The process and gear
Vision
This is where the art begins. I spend days on the microscope exploring objects searching to find a composition that resonates. It may be aesthetic, emotional, visceral, harsh, gentle, or simply intriguing - anything within that range is fair game. I imagine how the shot will look, sketch out the composition in my mind, and consider the lighting. From this point forward the process becomes technical, but the art still guides every decision.
Subject Prep
Once the shot is envisioned, the image construction process shifts to preparation. That might mean polishing the surface of a rock or carefully washing and positioning an insect. All of it serves the original vision. At these magnifications every step happens under a microscope. I even use a fine airbrush as a kind of power washer.
Photon Selection
Because I work at the quantum limits of reflected-light microscopy, the choice of light is critical. I use light sources that can extend into deep ultraviolet or emit at a single wavelength. This control over the light itself is essential. For safety I usually block UV, but some subjects fluoresce under UV, and that glow can subtly enrich the colors, like sunlight in its full spectrum. When it enhances the image, this becomes a step that cannot be skipped. I use the science not for its own sake, but always in service of the art.
Lighting
Lighting is as crucial here as in any form of photography, but I can't just place a two-by-three foot soft-box in front of my subject. Instead, I work with fiber-optic light guides and custom diffuser hoods. Space is tight, especially when complex lighting is required. Diffusion, blocking, polarization, and fill light all have to be planned in advance and executed with precision to realize the image I first imagined.
Optics
The choice of optical elements in the light path is primarily driven by the physical size of the composition I first envisioned. My goal is to capture the most detailed rendering of that idea. Many subtle factors matter like depth of field, numerical aperture, and more. Unlike conventional lenses, microscope objectives do not zoom, they are the equivalent of fixed prime lenses. Each one is chosen to meet the exact requirements of the shot.
Calibration
Every lens I use goes through a calibration process to measure and correct chromatic aberration, field distortions, and depth-of-field shifts. Beyond that, each shoot involves unique lighting conditions with its own spectral profile. I cannot rely on a standard color card at these scales - on paper it would reduce to dots on fiber. Instead, I use a specially made microscope-scale color standard to guarantee color accuracy.
Photon Collection
My Brownie 2A was impractical for this task - and so were many modern cameras. I experimented with many cameras from all the major manufacturers. The critical requirements were that there cannot be a mechanical shutter and the sensor must remain stable and consistent over time and temperature. Before each shoot, dark-frames are collected to identify the hot or dead pixels which every sensor has. Also, a precision nano-scale grey target is used to balance regional tonal variations of the sensor, and then an overall scan of all individual panes is made to determine the final exposure for the shoot.
Click click click click...
I built an automated rig to capture the thousands of individual frames needed for panoramic microphotography. The sequence includes depth scans for focus stacking and a grid of movements that defines the imagined composition. Image collection can last for days, and the shutter count can exceed 220,000 frames (my broken cameras prove it). Long shoot times demand light sources of the highest stability that are unaffected by fluctuations in grid power.
Supercomputers
Here is where it gets a little bit crazy. Processing all those frames on a typical workstation would take years. The only way forward is supercomputing. The workflow includes transfer, renaming, collation, conversion, point-spread function derivation and deconvolution, median sequencing, and more. These methods draw heavily on techniques developed in astrophotography and microscopy research. Then comes the actual focus stacking and panorama stitching. If everything holds together, the result is a massive and faithful realization of the original vision. The blue glow is me applying a magical enchantment to the otherwise ordinary supercomputer hardware.
Postprocessing
I do very little post-processing on my images. Careful planning before the shoot usually eliminates the need. Even if I wanted to, most software cannot handle images stretching to 180,000 pixels wide. However to compensate for specular highlights I often need to underexpose my shoots. To make final adjustments, I prepare scaled-down versions, generate HALD CLUTS from them, and then apply those profiles during a grid-based teardown and rebuild of the final image. I never alter or saturate colors - the subjects already provide everything. Only exposure curves and occasional crops are applied.
Printer Calibration
It is essential to me that the print matches what I see on screen. I meticulously calibrate my printers (Canon IPF8400 and IPF9400) to ensure the truest possible color. Each paper type requires its own calibration, and I repeat the process whenever new ink is installed to account for any potential ink formulation changes.
Paper
For large prints I almost always recommend face mounting. For that I use high-grade RC gloss papers from Fuji or Canon. Some works call for metallic papers, while others are best on a flat matte paper. For oversized prints not face mounted, I suggest satin finishes to minimize glare - when a piece fills an entire wall you can't simply shift your viewing angle, and a glossy surface also amplifies imperfections in mounting or framing. Still, if a glossy finish matches the collector's vision, I provide it. All paper is archival quality, paired with archival inks, and I stock a full range of paper types in all of the sizes that I print.
Mounting and Framing
At this final stage, the scale of the work poses real challenges. The 36-inch-tall version of the shot to the left stretches nearly 29 feet wide. Anything under four feet on the long side can usually be handled and shipped conventionally, but larger works often need special handling and the largest almost always need to be paneled, especially in museums or commercial spaces. Mounting and handling a single 12-foot-long flat piece is difficult, but I have done it. I work closely with clients to find practical solutions and ensure the installation succeeds.