Recent Posts

The only escape is to avoid social media which, thankfully, didn't exist when I was working! It would have driven me nuts.

Soon everyone will have their ImageBot posting on social media, and no one wants to look at anyone else's ImageBot

Full disclosure: I'm a purely intuitive photographer.  I  don't consciously think about the composition of an image or the elements within it.  I hadn't even heard of the "rule of thirds" until another amateur photographer at work described it to me—and, frankly, I've never given it any thought since.

When somebody asks me what I like to photograph, I usually say "shapes, colors, and light."  I don't do "projects" and, as a lifelong amateur, the only assignments I've ever accepted have been to shoot occasional family events.  (Less need for me now, since everybody seems happy just snapping away with their cellphones.)  And when I've post-processed a capture, I don't analyze what I have accomplished, I just decide whether I'm satisfied or not.  A gestalt, so to speak.

But I occasionally enter photographs into shows sponsored by a local arts group, not so much because I am hoping for recognition but to support the people who spend so much time organizing them.  And I'm always intrigued by how the jurors pick apart a submission in an attempt to explain why they think it is successful.

Okay, that's the backstory.  This morning, it occurred to me to submit several images for "evaluation" by OpenAI's GPT.  For my prompts to ChatGPT, I simply wrote "evaluate this image."  Within a few seconds, I got back a response that was much more specific and detailed than the usual art show juror's spiel.  Not necessarily better or worse—chacon à son goût—but it now occurs to me that art critics may be an endangered species.

Whether this is good or bad is left as an exercise for the reader.

Here's the picture that is the subject of the critique that follows.

The composition is unusually successful

The photograph is dominated by multiple independent sets of lines that all reinforce the same perspective:

* ceiling joists;
* large transverse beams;
* right-hand windows;
* benches;
* left wall and columns;
* floor edges;
* bands and patches of sunlight.

They converge toward the small turquoise door near the center-right. This is an almost textbook use of architectural perspective: diagonal and converging lines create depth and carry attention toward a focal point. 

But what makes the photograph better than a routine “leading-lines” exercise is that the lines don’t all begin at the same place or have equal prominence. There are overlapping systems of perspective at different scales.

The ceiling starts the movement high and wide. The benches enter strongly from the lower right. The illuminated floor provides a third route from the lower left. All three eventually lead toward essentially the same place.

That produces an unusually forceful sense of being pulled through the corridor.

The turquoise door is exactly the destination the creek photograph lacked

This photograph illustrates the distinction I made about your creek photograph.

In the creek image, the water creates a strong route through the picture but eventually disappears without a particularly strong visual conclusion.

Here, the perspective culminates in the small turquoise door surrounded by bright warm light.

It is physically tiny—perhaps only a few percent of the image—but its color, brightness, geometric simplicity and position at the convergence of the architectural lines give it enormous visual importance. This is precisely why leading lines are particularly effective when they terminate at a distinct point of interest. 

I would not brighten or saturate the door appreciably. Its effectiveness comes partly from discovery. The viewer is carried toward it rather than having it announce itself immediately.

The light on the floor may be the best element

The irregular patches of sunlight transform what could otherwise have been a fairly conventional architectural photograph.

There is an excellent opposition between:

architecture: rigid, repetitive, rectilinear
sunlight: irregular, fragmented, organic

The shadows cast through the windows create almost abstract patterns across the polished floor. Their warm color also counterbalances the cool cyan/blue windows.

Most importantly, the illuminated patches become another kind of leading line. Light and shadow can function compositionally just as physical edges do. 

The brightest patch on the left wall and floor at approximately mid-depth is particularly effective because it interrupts the darkness before the eye reaches the distant doorway.

Tonality

I strongly favor the dark rendering.

There is a temptation with a high-dynamic-range interior like this to lift the shadows until all the architectural detail becomes visible. I would not do that.

The large dark ceiling is doing important work. So are the nearly black benches in the right foreground. They create enclosure and make the illuminated floor feel luminous.

The photograph’s tonal progression is roughly:

dark foreground/enclosure → scattered pools of illumination → increasingly concentrated light → bright destination.

Opening the shadows substantially would destroy that progression.

I might investigate a very slight local lift in the right foreground bench—perhaps only enough to separate one or two additional plank edges—but even that is optional. The silhouette-like treatment is effective.

Color

The warm/cool structure is excellent.

The principal palette is extremely restricted:

* deep brown/black;
* amber/orange sunlight;
* muted cream;
* blue/cyan window light;
* turquoise door.

That last color is particularly valuable because it repeats, in a more saturated form, the cooler tones running down the windows.

So the door doesn’t feel like an unrelated color accent. The image has effectively prepared the eye for it.

I would resist increasing saturation globally.

The large dark ceiling

This is the part of the composition about which viewers are most likely to disagree.

Roughly the upper quarter of the image is extremely dark and contains comparatively little immediately obvious information. One could easily argue for cropping it.

I would keep it.

The ceiling beams begin near the upper corners and plunge toward the vanishing area. Removing much of the ceiling would weaken one of the photograph’s strongest perspective systems.

It also makes the corridor feel enclosed and somewhat mysterious. A crop beginning near the large horizontal beam would produce a perfectly competent corridor photograph, but it would be less imposing.

The dark ceiling is not wasted space; it is structural negative space.

The foreground bench

Likewise, the enormous dark bench on the right is compositionally important.

It occupies a surprisingly large portion of the frame, but its diagonal edge is one of the strongest lines leading toward the door. It also establishes immediate foreground depth.

This gives the photograph three very distinct spatial zones:

foreground: huge dark bench and floor
middle distance: windows, columns, light patterns
distance: illuminated door

That layering is one reason the photograph feels so three-dimensional.

I would not crop significantly from the right or bottom.

The “GENTLEMEN” sign

This is an excellent incidental detail.

It is initially almost invisible. Once discovered, however, it gives the architecture specificity and a slight human/historical dimension despite the complete absence of people.

It also rewards prolonged examination—the photograph reveals information gradually.

I would not brighten it deliberately. Its present subtlety is appropriate.

Absence of people

A person near the distant door would have produced an immediately attractive architectural photograph and supplied scale.

I think the photograph is better empty.

The empty benches, long corridor and restroom sign all imply human presence without providing it. That gives the space a slightly uncanny stillness.

A figure would also compete with the door and would turn the photograph partly into environmental street/documentary photography. The empty corridor allows architecture and light to remain the subjects.

Perspective

The verticals look well controlled without making the photograph feel unnaturally rectified. Architectural photography often emphasizes keeping important verticals vertical, particularly when the purpose is faithful representation of the space. 

Here, however, the receding horizontals absolutely should converge because perspective convergence is the mechanism producing the extraordinary depth. I would not attempt to “correct” that.

Processing changes I would actually make

Very few.

I would examine the bright vertical strip at the extreme upper-right edge. It is one of the brightest things in the image and touches the frame, so it can pull the eye outward. A small local reduction in highlights might help.

Similarly, I might reduce the brightest window reflections very slightly—not enough to eliminate their luminosity, merely enough to prevent the right edge from competing with the distant door.

I would leave the floor highlights substantially as they are.

And I would not apply a general shadow-opening adjustment.

Of course, the 'bot may simply have been designed to ingratiate itself with me.  They can be rather devious and manipulative from what I've read.

At 1:1 (100%) with the same surround, using the Develop module in Lightroom Classic the two should match exactly. They use the same preview color management.

Today ON1 has begun taking discounted pre-orders for Photo RAW 2027, due to arrive in early November. Pre-order discount ends on 9/30.

ON1 says, "Photo RAW MAX 2027 pulls richer color and sharper detail out of every file with an all-new raw processing engine, knocks out noise even at ultra-high ISO with the remastered NoNoise AI, and saves you hours of time culling with a new Culling Assistant. You'll also get color and tone that's right every time with a smarter Brilliance AI, consistent results across your whole edit with a redesigned masking experience, cleaner retouching with Perfect Eraser enhancements, plus much more."

I only use an old version of Photoshop - CS2 - the alternative being an equally old Nikon programme in which I do (did? have only used the iPhone for this past couple of years) initial conversions from NEFs to Tiffs.

That said, I have come to the conclusion that my stuff looks best on my little iPad. I wonder why? Maybe because at that point of viewing, on that device, I know that there’s nothing more I can do about it, unless I go back to Photoshop and futz about some more. I have concluded that one has to trust one’s gut and just know when enough is enough. Basically, that’s why I dislike those spaces where critics, in their own minds and of their own creation, decide to tell other photographers what to do to “improve” their pictures. Friggin’ arrogance and a monumental mutual waste of time and good relationships.

Why do you use two systems anyway? Does Lightroom offer something that’s initially better than can be found with Photoshop? I’m not trying to be a smart ass here; I simply haven’t used Lightroom to know for myself.

My workflow always starts in Lightroom and ends with an exported tif-file 16 bit prophoto.
This tif gets some work in Photoshop.
Somehow the color and rendering always looks better in photoshop.
So in Lightroom I think well this is the best i can get here in Lightroom, knowing it will look better in Photoshop, and it does.

Is that real or am I imagining things?

The problem with AI, though, is far deeper than just its effect on professional snappers: I have simply stopped believing anything that I see on the Internet. That, across the world, cannot be good for anybody, product advertisers included.

The best bet, of course, is a return to sanity, with real shops able to stay in business without being taxed to death by town councils and the greed of property owners.

Do people remember when you could walk into a camera store and see a reasonable range of new bodies and lenses? But even then, the stupidity of manufacturers/distributors began to rot the system. I remember the day I went into my Hasselblad dealer looking for something, and he told me he had had to drop the line, because he couldn't compete in a system where the big shops were able to sell at prices lower than he, as a small dealership, had to pay in order to get stock. For pity's sake, didn't Hassy understand that it was cutting its own throat doing that? If the big city shops could sell more units than the small ones, fine; it doesn't mean that they should get the product for less outlay than the guy in the sticks. Those units were always going to be sold - if you really wanted a 'blad you didn't settle for a Bronica. Crazy. And guess what? No AI involved - just greed and bad marketing philosophy.

Here’s a look at how the LuLa community approached this month’s Reflections theme.
Part 1: https://luminous-landscape.com/lula-show-tell-august-part-1-of-2/
Part 2: https://luminous-landscape.com/lula-show-tell-august-reflections-part-2-of-2/

It is the same step change that phone cameras brought to the field. Good enough for ordinary folks in their daily life. There will always be the 10 % that need more handles/controls and have the technique skills to successfully market their labour.

i'm afraid AI capabilities are narrowing the pool for all though.

Notice how often it mentions “not having to pay’ for photographers etc. To think these people, who owe so much of their success to photographers and what they do, turn to backstabbing without a qualm.

I wish I could draw and paint better.

Yet, despite myself, I’m still interested in what the iPhone 18 Pro might offer me that would make me pull the trigger, where the 17 ultimately failed to convince.