top of page
  • Ed Dozier

Nikon D850 Focus-Shift-Shooting Step Width Calibration

The Nikon D850 has a focus-shift shooting feature, but they provide no concrete information about how much of a “Focus step width” to use. Instead, they only give you a step-size to select, ranging from 1 (small steps) to 10 (big steps).

105mm Micro Nikkor at f/8.0 15 stacked shots, shift 6, 0.4X

For a shot like the kangaroo paw above, how do you determine what focus-shift step size to use? How do you determine how many shots to stack? What aperture should you use? If you have ever tried a macro shot with the level of magnification like this photo, you know that you can’t get this much depth of focus with a single shot.

Before you go out into the field to shoot macro subjects, it would be worth your time to know to configure your camera and lens. The following article shows you how to calibrate your focus-shift shooting for an optimum shift step-size.

The Focus shift shooting menu

D850, 105mm Micro Nikkor, and LED ring light

I used the gear shown above to take the close-up shots for the photos that follow. The battery-powered LED ring light provides really excellent shadow-less illumination. This setup works well even at 1.0X magnification. A tripod isn’t strictly necessary at long focus distances to do focus-stacking, but it (or some other camera support) certainly is required at high magnifications. You should use the “Silent photography” setting whenever possible, to minimize vibrations.

I did some testing with my Micro Nikkor 105mm f/2.8 lens on the D850 to dig a little deeper into these “Focus step widths”. It turns out that those Nikon engineers did some pretty smart engineering.

Although focus-shift shooting can be done with any autofocus Nikon lens at any distance, the main reason for this camera feature is for macro photography. Focus depth is incredibly small at high image magnifications, and most subjects turn out to be very disappointing with only a miniscule region in sharp focus.

Nikon partially fixed this narrow-focus-depth problem with the ability to take a requested number of shots while automatically changing the focus slightly between shots. A problem arises with the “slight” focus change you want to specify to the camera. When you change lenses and/or apertures, the amount of focus change is different with every combination. To some degree, Nikon simply punted and came up with a dimensionless step value that you can request.

Nikon only partially fixed the shallow-focus problem, because they don’t actually stack the photos. I’ll discuss later in the article how that problem can be rectified.

10 Stacked photos, Micro Nikkor 105mm at f/8, 1X

It’s hard to believe, but the shot above is actually a stack of 10 photos. The smallest divisions on the ruler are actually 1/100”. This means that the region in focus is about 0.15 inches (3.8 mm) along the ruler. Since this is a stack of 10 shots, a single shot has only about 0.015 inches in focus! The ruler was rotated to 70 degrees, so the actual focus depth is even a little less.

The combined stacked shots above, shot at f/8 with the Micro Nikkor 105mm, are at life-size magnification. Now, try to imagine if the lens focused down to maybe 2X magnification; almost nothing would be in focus. I chose f/8, since it’s the best combination of focus depth and resolution. The f/8 I mention throughout this article is the infinity-setting f/8, and it doesn’t take into account the increasing f-number as magnification increases.

Micro Nikkor 105mm at f/8, 1X single frame

The shot above is a single frame taken with the Micro Nikkor 105mm at f/8 and 1X magnification. There’s almost no depth of focus at all! Now you see why focus stacking is gaining in popularity.

I did a number of tests to find out how much the focus-shift “step size” should be to guarantee focus overlap, but not to overlap focus any more than necessary. With the 105mm at f/8 at this focus distance, the largest focus step width size that would still have a frame-to-frame focus overlap was 7. If I stepped larger than that value, then I was left with unacceptable bands of out-of-focus in the stacked photo. If there was any wiggle in the camera or subject, I once again had a ruined shot stack.

I determined that the optimal step size for life-size magnification with the 105mm at f/8 is 6. This gives a very slight safety margin for vibrations or image movement.

I used a tripod for all of my testing, so the Silent photography is a great option to minimize vibrations. If you try hand-holding shots instead, turn the Silent photography option off (always use a tripod or some other device to help steady your camera for really close shots).

After I had my desired test settings, I just pressed the “Start” option to begin automatically shooting the series of shots. After a brief “Processing” message, the D850 would then take the series of shots.You can also press the center button on the multi-selector to start shooting, once "Start" is highlighted.

10 Stacked photos, Micro Nikkor 105mm at f/8, 0.5X

My next set of testing was conducted at 0.5X magnification, with the same 105mm at f/8. Guess what? The optimal step size was 6 again! Stacking 10 shots, the range shown in focus on the ruler was about 0.65 inches, or 0.065 inches per shot.

This shows that the depth of focus went up by more than 4X at half-life-size, compared to life-size.

For typical subjects, you’d probably want to stack more than 10 shots at this magnification and f-stop. All that you would need to do is increase the “No. of shots” in the menu to maybe 20 or 30.

Micro Nikkor 105mm at f/8, 0.5X, single frame

I’m showing a single shot at 0.5X magnification above, using the 105mm at f/8. This single shot still isn’t enough to focus the width of most bugs, but the stacked shot of 10 would have been just fine.

10 Stacked photos, Micro Nikkor 105mm at f/8, 0.25X

Switching to 0.25X magnification, my testing showed that the optimal focus shift step size should be 6 once again. Simply amazing! Those Nikon engineers designed a system where a single focus shift step size would work at that whole range of distances!

At 0.25X magnification, the 10 stacked shots above now shows an in-focus range on the ruler of about 2.5 inches. The depth of focus for the 105mm at f/8 for 10 stacked shots is getting into a really useful range for many subjects.

Micro Nikkor 105mm at f/8, 0.25X, single frame

Even at 0.25X magnification and f/8, the depth of focus is still too shallow to be very useful for this 105mm without focus stacking.

Micro Nikkor 105mm f/8, 1/10X through infinity

For testing something on the scale of landscape photos, you could extend my concept to use a subject like a fence. This would make it easy to evaluate shot-to-shot focus overlap. When you take the shots, the camera will simply stop shooting once it reaches infinity focus (or the end of focus travel), so you don’t have to worry about requesting too many shots in the sequence.

Unsurprisingly, the same focus step shift size of 6 worked yet again for f/8. This stack took 29 shots to cover 1/10X through infinity. The camera took a few extra shots beyond infinity, stopping when it got to the end of focus travel. Remember to delete these “beyond infinity” shots before focus-stacking!

This shot above demonstrates a major weakness of stacking shots for a landscape: wind. You can see some little ghost branches, etc. It’s a cold reality that you’ll usually end up using things like a healing brush in your editor when stacking shots from the outdoors.

Micro Nikkor 105mm f/8, 1/10X. First shot of stack

Until you see individual frames from a focus stack, it’s hard to appreciate how much focus depth can be obtained. It’s kind of like getting cataract surgery.

How I stacked the test shots

I used the free CombineZP program, created by Alan Hadley, to stack all of the shots shown in this article. I used Windows10; you’re out of luck on Apple computers. Something like Helicon Focus (not free) should work on Apple systems.

Before the D850 “Focus shift shooting” feature existed, I used my Nikon PB-4 bellows for macro focus stack shooting. In that case, each photo was shot at exactly the same magnification. Using the D850 focus-shift shooting, the photos need to get “scaled” to be the same magnification. The CombineZP program has a few different macros that can solve this ‘magnification change’ problem.

If you choose DX mode shooting versus FX mode, you’ll find that the stacking process is maybe 4X faster. Also, CombineZP sometimes has problems with full-size FX images in the TIF format, which changing to DX mode solves.

Align (and scale) the shot stack in CombineZP

To make a stacked shot, there are basically 6 steps I followed.

1 Convert the raw shots into TIFF format. (I use 16-bit)

2 Open the shots in the CombineZP program

3 Run the Align and Balance Used Frames (Quick) Macro

4 Run the Do Weighted Average Macro

5 Draw a rectangle around the finished stacked shot to trim it

6 Save the rectangle in jpeg (or TIFF) format

Because the D850 does focus-shift shooting to create the photos to stack, each focus distance change causes a different magnification. If you merely combined the shots into a stack, none of the shots would match each other in size. The CombineZP program can fix this “scaling” problem, and simultaneously align the shots as well.

The CombineZP program doesn’t understand the D850 raw photo format, so I use the free Nikon Capture NX-D program “batch process” feature to easily convert the shots into TIFF format.

There are several stacking macros to choose from in this program; you can simply try another macro (or all of them) to see which one you prefer. Thank you Alan Hadley for this excellent program!

I do have to mention that the CombineZP program can save images in TIF format, but some of my other editing programs don’t like those files; that’s why I recommend that you additionally save into jpg format, just in case. At least try out your favorite image editor with a file saved as TIF from CombineZP to verify it works okay.

By the way, I run the CombineZP program under 64-bit Windows 10. There are other stacking program offerings out there, but I can’t comment on their relative ease or capabilities. You’ll need some other stacking software, such as Helicon Focus for Apple products.


For any given lens and f-stop, it’s pretty straightforward to take a series of focus-shifted shots using the Nikon D850 to find an optimal step size. By looking at neighboring shots afterward at 100% magnification, it’s easy to see if the shots share a common zone of focus. If the shots are shifted too much, simply decrease the shift step size and try shooting again.

Once you have a “calibrated” step width for your lens and f/stop, it simply becomes a set-and-forget scenario. You can shoot with the confidence that your stacked shots should combine perfectly.

If the last shifted photo hasn’t yet gotten enough of your subject into focus, then you can simply command the focus-shift “Start” option again, and the D850 will pick up focus where it left off. You can always increase the total number of shots to take, especially if your estimate was way too low to cover the whole subject.

Focus stacking is pretty slow, so you want to minimize how many shots you try to stack. You will find it much easier and quicker to take shots at a lesser magnification (e.g. 0.5X instead of 1.0X) and simply crop the final stacked photo. The D850 has such high resolution that cropping is almost never a problem. If you’re lucky enough to have the 105mm Micro Nikkor, it’s very, very sharp. If you change your shooting to the DX crop mode, the stacking will be much faster, and it will take a fraction of the disk space.

I’m amazed that the Nikon engineers designed a “Focus step width” algorithm that would allow a single setting to work with this 105mm lens at any distance for a given f/stop. This feature is going to revolutionize macro photography, which has historically been a very slow and tedious chore.

105mm Micro Nikkor f/8 shift=6, 12 shots, DX mode, 0.25X

I tend to use my LED ring light with my macro photos instead of a flash. I really like how smooth the lighting results are, compared to flash.



Recent Posts
bottom of page