If you hold it and shift it around so the light reflects off the glass, the surface of a Glimmerglass filter looks like it has been sprinkled with some sort of metallic dust, maybe gold. The particles aren’t really on the surface, they are sandwiched between two thin layers of glass, and they probably aren’t gold, though for the price of the filter I’m thinking maybe they should be.
It’s a diffusion filter, it goes on the front of the lens. You put it on your super-sharp, modern lens and you make pictures.
Why would you do that?
There are several things a diffusion filter will do but the primary thing that it does is add character. Designers of lenses for still cameras work hard to create lenses that are super-sharp, have little vignetting, and make colors pop. They have no choice since the Internet forums are filled with people—which is to say, customers—seeking the next increment in sharpness, the next justification for their next purchase.
“Not sharp” is not a neutral statement identifying a characteristic of a lens. It is a judgment of that lens, a dismissal of that lens.
And sharp they are. Lenses are judged now by enlarging an image on a screen to 100%—image pixel per screen pixel—at a size and a viewing distance few images are ever seen at—and the best lenses pass the test, exhibiting a crystal clear resolution that can bring childish mirth to your face. It’s fun, in a way, to see that little detail in the photo now clearly visible after tapping the zoom-in button. It makes you want to zoom in on every image, test every lens to see how sharp it is under maximum magnification.
If you are not careful the lens’s sharpness can become the true subject matter of your images.
In the filmmaking world you don’t have that same pressure to find the ultimate in sharpness. A moving image is extraordinarily forgiving of what would be unthinkable in still photography. A little unsharpness passes without comment in a movie while the still frame from that image would be cast to the darkroom floor. In the film days it was not uncommon for focus in a motion picture to be off a bit now and then (it’s still not as rare as you might expect). Even big-budget films went to print with misfocused scenes.
This isn’t surprising. Focusing a movie camera is a very different affair than focusing a still camera. On a still camera you can often let auto-focus do the work, monitoring it and correcting any mishaps, deleting the bad frames later. If you focus manually you carefully go back and forth, zeroing in the precise rotation of the focusing ring.
In a movie it is not as common to use autofocus—it’s just not good enough yet. How it works in many scenes is that the actor will rehearse their scene and the photography team will set certain marks for that actor to hit—points which the person focusing the lens, the focus-puller (that’s a specific job) will know beforehand where to focus the lens, how to change the focus during the scene. On a lens for video or film you can put marks on the lens, too, that match the marks the actor is trying to hit. This series of marks can be quite complicated and a lot can go wrong. And if you are shooting with a handheld camera or working on a project that is more spontaneous with little or no rehearsal plus shooting wide-open for those blurry backgrounds then you can see how all sorts of errors can creep in.
That’s why the errors occur but why do they keep them? Why doesn’t the director of the film just reshoot the scene, especially with a big-budget film? The reason is that sharpness is not the be-all, end-all deciding factor in the success of a scene. There are a host of things going on and a little unsharpness can be easily overlooked if the overall sense of the scene is powerful, if the actors did something really special, if all of the complex camera movements, lighting, everything, all came together to cast some magical spell, some primordial connection between the filmmaker and the audience. Or sometimes filmmakers just didn’t notice (they review scenes during the shooting on monitors that are much smaller than the big screen) or they noticed but they’d already moved on to something else.
Feeling is what matters, more than technical perfection. Sometimes ultra-sharpness contributes to that feeling. Sometimes a softness in the image contributes to that feeling. That’s where the Glimmerglass filter comes in. This filter, made by Tiffen, is one of a number of diffusion filters that do not simply “fuzz” the image but do something far more interesting. I bought one in a size to fit the new fast Fuji 50mm f/1.0 and thought it might be worthwhile to shoot this lens, which already has a sort of diffusion look built-in when shot at f/1.0, and to see what the combination would produce.
This look is going to look familiar to you. If you watch Netflix or AppleTV+ or Hulu or any of the rest you have seen diffusion, and sometimes lots of it. It’s everywhere as filmmakers turn to older lenses and other techniques to both add character to their work and to differentiate themselves from other TV series. The result, to my eyes, is that there are an awful lot of similar-looking shows out there, all with the missing reds, various intensities of glow around the light sources, flare and whatnot from the lenses.
Take The Handmaid’s Tale—I’m finally watching season two. It looks like almost every interior scene has a diffusion filter (the lights and windows are all aglow) and they are using an old Canon lens (I had to google to learn the particular lens) which shows all manner of colored rings and flares, sometimes as tall as the screen, often used by the filmmaker as compositional elements in the frame. Layered on top of that, the interiors tend to be mist-filled, so that sunbeams and truck headlights blast columns of light through the air.
It’s fine, it’s an interesting look, it just starts to sort of blend after a while. Filmmakers, just like still photographers, tend to be fad followers.
But followers have to, by definition, be following and that implies that somebody is out there in front. With diffusion filters that is certainly not me—adapting old lenses to modern cameras is a well-established corner of the photo world and diffusion fits into that same corner as well, although even there it’s sort of a niche, a corner within a corner. Perhaps we can add a bit to that little group—put the Glimmerglass on the fast Fuji 50 and see what happens?
My house sits on a street that leads to an old mountain road—the name changes but it’s the same road—that crosses Montara Mountain. My family and I have been walking this same walk, this same hike, for years and every walk is different. Different lighting, different people, different birds. So I picked out a tripod and headed out, running a bit late as the Sun was already well down toward the horizon.
In each of the image pairs that follow the Glimmerglass image is first, followed by the unfiltered image.
A few weeks ago this scene was so thick with underbrush you couldn’t see through it. There’s a trail down just past the main group of tree trunks and from where I stood for this image you’d just see an occasional head of another hiker appearing and disappearing if you stood there long enough. Then came the fires and smoke—not here—and the county powers-that-be in far-away Redwood City or the state powers-that-be in even farther away Sacramento began to clear forested areas near our small town. It was a shock to come upon this opened area after it was done. You can get a sense of that shock by noting the number of sawed-off limbs indicated by the fresh cut marks on the trees.
The effect of the Glimmerglass filter shows most obviously in the sun-bright portion at the top edge of the image where the white sun suddenly blooms, the white glow spreading out over the leaves and the tree trunk. There is a lowering of contrast—the Glimmerglass image might initially look dull in comparison. I’m not sure if there is a loss if resolution or not—higher contrast can give the illusion of sharpness and lowered contrast thus seems less sharp.
It’s the very bright areas that reveal the most pronounced effect. In this image of a tree, three-fourths of the image shows a loss in contrast but is otherwise little changed. The top left quadrant, with the sun shining through, shows again the blooming that gives that area of the image a different feel. Note also the dark areas on the tree branches in the upper left—we might be seeing a slight increase in detail in the blackest areas.
Film had a threshold effect with exposure. On a graph you’d see that small amounts of exposure would produce a small amount of density in the negative but in practice a small amount of exposure would produce no density at all. The shadows, which would theoretically show detail, were full-on black. One trick to gain back that shadow detail was to pre-expose each negative briefly, shining a light upon each sheet, giving it an overall exposure that put the negative just above that threshold level. That way, any additional exposure, even a tiny bit of exposure to the very dark but not black part of the scene, would register on the negative with a little bit of added density.
I suspect something like that is happening here, although happening at the same time with the main exposure, rather than in two steps. The glow from the light source as seen through the Glimmerglass is adding a bit of exposure to the entire image area, or at least broad areas around the highlighted area, pushing those deep blacks up just a tad so that detail will now register.
The ocean here doesn’t seem much affected by the filter. Perhaps it lacks the fine detail that would reveal the presence of the filter and with the already high contrast of the Sun’s reflection on the waves a slight loss in contrast is not noticeable. The Sun, on the other hand, shows a pronounced difference. In the filtered image the Sun is radiant, a sort of gossamer corona extends out from it, gently blending into the sky. The Sun, although at the same ISO, shutter speed, and same aperture in each image, is fully overexposed and gives the Sun a sense of brightness, a little blinding, really. In the unfiltered image the Sun is less exposed, more accurate to what the eye saw. Compared to the filtered shot, this Sun has a dark ring around it, augmenting its more defined edge.
One is not better than the other, one desirable, the other not. They are just different possibilities.
Intermission time. Before look at additional photograophs with and without filtration, let us think briefly about sharpness (again). In some places, when you test a lens for sharpness, you might consider the use of a brick wall. Here, on the coast of California, we use dolphins.
Here’s a full-frame image, the purple water actually colored purple by the just-set sun.
And zoomed in at 100%. This is at f/1.
And a similar image at f/1.3
Even after I introduced the thought that we were looking at sharpness, even after I set the stage, were your first thoughts when you looked at the images about sharpness? Or were they about color, about the ocean, and most of all about the dolphins?
Another image and another close-up
This looks to be a humpback whale—there may be two of them. They are here unseasonably late in the year, the majority of the whales passing for the north months ago. This whale appeared to be doing a solo version of lunge feeding where a group of whales, acting in concert, will swim around a school of fish which, frightened by the whales, pack themselves into dense defensive balls. Poor, dumb fish. Then several of the whales will swoop up from underneath, right into the ball, mouths open, and fill themselves as they lunge out of the water. The whales are said to take turns corralling and swooping so all can eat.
This whale was moving in a circle, too, in shallow water, apparently both corralling the fish and then coming up on them to eat them. I saw him lunge out of the water but neglected to take a picture. It is also unusual to see whales active and feeding like this after sunset—normally about a half-hour prior to sunset the whales tend to wrap things up and head to wherever it is they go at night. But there was a lot of fish in the water, witnessed by the many birds on the surface and the dive-bombing Pelicans crashing into and under the swells, and so that may explain the late activity, though I worry that this whale may be here now more out of weakness and hunger than out of choice.
The Moon in this image dominates the composition and the difference between the Glimmerglass-filtered image and the straight image couldn’t be starker. They are different photographs. The dark foreground areas of the photograph look identical in each and yet in one the diffuse glow around the Moon renders a different Moon entirely. Again, one is not better than the other—that depends on the photographer’s intent and the viewer’s taste—but the different possibilities are clearly, and literally, illuminated.
Note in the Glimmerglass photo the ball of greenish light—lens flare from the surface of the filter. The flare is not evenly toned—more on that later.
There’s a slight mist in the air here—actual water vapor, not residual smoke from the forest fires—which gives its own diffusion filter effect in the light. This is greatly augmented by the filter, as you can see, creating a glow all across the houses in this small town. It’s worth noting that Glimmerglass comes in several strengths—the one I am using here is the weakest.
When a light source is overexposed it blooms—it spreads out over the adjacent parts of the image. The diffusion filter strengthens that spreading and here you can see the Moon has bloomed out to many times its original size. In this particular case I find this extreme blooming undesirable although no doubt good use could be made of this effect.
Earlier we saw flare from the filter and here it is again, though brighter and clearer. And again we see that it is not even—is there something in the flare? And there is—a picture of the Moon. With greatly reduced exposure the details of its surface are clear, the focused light having bounced off the flat filter in some way. There’s a coolness factor here that cannot be denied.
That small dot between the bloom of the Moon and the flare spot is not another flare spot. It is Mars, a little over-exposed into white but hints of red remain at its edges. As this time Mars rises with the night and Jupiter and Saturn pair up higher in the sky—you can see several of Jupiter’s moons and Saturn’s rings with ordinary binoculars—the Moon racing by them and then leaving them as the month goes on.
It’s a busy night sky.
Our final example is of a neighbor’s house and her Halloween decorations. It is shot at f/2.2 so we aren’t getting any of that glow in the lights from the lens itself—that effect is most pronounced at f/1.0 and goes away quite quickly as you stop down. Instead, looking at the unfiltered image, the glow is from overexposing the lights. The camera’s sensor—which can be thought of as a tightly packed grid of buckets being filled with photons—spills out some of its electrical charge when too many photons reach it, spilling from one pixel into adjacent pixels. At first glance it looks like filtration or mist in the air or some clever combination of the two.