Computational Photography for the Purist: Augmenting Reality, Not Replacing It

The term ‘computational photography’ makes most purists flinch. It conjures images of fake skies and plastic-looking portraits from a smartphone. It feels like a betrayal of the craft, a shortcut that bypasses the hard work of understanding light and optics. For over 15 years, I’ve built my career on the principle of getting it right in-camera. My time in the print shop taught me that what you capture is what you have to work with; you can’t invent detail that isn’t there. So you can imagine my skepticism.

But not all computational photography is a lie. It’s not one monolithic evil. It’s a spectrum. On one end, you have digital fabrication—creating something from nothing. On the other, you have augmentation—using processing power to overcome the physical limitations of a sensor and lens to better capture reality.

That’s the line. That’s where a purist can engage without selling their soul.

Augmentation: The Tool That Serves the Truth

Think about the techniques we’ve accepted for years. Is focus stacking a betrayal of the art? I do it all the time for macro work with my Sigma 105mm f/2.8. I’ll take a dozen shots at different focal planes and merge them to get a depth of field that is physically impossible in a single capture. The final image, however, is a more truthful representation of the subject than any single, shallow frame could be. The computation serves reality, it doesn’t invent it.

The same goes for intelligent HDR. I’m not talking about the garish, over-processed HDR from a decade ago. I’m talking about bracketing exposures to capture the full dynamic range of a scene that exceeds what my Nikon Z6 III’s sensor can grab in one go. The camera merges them to create a file that contains the shadow detail I saw with my own eyes and the highlight detail that was also there. It’s a technical solution to a hardware limitation. It augments the capture to better match the real world.

This isn’t about making things easy. It’s about making things possible. The goal is to faithfully render the scene in front of the lens, not to build a new one from a menu.

Even the advanced autofocus in modern cameras like my Z6 III is a form of computational photography. It uses processors to identify and track a subject with a speed and precision the human eye can’t match. It’s not faking the image; it’s helping me nail the decisive moment. This is technology in service of the craft.

Replacement: The Crutch That Creates a Lie

The other end of the spectrum is where I get off the train. This is the world of one-click sky replacements, AI-generated people, and algorithmic portrait “enhancements” that scrub away every pore until the subject looks like a porcelain doll. This is not photography. It is digital illustration masquerading as photography.

When you replace a sky, you are replacing the light. You are fundamentally altering the entire context and mood of the image in a way that is dishonest. The light on your subject no longer matches the light in the sky. The narrative is broken. You’ve created a composite, not a photograph.

This is the core of the problem. These tools encourage photographers to ignore the fundamentals. Why wait for golden hour when you can just drop in a stock sunset? Why learn to use a shallow depth of field with a fast prime lens like my Sigma 24mm f/1.4 when your phone can just blur the background with an algorithm? It devalues skill and replaces it with a cheap trick.

The Ansel Adams Parallel

This debate isn’t new. Think of Ansel Adams and his Zone System. He was a master technician who used extensive, complex darkroom techniques—dodging, burning—to craft his final prints. No one would call his work “fake.” Why? Because his manipulations were done to realize his vision of the reality he stood in front of. He was augmenting the information on the negative to match the grandeur he witnessed at the scene. He wasn’t adding mountains that weren’t there.

That is the mindset we should adopt. We can use computational tools as a modern darkroom to bring our captures closer to the truth of the moment. We can use them to solve technical problems, but we must stop when they start telling lies.

The camera is a tool for capturing light and time. The moment we ask it to invent light and fabricate moments, we’ve stopped being photographers.

Where I Land

• Augment, Don’t Fabricate: I’ll use tools like focus stacking and legitimate HDR to overcome hardware limits and capture more of reality. I will never use a tool to invent elements that were not there.
• Technology Should Serve Craft, Not Replace It: Computational aids like advanced autofocus help me execute my craft better. Tools like sky replacement encourage me to ignore it. I choose the former.
• The Goal is Truthfulness: Even a heavily stylized image should be truthful to the moment and the subject. Computational photography becomes a problem when it prioritizes algorithmic perfection over authentic storytelling.