photography. While others may have a totally different approach, I believe all the main observations still apply, but I think that things will be much clearer if you understand where I am coming from.
The Theory of The Unseen
I have a theory. My theory is that one of the main characteristics of successful photographs is that they contain something that was unseen by the observer before the photograph was exhibited. Let me elaborate with a somewhat extreme example:
Let's assume that someone took a naked picture of Queen Elizabeth and published it (No offense to Her Majesty intended, this is simply used as an extreme example!). The image itself may not be the prettiest thing to look at, but I think we can all agree that it would cause a sensation and generate a huge amount of interest. Why? Two main reasons:
1. Because it is something heretofore "unseen"
2. Because it is something that most viewers never expected to be able to see.
If we think about it and go back in history, it is quite interesting to note that the very first pictures of horses showing that all four legs can be up in the air at the same time, the early pictures of wildlife, the pictures of the Civil War in the United States, the first pictures of distant lands and distant landscapes, or the very first pictures of bullets piercing through light bulbs and apples all possessed the above two characteristics. And they indeed generated an enormous amount of interest.
But let's continue to think about this:
I would make an argument that from photojournalism to landscape photography to portraits to commercial to architecture to sports to wildlife and all other kinds of photography, a key element of many of the best images in the world is that they depict something that was previously unseen, and they surprise the viewer by showing something the viewer never thought he or she would ever be able to see. It can be an amazing event or very unusual natural light or a lens and an angle that allows to see things in a different way, or freezing the action, or taking a 365 day exposure, or a view through a microscope at extreme magnification, or having a very unusual subject agree to be photographed, or finding a new animal species, or taking a picture from an amazing vantage point such as the earth from the moon or something inside the human body, or some other element that immediately screams: This was unseen until right now.
Therefore, whenever I am out in the field making an image, I always repeat to myself that the image must depict the unseen. It is amazing how this simple thought can improve and differentiate your images.
There is another way to show the unseen. It is also a way to dramatically enhance the psycho-visual impact of an image. I call it Hyper-Reality. I saw it for the first time when I went to an exhibit of Edward Weston's work. I had never seen large format photography before. The 8x10 inch contact prints in the exhibit absolutely blew my mind. Their impact was so profound that I could not sleep until I learned how to shoot large format. After that, I shrugged off the smaller formats and I shot pretty much all my work with 4x5 and 8x10 cameras until I made the switch to digital.
Later on I realized that one of the key reasons for this tremendous psycho-visual impact was that large format photographs, unlike smaller formats, showed a hyper reality that the naked eye normally cannot see. In other words, the extreme quality of the format itself was able to show the unseen in ordinary objects. To be more specific, the combination of a good lens and an 8x10 inch negative had such fine resolution that the prints showed details that were usually lost to the human eye. The format could also capture full detail in dynamic ranges that smaller formats could not. More importantly, with the use of rise/fall, shift and swings and tilts, the format could achieve depth of field far exceeding that of the human eye. Add to the combination the exquisite very fine tonalities that large format can capture, plus the lack of an enlarger in Edward Weston's images, and there it was staring me in the face: Hyper-Reality in its finest form: Being able to see the unseen for the first time in a sand dune, a nude, a toilet or a green pepper.
In other words, what I like to call Hyper-Reality is the ability to show detail, dynamic range, color, depth of field and fine tonalities that either exceed the human eye or are so far above what the eye/brain system expects that it has basically the same effect as seeing the unseen for the first time.
I would assume that it is partially because of these kinds of experiences in my youth and partially because I am primarily a landscape photographer that I have clearly gravitated to make Hyper-Reality an integral part of my photography. Therefore, I shoot almost exclusively Medium Format Digital with a technical camera. I need and I crave that additional quality that is missing in the smaller formats, and I will take the increase in image quality of a technical camera versus an SLR whenever possible.
Later in the essay, I will show that even on a computer screen, at very small size (600x800 pixels), the look of a Medium Format Digital image is totally different from a smaller format. This is the first myth that needs to be debunked. Even at very small sizes and even at low resolutions, Medium Format images look much better than the smaller formats.
When people ask me if I shoot Medium Format digital to make huge prints, or to spend lots of money, or to show off, or whatever other reason they may think of, my simple answer is:
No. I shoot Medium Format Digital in order to have Hyper-Reality be an integral part of my images.
Leica S2 with 180 mm lens
It Is Still Not Sufficient
The question at this point is as follows: If I can produce an image that depicts the unseen, and I add to that image a good dose of Hyper-Reality, is that sufficient to produce a truly good image?
And the obvious answer is no. It is not sufficient.
While images that show the unseen produce immediate interest, for many of these images the interest will decrease after a while and they will become more of a curiosity or a historical document. Very few will endure as true works of art. The ones that will endure as true works of art are the ones that in addition to showing the unseen have a certain degree of technical excellence, an artistic message, a personal style and all the other characteristics one expects of a true work of art. I must repeat myself again, everything matters; there are no shortcuts and no simple recipes.
For instance, among the huge volume of war photographs that have been taken through the years, there are very few that remain and endure as true works of art. But the ones that remain as such are truly breathtaking and they are indeed a combination of showing the unseen with all the other elements necessary to make them what they are.
Similarly, of the huge numbers of landscape photographs taken every year, there are very few that will endure as works of art and survive the test of time.
The same can be said for any other kind of photography, be it sports or fashion or product photography.
What I can say is that an image that does not show the unseen in some form, is highly unlikely to succeed. Particularly given the billions of photographs shot every year, such images are likely to end up in the oblivion pile.
The First Myth
One of the most common myths in my experience is that Medium Format Digital is a waste, particularly for relatively small prints, web images and so on. Related myths include statements that if, for example, you are going to show an image on a screen with a resolution of, say 1028x760 pixels, all you need is a sensor with the same number of pixels. Anything beyond that is a waste.
So, let's look at two images of very small size (600x800 pixels) on a screen. My wife had a birthday recently and she received some flowers, so for this simple test I took two quick pictures of the flowers with two different cameras and equivalent focal length wide angles. The images were shot almost simultaneously. I wanted to shoot almost simultaneously because they were shot under natural light and I did not want the light to change. This necessitated one of the cameras to be mounted just above the other, so there is a difference in the angle of the photographs, but this does not affect the observations below. Both cameras were mounted on very sturdy tripods.
The images received no processing adjustments. I exposed them as carefully as possible, did a direct conversion from RAW to TIFF with no adjustments and since both images had a lot more pixels than 600x800, I reduced the image size in Photoshop using the standard "bi-cubic sharper" algorithm.
I assume that many readers will be looking at these images on uncalibrated laptop displays, so here we have the extreme scenario: Small image size, only 600x800 pixels on a low resolution uncalibrated and reduced color gamut medium.
What are the differences I see between image A and Image B?
For starters, the lens in image A distorts a lot. The lens in image B looks much more natural.
What else? Image A seems to have a lot more contrast. Also areas of the pink flowers near the top and the sunlit white flower on the left side are completely blown out in image A, Image B preserves the tonality and the delicacy of the flowers and preserves full detail everywhere. Looking more carefully at image A, and comparing it to image B, it becomes quite obvious that image A does not have the color richness of image B. The big red flower at the bottom center looks dead in image A compared to image B. Also notice the vastly reduced color palette in image A. Image B shows many more shades of green, red, pink, etc., while at the same time capturing the full dynamic range of the image.
Looking at the shadows, image B has tons of detail; in image A the deep shadows simply go black. Finally, if you look carefully at the big red flower at the bottom center (where I focused), Image B looks sharper with much better edge delineation, and it also shows a lot more resolution of the fine details of the flower. Besides the higher resolution of detail, there is also higher resolution in the fine transitions from one shade of color to another. Image B is smooth and delicate. Image A is quite harsh by comparison.
Finally, the colors in image B are much more accurate.
In conclusion: There is a huge difference between Image A and Image B.
The punchline is that Image B was shot with a Medium Format camera. Image A with a smaller format camera. Let's momentarily put aside the fact that the Medium Format lens is much better. Image B is still clearly superior in every other area, and the visual impact of this is immediate.
(Note: Medium Format wide angles, particularly those for technical cameras are usually vastly superior to smaller format wide angles, as these two images clearly show. In my personal experience, the only exception to this general rule are the Leica M wide angles).
Is there an explanation? Yes there is. And just to make sure, I contacted a very experienced digital camera designer. I asked him the following two questions:
- Could he also see these differences?
- Did my explanation make sense?
The response I got was a resounding yes on both counts.
Not only are the differences visible to him and his staff, but they are also measurable with their lab equipment. I sure was glad to have technical verification from an experienced digital camera designer that Medium Format delivers better prints and better images on screen. He emphasized to me that they can clearly see and measure these differences at any size, from very small prints to very large prints particularly with images that have high dynamic range, lots of detail, large color gamut, etc.
Of course, one could always concoct some image with little detail, small gamut and small dynamic range that makes the differences extremely small and almost invisible in a small print, but in general, I think it is safe to say that one can expect to see quite noticeable differences between Medium Format and smaller formats.
The basic explanation is that Medium Format cameras have much higher quality pixels. There is less noise for each pixel, so an MF camera is able to record smaller transitions and has more precision in the measurements of what is happening in each pixel. This allows for an abundance of micro detail, as well as higher resolution and dynamic range. The higher signal to noise ratio allows for better shadow detail, it also means that more shades of gray and more gradations of each color can be resolved (leading to a more accurate color rendition) and so on. All of this information can be preserved when scaling the image, as long as the scaling is done correctly. Add to that a much larger number of Megapixels versus the smaller formats, and you also have a much higher resolution system (all of this assumes, of course that the lenses are up to the task, which they are if you make the right choices).
When scaling to much smaller sizes, this abundance of information still produces images that are far superior. At a simple level, this can be understood by just thinking about, say, averaging the values of a number of the original pixels for each pixel in the final image. It is fairly evident that averaging many original pixel values will change the final values versus averaging only a few original pixel values. Because of the higher quality pixels and the larger number of pixels of a Medium Format sensor, the number of values used for averaging is much larger. In other words, the final image will be different and these differences translate perceptually into higher sharpness, more resolution, better color and all the other advantages we mentioned above.
I should mention at this juncture that there are many third party software products that perform scaling with much more sophisticated algorithms than the bicubic algorithm. Perhaps the most famous one of these is Genuine Fractals, but there are many others. Often times, when using a more sophisticated algorithm to downscale a Medium Format image, the visible superiority of Medium Format is even more pronounced.
The bottom line is that Medium Format Images are superior to smaller format images regardless of the size of the print or the image on the screen.
First myth? Busted!
Summary of Observations and Recommendations
There are a few things that I hope you will always keep in mind after reading this article:
- Everything matters. It is all about the small details. Celebrate these details. They are what makes photography so exciting.
- Do not ignore details. Even the smallest ones can be crucial. The weakest link in a chain will always determine the ultimate quality of the entire chain.
- Do not rely on specifications, measurements or marketing claims. Trust only your eye/brain system.
- Search for the unseen. The first step in creating a great image is to show something heretofore unseen.
- It is not only the subject matter that contains the unseen. It can be a special angle, a special view, unusual lighting, a distinct vantage point, Hyper-Reality or something else.
- Prints and screen images from larger format captures always look better, regardless of the size of the print or the screen image.
- Avoid myths. This is why I busted the first one and will bust a few more in the second part of this essay.