What the Fujifilm GFX 100 II Gets Right (and Wrong) for Insect Macro

I spent most of a June evening lying on a damp riverbank trying to get a robber fly to stay still long enough for the Fujifilm GFX 100 II to lock focus, and by the time I packed up I understood something that the spec sheets never mention: medium format and insect macro are not natural friends. They can work together, and when they do the files are extraordinary, but getting there means fighting the sensor's own physics more than you would with a smaller, faster system. This isn't a review. It's an account of what actually happens when you point a 102-megapixel medium format body at something six millimetres long that does not want to be photographed.

The depth-of-field problem nobody warns you about

Bigger sensors give you shallower depth of field at equivalent framing, and at macro distances that difference stops being a stylistic choice and starts being a practical obstacle. On a Micro Four Thirds body at 1:1, you can often stop down to f/8 and keep a decent slice of a beetle's face in focus. On the GFX 100 II's 43.8 x 32.9mm sensor, the same framing at the same f-stop gives you a noticeably thinner slice, maybe a third of the depth, because the physical aperture diameter needed to hit f/8 is so much larger relative to the subject.

In practice this means that at true macro magnifications, an eye and an antenna are rarely both sharp in one frame unless you stop down hard or stack. I've had single frames of dragonfly heads at f/11 where the near compound eye is tack sharp and the far one has already started to soften, on a subject that's maybe eight millimetres deep. That's the trade you're making for the resolution and the sensor's tonal range. It's not a flaw in the camera, it's just a fact of the format that a lot of photographers moving up from APS-C don't budget for.

Living with the GF120mm, Fujifilm's only real answer

Here's the part that surprises people: Fujifilm's own GF120mm F4 Macro R LM OIS WR only reaches 0.5x magnification natively, not 1:1. That's half the reproduction ratio that most dedicated macro shooters expect from a "macro" lens, and it matters enormously for insects, where you often need to fill the frame with something the size of a grain of rice. Compare that to Canon's RF 100mm F2.8L Macro, which does 1.4x, or the Sony FE 90mm F2.8 Macro G OSS at a true 1:1, and the GF120mm starts to look like it's playing a different game entirely.

What that 0.5x limit does in the field is push you toward one of two workarounds: crop hard in post and lean on the sensor's 102 megapixels to survive it, or add extension tubes to physically increase magnification past what the lens does on its own. Both work, but neither is free. Cropping means you're throwing away the resolution advantage you paid for in the first place. Extension tubes cost you light (roughly a stop to a stop and a half at the lengths you'd need for a meaningful magnification bump) and they kill infinity focus, so you're committed to close work only until you take them off again.

Extension tubes and the adapter route

Fujifilm doesn't sell a teleconverter rated for the GF120mm, so extension tubes are really the only supported way to push past 0.5x. I run a 45mm Fujifilm-compatible extension tube behind the GF120mm for most jumping spider and small beetle work, which gets me to roughly 0.85x. It's still short of true 1:1, and the working distance shrinks to somewhere around 9cm from the front element, which is close enough that skittish subjects like hoverflies often bolt before you're set. Some shooters go further with stacked tubes or an older adapted lens (bellows-mount enlarger lenses reversed onto a GFX adapter show up in a few macro forums), but at that point you're troubleshooting an improvised optical chain rather than photographing insects, and I'd only recommend it if you already enjoy that kind of tinkering.

How the GFX 100 II actually compares for this specific job

Numbers help more than opinions here, so I put together what I've measured and experienced across the systems I've either owned or borrowed for macro sessions over the past year. Treat the depth-of-field column as a practical, field-observed approximation rather than a lab measurement; it's consistent enough between systems to be useful for planning, not precise enough to argue about in the comments.

What jumps out is that the GFX combination needs the fewest frames per stack, largely because you're starting from a shallower base magnification, not because the depth of field is more forgiving. Once you add the extension tube and actually reach comparable magnification to the other systems, the frame count climbs to match theirs, and the working distance drops to something that makes flighty insects harder to approach. The OM System body, for what it's worth, is the one I reach for when I specifically want live, uncooperative insects and don't have time to set up a stack. It's smaller, faster to focus, and the crop-sensor depth of field is genuinely kinder at high magnification.

Where the GFX 100 II actually wins: stationary subjects and resolution headroom

The camera stops being a liability the moment your subject holds still, which happens more often than you'd think with insects if you shoot at the right hour. Damselflies and bumblebees are close to immobile for the first twenty to thirty minutes after direct sun hits them, while they're still too cold to fly. That window is when the GFX 100 II earns its keep. The built-in focus bracketing mode lets you set a step size and frame count and it will rack focus through a full sequence automatically, which I've used to pull off 35-frame stacks of a torpid damselfly's whole thorax and wings without touching the camera between frames.

The other genuine advantage is that 102 megapixels gives you room to be wrong about framing. If a mantis is slightly too far away and you don't have time to recompose before it moves, you can back off, shoot wider than you'd like, and crop in post to something that still holds detail a 24-megapixel body couldn't match after the same crop. I've salvaged shots this way that would have simply been "too loose" on smaller sensors. It's not a substitute for good technique, but it's a real safety margin that changes how aggressively you can chase a subject.

Diffraction sets in earlier than the sensor size suggests

This one catches people off guard. Medium format sensors are physically large, so there's an assumption that diffraction is less of a concern than on a full-frame or APS-C body. It isn't, because what matters for diffraction is pixel pitch, not sensor size, and the GFX 100 II packs its pixels almost as tightly as a high-resolution full-frame sensor. In practice I start seeing pixel-level softening creep in around f/11, and by f/16 it's obvious on a full-size export, even though f/16 would look perfectly clean on a lower-resolution body of the same physical dimensions. For macro work, where you're constantly tempted to stop down for depth of field, that's an uncomfortable ceiling. It pushes you toward stacking multiple frames at a moderate aperture like f/8 or f/9 rather than trying to brute-force depth with one frame at f/16 or narrower.

Flash, diffusion, and the weight tax

A GFX 100 II with the GF120mm and a tube attached is not a light rig, and that matters more for insect work than for landscapes, because you're often holding the camera one-handed while positioning a diffuser or a reflector with the other, crouched at an awkward angle for several minutes at a time. I use a small twin-flash bracket with a folded diffusion panel rather than a ring flash, since a straight ring light tends to flatten the texture on chitin and wing veins that's usually the whole point of shooting insects this close. The GFX 100 II's mechanical shutter syncs at 1/800s, a real jump from the 1/125s ceiling on the older GFX100S, and dialed down two stops from full power that's my starting point for anything in open shade; direct sun days I mostly skip flash entirely and rely on natural light with a small reflector card to fill shadows.

What a stacking day does to your evening

The unglamorous part of this whole approach is what you're left with afterward. A single productive morning of focus-bracketed insect shooting on the GFX 100 II routinely leaves me with 800 to 1,200 raw frames, most of which are near-duplicates from a stack where only the plane of focus shifted a fraction of a millimetre between shots. Sorting that manually, frame by frame, at full resolution on files this large is genuinely painful, and it's the reason I moved my culling step to imagic rather than scrolling through Lightroom's grid view for two hours every night.

Because imagic's sharpness and focus scoring runs locally against the actual pixel data rather than a thumbnail, it's reliable at telling me which frame in a five-shot burst of the same damselfly is the one where the eye is genuinely sharpest, which matters when the differences are this small. It also groups the burst and stack sequences automatically so I'm not hunting through a flat folder of a thousand thumbnails trying to remember which fifty belong to which stack. Everything runs offline on the laptop I bring into the field, which matters when I'm reviewing shots at a campsite with no signal to speak of.

If you want the longer explanation of what that kind of automated culling is actually doing under the hood, we've covered how AI photo culling works in more detail elsewhere on the site. For anyone shooting this much volume regularly, it's also worth reading through some general habits for cutting review time down, which we laid out in 10 tips for a faster photo workflow.

The other thing that's changed my evenings is training imagic's apply_my_style preset on a batch of insect macro edits I was already happy with. Once it learned my usual white balance correction and the contrast curve I lean on for wing translucency, it applies something close to my own edit as a starting point on new imports, which saves a real amount of time when you're editing forty keepers from one outing instead of five. It's a one-time purchase rather than a subscription, which mattered to me since this is a hobby-adjacent workflow tool I didn't want another recurring line item for; the details are on the pricing page if that's relevant to how you budget for software.

Frequently Asked Questions

Is the GFX 100 II actually a sensible choice for insect macro, or is it just a novelty?

It's sensible for a specific kind of shooter: someone photographing mostly stationary or slow subjects (dew-covered damselflies at dawn, resting butterflies, spiders in web centers) who values maximum resolution and tonal range over speed and reach. If you're chasing fast-flying, skittish insects handheld all day, a smaller, faster-focusing system with a true 1:1 or better macro lens will frustrate you less and put more usable frames in the folder.

Do I need a dedicated GF macro lens, or can I get away with extension tubes on a standard lens?

Extension tubes on a non-macro GF lens will get you closer, but they don't fix the optical issues (softness, chromatic aberration) that a purpose-built macro lens like the GF120mm is corrected for at close range. Tubes are a reasonable way to extend what the GF120mm already does, but they're a poor substitute for a proper macro lens if you don't own one already.

How much does the shallow depth of field actually cost me at high magnification?

At true macro distances with this sensor, expect usable depth of field measured in fractions of a millimetre once you're past roughly 0.6x magnification, even stopped down to f/11. For anything with visible depth (a beetle's back, a spider's whole body), that means focus stacking is close to mandatory if you want the entire subject sharp, rather than an optional refinement.

Is focus stacking mandatory, or can single frames work for live, moving insects?

Single frames work fine for flat or side-profile subjects where the plane of focus can align with most of the insect's visible surface, like a butterfly's wing shot straight-on. For anything with real depth facing the lens, like a mantis head-on or a beetle from a three-quarter angle, a single frame will usually leave part of it soft. At that point your honest options are stacking it properly or accepting the softness as part of the shot.