When you shoot multiple shots to stitch into a panorama, you should pivot the lens around what’s often called the entrance pupil . This magical pivot spot is the only  location that will eliminate parallax error. This spot is also known as the nodal point .   If you don’t pivot your lens in the correct location, you’ll end up with overlapping shots that don’t match up precisely. Stitching programs try to compensate for this error, but it’s better to get the right shots in the first place.   Unfortunately, lens manufacturers don’t normally provide the information about the lens’ entrance pupil. In this case, you’ll probably have to figure it out for yourself. Believe it or not, sometimes the entrance pupil can be located outside  of your lens.   To set up a lens for proper panorama shooting, you’ll need a tripod head that can rotate about the vertical axis, and you’ll also probably need an attachment plate that you use to offset the camera tripod screw from the tripod’s pivot axis. The best attachment plates are Arca-Swiss (your tripod head needs to have an Arca-Swiss clamp). With these plates, you can easily slide the camera/lens and then securely clamp them in place.   There are also special ‘panorama heads’ you can buy to make panoramas, instead of the more generic Arca-Swiss clamps. You still need to locate the pivot-point(s) for your lens to make proper use of them, however. Beware that zoom lenses have a different pivot point for each focal length. Camera attached to a long Arca-Swiss plate   In the photo above, the camera is attached to a long Arca-Swiss plate on top of a tripod. The long plate allows the camera to be shifted far from the tripod rotation centerline. Now, when the tripod head gets rotated the camera/lens will rotate around the vertical red line instead of the camera tripod socket.   Usually, really long Arca-Swiss plates are a pain and are best avoided. Here’s a case, however, where a long plate is exactly what’s needed. As you’ll see below, sometimes even the hardware shown above isn’t enough to achieve a correct pivot.   Method 1 to Locate Pivot Location   To gather the information about the ideal pivot location, you need to start by taking photos with subjects at medium and long distances that are aligned with each other. Start by aligning the near and far objects in the exact frame center. Your camera should be perfectly level. Now, rotate the tripod head to get those objects on the left-hand side of the frame. Take a shot with the distant detail on the left frame edge and then rotate on the tripod to the right edge of the frame and take another photo.   Repeat the process, but first shift the camera/lens combination in the tripod’s Arca-Swiss clamp by a large amount. Make notes about which shift location was used for each pair of shots.   With a mirrorless camera, you might get away with just switching to a magnified view to check for left/right alignment and skip having to review photos. The first attempt: pivot around the camera tripod socket   I used an 8-meter-away window frame for the medium object, and a 500-meter-away palm tree for the distant object. I adjusted the camera position until the medium and far objects were right next to each other in the exact frame center.   Next, I pivoted the camera to place those objects on the left side of the frame. I captured the first shot. Close object aligned with distant detail on the frame left edge Distant detail appears shifted right  from the window frame       Next, I pivoted the camera to place those objects on the right side of the frame and took another shot. Close object aligned with distant detail on the frame right edge Distant detail appears shifted left  from the window frame   The results from pivoting over the camera’s tripod socket were terrible. The distant object shifted severely, compared to the window frame. It slid from one side of the window to the other, just by photographing it from opposite sides of the frame.   What’s shown above is classic parallax error. The distant object keeps moving, relative to a closer object, when shot from different left-right locations in the frame. The next test: pivot around the lens as shown     Since I am using a mirrorless camera for the testing, it’s pretty straightforward to just iterate as follows, once I have my near/far objects aligned in the frame center using a magnified viewfinder image:   1)    Pan left or right and observe any shifting of the far object. 2)   Slide the camera/lens forward or backward in the Arca-Swiss clamp by maybe 10mm and lock it down. 3)   Repeat the panning to see if the far object shifting error grows or shrinks. If it is growing, then slide the opposite direction in the clamp. Go to step (1).   In my testing, the trends indicated that moving the pivot point more toward the lens front got better and better results. Finally, I ended up with the configuration shown above. The entrance pupil is roughly between 90 and 100mm from the camera sensor at the 24mm focal length. I couldn’t tell the difference after I got to within a zone of maybe 5mm at this focal length. Close object aligned with distant detail on the frame left Distant left-detail appears the same as in the frame center!   Close object aligned with distant detail on the frame right Distant right-detail location matches the frame center and left!     After I located the proper lens pivot location, I attached the Arca-Swiss plate onto the camera such that one end of the plate exactly aligned with the Arca-Swiss tripod clamp edge. This makes it easy to reliably place the camera onto the tripod with perfect alignment for correct panorama pivots.   I got lucky with this 150mm Arca-Swiss plate. If it were any shorter, then I wouldn’t be able to quite reach the lens nodal point with the plate fully gripped inside the tripod head’s clamp.   The top photograph shows the correct nodal point (entrance pupil) for the Nikkor 24-120mm f/4 S lens when zoomed to 24mm. Same lens zoomed to 120mm: entrance pupil now located at 59.5mm I got my smug smile wiped right off of my mouth when I now tried zooming the lens from 24mm out to 120mm. Big-time parallax error had returned. After repeating the panning procedures at this zoom setting, the new nodal point was located as shown above. Method 2 to Locate Pivot Location   The following technique can at least get you close to finding the lens nodal point (entrance pupil). This technique is more challenging and time-consuming than the first method.   You need to draw the lens horizontal field of view (FOV) angle on paper, with a line splitting that angle exactly in half. In the case of this 24mm setting, that’s 73.7 degrees, with the split line at 36.84 degrees. Use a protractor to measure the angle. Lay this paper on the floor and lay some long straight guides along these lines. You might find that using string taped to the floor will make a good guide line. Mark the exact field of view angle and lay guides     Next, place your camera on the floor over the intersection of these guides, and shift it forward and backward until both guides are seen just along the vertical frame edges in the viewfinder. Switching to Live View may make this job easier. Be sure to stop down the lens aperture to get a deep depth of focus.   Also, you’ll want your camera to be level. I have a handy little bubble level that I keep in the camera’s hot shoe to check for level. Align the camera to split the angle Slide the camera to get the guides parallel to frame edges   Notice that the thin black guide lines are just along the frame edges in the shot above. Where the guides intersect under the camera/lens is the location of the correct pivot point (entrance pupil).   To verify the pivot point, go back to Method 1 using this location as a starting point to see if it's correct. You may need to fine-tune the location a little. 15-frame panorama after pivoting around the ‘entrance pupil’   I stitched the photographs together using Capture One  with ‘cylindrical projection’. This editor has several projection options. I like using Lightroom  for making panoramas, too. Nikkor AF-S 24-70 f/2.8 E ED VR at 24mm zoom setting     Shown above is another lens also zoomed to 24mm. Its entrance pupil is quite different than the 24-120mm lens. It should be obvious that the kind of severe offset of the camera and the center of gravity shown above requires a very sturdy tripod. The last thing you want is to have your tripod fall over with your camera and lens.     Entrance Pupil the Easy Way If he happens to include your lens, Bill Claff’s Photons to Photos database of lens information has already compiled the pivot information you need. Bill’s wonderful site has many lenses analyzed here, and you need to find the value of ‘P’  (entrance pupil) and ‘ I ’ (image plane) for your lens at the correct zoom setting. For the 24-70 lens shown above, his site link gives the values of P=34.76mm  and I=220.25mm at 24mm.  Just do the subtraction  I-P or  (220.25 – 34.76)= 185.5mm  to find the distance from the camera sensor to the tripod pivot location. At 50mm zoom, the pivot is at 181.25mm , and at 70mm zoom, the pivot is at 172.3mm .   These pivot numbers are pretty huge, requiring hardware with a long range of adjustment. This also requires strong tripod hardware.   I found out that Bill had also analyzed my 24-120 f/4S lens, unfortunately only after  I did the experiments mentioned above, and I got essentially the same results as Bill did. Bill got I@24mm = 130.96mm, P@24mm = 30.8 and I@120mm = 185.96mm and P@120mm = 126.43mm. This means the sensor-to-pivot at 24mm = 100.16mm. For the 120mm zoom setting, this means the sensor-to-pivot value is 59.5mm .   Bill has my 24-120mm lens listed as “Nikon 25-117mm f/4 IF”, which is why I didn’t find it at his site initially. He lists it according to the measured focal length versus the actual lens name.   Bill has lots of very nice information on his website, although much of the site is targeted more for scientist types.   Summary   Probably 99% of photographers would simply use the camera tripod socket location on their tripod for making panoramas. As demonstrated above, that would simply produce incorrect results. Don’t be that photographer; you’re better than that.    The procedures shown above could also be done with the camera in portrait orientation, but you better be sure that you have pretty heavy-duty hardware, a camera ‘L’ bracket, and a strong tripod.

I thought it would be interesting to show you how to find out (or verify) your lens’ focal length and its field of view.   One reason you might like to measure this stuff, versus just reading the lens focal length stamped on the lens, is that manufacturers often lie about the true lens focal length. This is particularly true of long zooms, which might not be as long as claimed. It’s often the case that a lens focal length reduces drastically when focusing close (called focus breathing), and these techniques could attach real numbers to that.   The diagrams above show the setup to align and measure your lens to get the information for calculating the focal length and field of view ( FOV ).   You only need to take 6 different measurements to calculate everything! The trick is in setting up the targets to make those measurements accurate.   The mathematics used in the calculations involves both algebra and trigonometry. Any “scientific” calculator has the necessary features to do this math. You can always download a scientific calculator app onto your smartphone. Fear not the math.   To get all of the necessary information, you’ll need to set up some targets on a wall (I used painter’s tape) and then take distance measurements at two different camera-to-wall distances. Carefully place tape at center and edges of the field of view   As shown above, I put some tape targets on the wall that point at the edges of the field of view and the exact center of the frame. I made sure that the distances on either side of the middle of the frame exactly matched. I also made sure that the lens was at exactly the same height as the middle wall target. Using high magnification via Live View or the viewfinder helps getting the targets more precise.   My setup was on a tile floor, and I used the tile grout lines to make sure the lens axis was exactly perpendicular to the wall. You could also temporarily tape a little mirror against the wall and line up the camera until you see your reflection in the center of the camera’s field of view. Laser for even better measurements   I happen to have a fun little laser that can make extremely accurate measurements over long distances. I used this device to get my targets placed within a millimeter, and to measure the camera-to-subject distance.   As I often say, “garbage in, garbage out”. The more accurate measurements you make, the more accurate your results will be.   I made all of the distance measurements relative to the lens axis and to a spot on the lens that I guessed to be near the vertex of the red arrows shown in the diagram. You don’t really need to know this location, since these techniques will actually calculate the location.   The nearer wall distance setup is called ‘ x ’. The farther wall distance setup is called ‘ X ’. Similarly, the distance along the wall from the frame middle to the frame edge in the nearer wall setup is called ‘ y ’, while the farther distance setup is called ‘ Y ’.   For both setups, the angular measurement from the lens axis to the frame edge is called ‘ Angle ’, measured in degrees. The FOV  of the lens is twice the value of ‘Angle’.   k  = Error Distance from lens measurement location to the true angle vertex.   k = ((y * X) – (x * Y)) / (Y – y)   x  = near_wall_distance from lens measurement location X  = far_wall_distance from lens measurement location   y  = (near_view_width) / 2 Y  = (far_view_width) / 2   Angle  = FOV / 2   Angle = ArcTangent (y/x)   x = y / Tangent(Angle)         and   X = Y/Tangent(Angle)   y = Tangent(Angle) * x         and   Y = Tangent(Angle) * X   FOV  = 2 * ArcTangent(y/x)  and FOV = 2 * ArcTangent(Y/X)   x_Actual  = x + k                   X_Actual  = X + k   Focal_length = (sensor_width * x_Actual) / (y * 2)       Example   I used my Nikkor 24-120mm f/4 S zoom at the 24mm position.   x = 1306mm y = 1085.85mm X = 2141mm Y = 1711.325mm   k = ((y * X) – (x * Y)) / (Y – y) k = ((1085.85 * 2141)-(1306 * 1711.325))/(1711.325 – 1085.85) k = 143.594mm   FOV  = 2 * ArcTangent(y/x)  FOV near = 2 * ArcTangent(1085.85/1306) = 79.48 degrees   FOV far = 2 * ArcTangent(1711.325/2141) = 77.27 degrees   x_Actual  = x + k x_Actual = 1306+143.594 = 1449.59mm   X_Actual = 2141+143.594 = 2284.59 mm   FOV_Actual = 2 * ArcTangent(1085.85/1449.59) = 73.67 degrees   Focal_length = (sensor_width * x_Actual) / (y * 2) Focal_length = (36 * 1449.59) / (1085.5 * 2) = 24.03mm   My angle vertex ‘guess’ at the front of the lens wasn’t very close. I had a calculated error of about 144mm.   It just so happens that the horizontal field of view of 73.7 degrees is what a 24mm lens has with a full-frame camera!   Summary   Using these techniques, you could actually find out what an un-marked lens’ focal length is. You could also verify (or debunk) manufacturer claims about the real focal length of a lens.

It’s in Nikon’s documentation that you can only use Pinpoint AF  while in the AF-S  focus mode. It turns out that even in this AF-S mode that you’re not allowed to select Pinpoint focus when you have assigned the “ Cycle AF-area mode ” to a button. I’m talking about assigning a button inside the Custom controls (shooting)  menu option.   I’m using the Nikon Z9 firmware version 5.0 the Z8 2.01 version. I just noticed this bug, but I assume it’s also a problem with older firmware.   If, instead, you go to the Photo Shooting Menu  and select Pinpoint AF inside the AF-area mode  there, you’ll be able to enable Pinpoint focus (assuming you are using Single-AF focus mode).   When I tried pressing the Sub-selector center button, it would cycle right past Pinpoint and invoke all of the other requested AF-area modes that are legal while in AF-S focus mode. 3D-tracking, for instance, isn’t allowed while in AF-S focus mode. Set Pinpoint AF here Custom controls (shooting) button assignment Assign Cycle AF-area mode Pinpoint AF gets ignored!! Pinpoint focus indicator   I hope that Nikon will eventually fix this bug.

I have known since the first day I got my Nikon Z9 that it focused better than any Nikon DSLR I have ever owned. When I got the Nikon Z8, I noted that it seems to focus exactly like the Z9.   I was a long-time holdout against mirrorless cameras, both because of their short battery life and their sluggish viewfinders. I didn’t see any particular advantage over DSLRs to persuade me to stop using my D850 or D500. Nikon eventually solved all of the mirrorless shortcomings that I knew about, and then some. I didn’t know just how much they had improved the focus system compared to DSLRs.   Instead of just ‘feeling’ like they focus more consistently, I did some tests to get some actual numbers comparing one of my mirrorless Nikons (the Z8) against my best DSLR: the Nikon D850. The camera sensors have the same resolution, so any measurements can be compared directly.   To eliminate any doubts about the lens being a limiting factor, I decided to use my 500mm f/5.6 PF Nikkor, which I consider my best lens, for the tests. Both cameras used the same ISO and exposure. I used a wired shutter release, and the D850 shots were done in “mirror-up” mode.  I manually de-focused the 500mm lens between each test shot, to force the cameras to do a full re-focus using AF-C.   My benchmark to evaluate focus accuracy is the lens resolution measurement, since higher resolution measurements correlate perfectly with accurate focus. I used a precision razor blade edge as my focus target, since it always gives me either equivalent or superior resolution measurements compared to my printed targets. I use the MTFMapper  program to evaluate my focus target photos.   I tested both cameras using phase-detect 3D focus. I’m not interested in comparing slow contrast-detect focus, even if it might be more accurate. My D850 has been autofocus fine-tuned with the 500mm lens; my Z cameras don’t seem to need this crutch, although they include that option in their menus.   I took many raw-format photos of a precision razor blade that is in silhouette. I don’t want to use any automatic sharpening (such as when shooting jpeg) to affect the resolution measurements. The camera/lens is on a heavy support, so that each shot has the blade silhouette in exactly the same location. Make sure that the razor blade is perfectly parallel to the camera sensor. Mirrorless Z8 versus DSLR D850 The focus subject is a slanted high-precision edge   The shot above shows the MTFMapper  dialog used to perform “manual edge selection”. I have a razor blade with strong back-lighting as a target, and I put the camera focus point over the center of the razor’s sharpened edge. After using the computer mouse to select the portion of the blade image to use, I let the MTFMapper  evaluate this edge to determine the lens resolution at this position and orientation.   The MTFMapper  program lets me select all of the test photos at once; it assumes the target in each shot is at exactly the same position. I make sure that “ reuse ROIs ” is selected, and then click the “ Accept Queued ” button to have the program evaluate all of the photos without further intervention by me. A typical shot after MTFMapper  has finished evaluation   You can see a little cyan-colored MTF50 resolution measurement superimposed over the location on the blade where I requested the program to evaluate the resolution. In this shot, the MTF50 resolution measurement is 84.8 lp/mm.   I put all of the measurement results from MTFMapper  into an Excel  spreadsheet. I then plotted the resolution measurements against the camera filename (frame number). D850 resolution variation with 500mm f/5.6 PF, AF_tune -4   The plot above shows the variation in resolution measurements for the Nikon D850. Again, I de-focused the 500mm lens and then refocused between each shot, to force the camera to perform a new phase-detect, AF-C focus in 3D-tracking mode.   The D850 results got an average MTF50 resolution of 63.2 lp/mm , with a peak reading of 81 lp/mm  and the standard deviation (spread of measurements) was 8.12 .   Since there was a single measurement higher than the others, the autofocus fine-tune setting could possibly be re-adjusted to get overall better focus, although the photo EXIF data in every frame indicates the same 4.47m focus distance. The spread  of the measurements isn’t affected by the autofocus fine-tune setting.   Bear in mind that resolutions above 40 lp/mm will all appear pretty sharp, so these results are still good. Nonetheless, I decided to change the AF-tune and re-try the shots. D850 resolution variation with 500mm f/5.6 PF, AF_tune -2   The D850 with the new AF-tune -2 results got an average MTF50 resolution of 71.7 lp/mm , with a peak reading of 84.8 lp/mm  and the standard deviation (spread of measurements) was 10.3 .   The slight AF-tune adjustment change definitely helped! Notice that there is sort of a split in the data. When I was shooting the target, I would de-focus the lens by alternating too near and then too far shot-to-shot. The camera seems to have a bias in focusing based upon the direction to achieve focus.   I don’t know if something might have shifted or drifted over time with the focus calibration. It was definitely worth re-checking this, since it resulted in a typical gain of about 13 percent resolution. Z8 resolution variation using 500mm f/5.6 PF   The plot above shows the variation in resolution measurements for the Nikon Z8. Again, I de-focused the 500mm lens and then refocused between each shot, to force the camera to perform a new phase-detect, AF-C focus in 3D mode.   The Z8 results got an average MTF50 resolution of 74.0 lp/mm , with a peak reading of 98.2 lp/mm  and the standard deviation (spread of measurements) was 6.7 .  With the exception of the single reading of 98.2, the data was much more tightly grouped than the D850 results. I think that the 98.2 reading is some kind of a fluke which should probably be ignored.   If you’re interested, the average resolution of 74 lp/mm is equivalent to 3537 lines per picture height. I prefer using ‘lp/mm’ units, since it applies to any size sensor.   I didn’t notice any bias in resolution related to focusing from near-to-far versus far-to-near, either. The Z8 (and Z9) are just smarter about focus.     Summary   The Nikon Z8 (along with its big brother the Z9) has noticeably better focus consistency than the D850. After seeing how the D850 results dramatically improved with such a small AF-tune adjustment, I should probably explore using that option on my Z cameras, too.   Before testing with my Z8/Z9 cameras, I would have said that my Nikkor 500mm f/5.6 PF lens had a typical center resolution of about 63  lp/mm when using 3D continuous autofocus on my D850. After studying those results and modifying the autofocus fine-tune, I changed my mind to conclude that the resolution was closer to 71  lp/mm. Now, I can say that it has a typical center resolution of roughly 74  lp/mm when using 3D continuous autofocus on the Z8/Z9 cameras.   If you only shoot using slow contrast-detect focus (via Live View on the D850) then these focus (resolution) differences will largely disappear. If your lens has focus-shift issues when changing the aperture, then the focus variation differences between DSLRs and mirrorless cameras will increase . The Nikon mirrorless cameras focus at the shooting aperture when using phase-detect (through f/5.6), but the DSLRs don’t.   You need to shoot many tests and study those results to guide yourself toward optimal calibration, in the case of DSLRs. Statistics exist for a reason; you can’t really know how your gear performs without doing lots of testing. It’s possible that your mirrorless Nikon could squeeze a bit more sharpness out of a lens with autofocus fine-tuning, as well.   Mirrorless definitely wins the war of consistency over DSLRs when using phase-detect focus.

I originally purchased ON1 Photo Raw  to get the ability to easily substitute interesting skies for boring skies in landscapes. Besides just using the sky images included in the ON1 installation , the editor lets you easily add your own skies.   Another great way to make photos much more interesting is to include the Moon in them. If you have ever worked on planning shots to include the Moon, you appreciate how difficult it can be to achieve. Being in the right place at the right time is rare, and those clouds can still move in to ruin your best laid plans. Is it real?     If you use a really big lens for moon/landscape shots, you’re well aware of how it’s typically impossible to get the foreground and the moon both in focus. If the moon is low in the horizon, then atmospheric effects ruin the sharpness.   Your substitute ‘sky’ can just as easily include the Moon. In this case, you could shoot photos of the Moon with a longer focal length and also take the shots when it’s high in the sky to get better details. It’s also a good idea to take both landscape and portrait-orientation shots for more options while editing later.   The ON1  editor is pretty good at masking complicated images, which allows you to even have tree branches partially blocking the Moon.   Have you ever taken landscape shots with wide-angle lenses, but you wanted to have the Moon appear larger in those shots? With substitute Moon/sky shots, now you can finally combine wide-angle landscapes with a large moon in them!   I should mention an old article I wrote that gives you the details on how to add your own custom skies to the ON1  editor. If you don’t know how to do this operation, then this article  should help. A Moonset shot that never really happened     Notice in the shot above that the Moon is partly hidden by tree branches. The ON1  editor is capable of pretty sophisticated masking. Inside the ON1  editor   The screen image above shows how I adjusted the sliders to control how a Moon shot was combined with a landscape. The landscape here was shot at 32mm, while the moon was shot using a long telephoto.   Once I find the desired Moon ‘sky’ and add it where I want it, I can then go over to the ON1   Develop tab to edit the shot further. A typical Moon shot to add to the ON1  editor   I have many Moon images added to my ON1  editor in its sky-swapping feature. I took my Moon shots while the Moon was high in the sky, so that I got rid of most atmospheric sharpness-killing effects.   There are several controls in the ON1  editor Sky  feature that let you have some control over where the Moon gets placed. This way, you don’t have to add too many shots into your ‘custom sky’ library. These controls include Position (left-right flip), Shift Horizon (vertical shifting), Level  (small horizontal/vertical movements), and Magnify (size change). The Moon adds a little spice to this harbor shot How the Moon’s reflection got added   For a more enhanced realism, the ON1  editor even lets you add a reflection if you want. I tend to prefer this particular shot without a reflection, but with this feature you get to explore adding a reflection or not. Don’t go there     Please avoid getting too outrageous with big moons, just because you can.  I suppose you could be working on some sci-fi assignment…     Summary   I know many photographers cringe at the thought of altering their shots in any way. If that’s your personality, then this technique is certainly not for you.   The ON1  editor has lots of fun and creative features in it, and using its sky-swapping capability to add the Moon can really make some blah shots a bit more interesting.

If you’re interested in seeing how sharp your camera lens is, but you don’t have a test chart to use, read on.  The free program MTFMapper has a feature called “Open with manual edge selection”. Using this program option, you’re given the freedom to select anything in a photograph and potentially use it to measure your lens resolution. The excellent MTFMapper program is authored by Frans van den Bergh, and can be downloaded from this site. MTFMapper has been used by NASA to measure lenses that are now in use on the planet Mars. There are rules (as always) for what you can use as subjects for testing lens resolution. For instance, you need a dark, straight, slanted edge against a light featureless background. The straight edges are preferably black, but if you have a strong backlight to get a silhouette effect, then the object with the edges doesn’t have to be any particular color. The MTFMapper program doesn’t like either vertical or horizontal edges, nor does it like edges at 45 degrees, due to the way the program’s mathematics work. To abide by this rule, all you need to do is slightly rotate your camera or your subject before photographing it. You’ll get different resolution measurements of the same target, depending upon which angle the target edge is at. Almost all lenses resolve differently in the sagittal (spokes away from the lens center) and meridional (perpendicular to sagittal) directions. Rotate the target or camera to enable getting the kind of measurement (sagittal/meridional) you want. To get the best results from this program, you’ll want to shoot your photographs in raw format (definitely NOT in jpeg format). I have some raw-format camera files that MTFMmapper can’t understand, because it uses the library called LibRaw to decode the raw files. My Nikon Z8 and Z9 ‘high efficiency’ compressed raw files can’t be used as-is by MTFMapper, but all I have to do is convert them into the DNG format. The MTFMapper program (using the Libraw library) understands DNG, and you can use the free Adobe DNG Converter program to convert nearly any raw format into DNG. Resolution measurement of a 500mm lens with 1.4X teleconverter As shown above, the MTFMapper program was able to measure the resolution at the edge of a clamp I attached to a fence. It got a resolution measurement of MTF50 50.3 lp/mm, which is the same as 2404  l/ph (lines per picture height). I used my Nikkor 500mm PF with a 1.4X teleconverter on my Nikon Z8 to take this 700mm f/8 shot from 58 feet (17.7m) away. I wasn’t able to set up my resolution test chart this far away, so I improvised with a clamp to get to the desired distance. I rotated the lens/camera combination so that the clamp wasn’t pointing quite vertical. I made sure that the background was much lighter than the dark clamp, and the background is smooth and out of focus. This clamp is large enough to make it easy to place the camera focus point on it, even at long distances. If the camera I’m using has it, I like to use the feature called “Pinpoint focus” to get the most accurate focus. Even the slightest focus error will cause a drastic drop in the resolution measurement. Pinpoint focus also helps ensure that the camera isn’t focusing on the wrong thing when I’m far from the target. The whole field of view for my test setup MTFMapper program option to pick edge(s) As shown above, I used the File | Open with manual edge selection… menu option to open up a suitable raw-format photograph. Select an edge to measure I mouse-clicked two locations on the edge that I wanted to measure. After clicking the first edge location (little orange circle above), I watched the histogram display (lower left of dialog) while I selected the second location (mouse move) to complete the edge selection. The histogram feedback helps you ensure that you get a clean selection that can be properly analyzed. If you decide to change your mind after selecting an edge, then you can click Clear ROIs (Region Of Interest) and select another edge location. It’s also possible to select multiple edges (regions) to analyze. You can fine-tune your edge selection by mouse-dragging the little orange circle to another location after your initial selection. Just watch the histogram while dragging the orange circle to optimize your selection. Slightly different location, different measurement I re-measured the edge, using a different section of a shorter length, and got a different answer. Here, I got an MTF50 resolution measurement of 55.7 lp/mm (2663 l/ph). This probably means that the edge I’m measuring looks more consistent over a short selection length of the target. This is a pretty cheap clamp, and not a super-precision edge. Multiple short regions could have been selected along the edge to compare them, too. After you have selected a suitable edge, then click Accept and allow the program to calculate the resolution at your selected edge(s). Make sure that you set up the correct options for your camera sensor (under Settings) before doing the manual edge selection, including the sensor pixel size. Selecting lots of edges of a printed resolution chart The screen shot above shows how I have selected numerous places in a photo of a test chart to analyze. Some of these selections are poor/illegal, to demonstrate what kind of feedback will result. If you select an unsuitable edge for the program to analyze, you might get an error message when you request that the program begins the analysis. You’ll either need to select another location or another subject to measure. Program feedback Successful measurements (looking at the annotated results) are displayed in cyan. Unreliable answers (45-degree sloped edges) are displayed in yellow, and failed answers (vertical or horizontal edges) are shown in red. I set up my program options to display the answers in MTF50 lp/mm units. Another kind of target I sometimes use a razor blade as a target when I’m shooting at fairly close distances. I use an LED light to back-light the blade, so it shows up as a silhouette. In the shot above, I requested measurements on 3 sides of the blade. If I’m far enough away, it doesn’t matter if I select the sharpened razor edge or another edge. I placed a white background directly behind the blade to get good contrast. Using a precision edge from a caliper (indoors) I have a handy little dual-clamp device with a pair of adjustable alligator clips that I used to hold a precision caliper. The program feedback (under the histogram) indicated that the caliper’s edge selection was 114.5 pixels. The minimum allowed selection length is about 40 pixels, but I’d recommend always using more than this. The annotated resolution result Again, when I select shorter sections of the edge I almost always get a higher resolution reading since the program sees less overall edge variation. The MTF50 result here was 58.2 lp/mm (2782 l/ph), or within about 4 percent of the reading from my Workpro clamp outdoors with the same 700mm lens/teleconverter combination. This shot was done at 40 feet (12.2m). The caliper’s edge is un-coated stainless steel, but I set up a strong backlight to get a good silhouette. Whatever target you try to use for getting resolution measurements, the choice had better be carefully thought out. Rough or crooked edges just won’t cut it.  Let the program’s histogram feedback be your guide for selecting a good edge. As always, this is a ‘garbage in garbage out’ scenario. Using lone edges like this won’t give you the comprehensive results that a proper resolution chart will, but if all you want is a handful of measurements then this should work fine. This is also a good way to compare a printed test chart’s quality against a precision edge. Summary It's hard to beat free. If you can locate a suitable subject with a good edge on it, you should be able to get lens resolution measurements. As I have said many times on my site, thanks again to that very clever Frans van den Bergh!

The Nikon camera “Pinpoint” focus mode is supposed to be the most accurate way to focus their lenses. It’s only available when in “AF-S” mode (not AF-C), so it’s only useful for static targets. On DSLRs (D850 & D780), it’s only available in Live View; I think all of Nikon’s Z cameras support Pinpoint focus. Is this mode really better than other focus modes? That’s what I am going to analyze in this article. My personal favorite focus mode is “3D-tracking”, although I use various other modes if unusual subjects get focused better with the other focus options. In the tests that follow, however, I will compare Pinpoint focus against just the 3D focus using the Nikon Z9 camera. I chose to use the Nikkor 24-120 f/4S lens at 120mm for testing. Nikon cautions that Pinpoint focus might be a little slower than other focus modes, but I didn’t notice any slowdown while shooting with anything except my 700mm f/8 setup (500mm + 1.4X TC). It supposedly only sees a quarter of the area of single-point focus mode. Pinpoint is supposed to use contrast-detect (hence slower), while the other focus modes are strictly phase-detect on their mirrorless cameras. Pinpoint AF-area mode selection Pinpoint mode indicator Note that you have to set the camera to AF-S (single AF focus mode) to have Pinpoint AF as an option. To conduct the tests, I placed the camera onto a sturdy tripod and attached a wired remote release to activate the shutter. I was 17 feet (5.3 meters) from the resolution target. In between each shot, I would alternate de-focusing the lens either too near or too far before pressing the AF-ON button. I waited at least 5 seconds before taking each shot, to allow any vibrations to stop. The whole resolution chart I placed the focus point in the center of the frame, so that’s where I used the resolution measurements for my analysis. I shot 19 photographs with each focus type, in order to get a good feel for the consistency of focus.  In none of the shots that I took was I able to visually tell the difference in sharpness. The only way to note a difference was to use my MTFMapper software to analyze the resolution target photographs. Resolution results for the central portion of the test target As shown above, I had my MTFMapper program analyze each test target raw-format photo and then display each measured target edge. I would pick the best reading inside the red square for my data. The resolution readings shown are in units of MTF50 line pairs per millimeter. Resolution plots showing the entire FX frame The plots shown above provide resolution data about the whole lens field of view, from edge-to-edge. There’s one plot for meridional direction (tangent to a circle) and sagittal direction (like wheel spokes) measurements. I concentrated on just the 32 resolution readings around the center of the lens (both meridional and sagittal), since that’s where I had the camera focus point placed. Test Results The Pinpoint focus data from the 19 test shots had a peak MTF50 reading of 72.7 lp/mm, with an average of 69.2 lp/mm. The standard deviation was 1.93 lp/mm.  Very consistent focus. The 3D-tracking focus data from the 19 test shots had a peak MTF50 reading of 72.7 lp/mm, with an average of 68.5 and a standard deviation of 3.24 lp/mm. Note that both focus modes obtained the same peak resolution of 72.7 lp/mm, but 3D focus was a little sloppier and averaged slightly worse resolution as a result. I have to admit that Pinpoint focus is in fact more consistent, and on average gets marginally sharper shots. The spread of data is pretty small, and you couldn’t notice any sharpness differences by looking at the actual test chart photos for either the Pinpoint or 3D shots. The MTFMapper software is super picky, and it notices subtle differences you can’t perceive yourself. In case you wondered, in the past I compared 3D-tracking against other AF-C focus modes and I didn’t note any particular sharpness differences in the various modes. I just prefer how frequently 3D will hold focus on the desired subject, compared to other modes. The Nikon Z8 and Z9 camera autofocus accuracy is very good and quite fast, no matter which focus mode you choose. Since these cameras focus at the shooting aperture (through f/5.6), it also means that you don’t pay a sharpness penalty using lenses with focus-shift (spherical aberration), either. If you’re after the consistently sharpest shots you can get, then Pinpoint focus is the way to go. If you happen to have the Nikon Z8 camera, then combine Pinpoint focus with pixel-shift shooting for really sharp shots when using a tripod and there’s no subject movement (use either the 16 or 32-shot options).

So what exactly is lens diffraction? Why do photographers hate it so much? How do you get rid of it? That’s the mystery that this article will unravel. That evil Airy disc Camera lenses, even if they’re built to absolute perfection, still make fuzzy images when you stop down their apertures too much. A pinpoint of light, after it travels through your lens aperture on its way to your camera sensor, gets ‘diffracted’ . Instead of hitting the sensor as a pinpoint, that spot of light ends up looking like the picture above. This light-dark-light circular pattern is called an ‘Airy disc’. In three dimensions, the Airy disc would look like ripples after a stone gets dropped into calm water. A guy named George Airy (1801-1892) first developed the mathematics for this diffraction phenomenon, and from then on it’s been known as an ‘Airy disc’. Sir George Airy was a professor of mathematics at Cambridge University. He became an expert in Latin, ancient Greek, architecture, astronomy, and engineering, just to name a few of his skills.  He even supervised the construction of London’s Big Ben chimes. But I digress. If you look at the picture above, the exact size of the Airy disc is ambiguous.  It just gets dimmer and dimmer at the fringes. When a lens aperture is stopped down, the size of this Airy disc starts growing in diameter. The Airy disc diameter is only a function of the aperture f-number and the color (frequency) of light. When apertures are idealized as being a perfect circle, the Airy disc diameter, measured in microns, can be estimated to be 1.34 times the aperture f-number for green light (549nm or 0.549um). For blue light, for instance, the Airy disc diameter is smaller. Green light can be between 500nm and 600nm, but 549 was chosen here. The Airy disc formula is: Airy_disc_dia = 2.44 * frequency_um * F_stop For the above, frequency_um = 0.549 This is why integrated circuits that get made by projecting an image onto silicon use ultraviolet light. This very high frequency light (short wavelength) produces a really small Airy disc diameter. The Airy disc is the culprit in making even ‘perfect’ lenses produce soft images, if they get stopped down far enough. For you to notice the image getting fuzzy, the size of this Airy disc has to grow until it covers more than a single pixel on your camera sensor. A rule of thumb is to start getting concerned about diffraction when the Airy disc grows to be two pixels across or more. For making prints, this rule can be loosened up considerably. For camera sensors that have anti-alias filters (to help rid any Moire effects) the images are even fuzzier. For an example, I’m going to pick on the Nikon Z9 camera, which has pixels that are 4.35 microns and NO anti-alias filter. Two pixels, then, cover 8.7 microns. We don’t care about the overall size of the sensor or how many megapixels it has, either; we only care about the distance between one pixel center and the next pixel center. If that pesky Airy disc covers a pair of pixels, then diffraction can be seen. Diffraction versus F-stop As shown above, when the lens (any focal length) gets stopped down to f/8 or narrower, the Airy disc diameter grows larger than two pixels (8.7 microns for Nikon Z9). As soon as this 2-pixel threshold is reached, some image softening starts. At f/8, diffraction is barely noticeable. Diffraction starts growing in leaps and bounds by f/16, and image quality suffers. Don't even ask about f/32. You have to decide if getting that large depth of focus is worth it. On cameras that support focus-stacking, it’s ideal to stick with f/5.6 or wider and take multiple shots to later combine them to get a large depth of field at optimal resolution. Just pick your lens’s sharpest aperture, and hope there’s no wind. Tripod required. You can manually re-focus between shots, if your camera doesn't have focus stacking. This is a trick to sidestep diffraction effects when there's no subject movement. If you buy a camera with huge pixels, you might get away with one or two extra f-stops before diffraction rears its ugly head, but eventually it will show up. When you hear the phrase “diffraction–limited”, it means that any lens aberrations except for diffraction have been essentially eliminated, so that any remaining aberrations all fit inside that darned Airy disc.  Making the lens optics even more perfect is pointless. Diffraction is always lurking. This lens fuzziness can’t be solved by money. It’s physics. Thanks, George (I guess).

Here’s an idea that probably most photographers have never explored: reflections by themselves. Nearly everybody has shot subjects that include reflections in water to get a nice symmetry. How about skipping the actual subject and just photograph it’s reflection in isolation? The main technique here is to either flip or rotate the reflection to get it upright. You don’t want a ‘perfect’ reflection, either, because then it would just be a repetition of the subject. I like to shoot reflections in infrared as well, to create an even more unique look. With color infrared, I like to switch around the color channels or make hue shifts, so that the sky is blue. I like using the Capture One editor when working on color infrared, and a link to how I use this editor for infrared is here. With water reflections, timing is everything. If the water’s surface is still, the reflection is boring. If it’s windy, the reflection is ruined. If the water has minor ripples, you’ll never get the same shot twice. Contamination floating on the water’s surface can give the illusion of a starry sky. May the wind not be at your back. 590nm infrared reflection, edited in Capture One 590nm infrared reflection, edited in Capture One

When I got my Nikkor 500mm f/5.6 PF lens, I was excited to try it out with my 1.4X teleconverter (mine is the Sigma TC-1401). This would give me a 700mm f/8 lens. At the time, my best camera was the Nikon D850, which reviewers have praised for its auto-focusing capabilities. Outside in bright light, this 700mm combo worked just fine, but then a bird flew into the shade. Disaster ensued. Nikon Z8 with 500mm f/5.6 PF and 1.4X teleconverter Nikon claims that the D850 autofocus system is the same as what’s in the D5 and D500 cameras, and focuses down to -4EV. In my own experience, I haven’t gotten even close to this. Additionally, they don’t specify how slow the autofocus gets when illumination gets low. I think that somewhere in the fine print Nikon mentions something about this only working with an f/1.2 lens, and I’m fresh out of f/1.2 700mm telephotos. The Nikon Z8 and Z9 cameras are specified as being able to autofocus down to -7EV (-9EV in Starlight mode). I just know that with a real telephoto in actual low-light shooting conditions, it just plain works. I quickly gave up using this unreliable D850/700mm combination, and instead stuck with the 500mm f/5.6 by itself. As much as you try, animals just don’t obey your wishes and stay in good lighting conditions. It occurred to me that I haven’t re-tried using this 700mm setup with my Nikon Z8 or Z9 cameras; big mistake on my part. Trying to crop 500mm shots to what a 700mm combination (500 + 1.4X teleconverter) gives you is never as good. I’m not saying crops of the 500mm shots are bad, but the quality takes quite a bit more of a dip than using a teleconverter. With birds especially, it seems like you can never have too much focal length. Tracking that bird at 700mm is quite another story, however! Before wasting time doing a potential known-to-fail test, I tried out the Z8 with the Nikkor 500mm PF with the 1.4X teleconverter on birds under heavy clouds conditions. Amazingly, the lens focused extremely quickly and didn’t appear to suffer from focus-hunting, either. The quality of light in these conditions can be really nice, which is why I’m interested in shooting like this in the first place. Beware of shooting flying birds against a cloudy sky, however; you risk getting nothing but near-shilouettes or else burned-out backgrounds. Z8 with Nikkor 500mm f/5.6 PF and 1.4X teleconverter The shot above is 1/2000 shutter, f/8, ISO 9000 with the 700mm combination. This shot hasn’t had any light or color adjustments, to demonstrate the quality of light in cloudy conditions. This light level equates to EV 9.4, and the Z8 focused the lens quickly in these conditions. The D850 does quite a bit of focus-hunting in these conditions, and successfully finds focus only about half of the time. The Z8 viewfinder is bright, while the D850 viewfinder is quite dim in this deep shade. Also, the D850 focus points away from the viewfinder center are even less sensitive and capable, although Nikon claims that 15 of the points support f/8 lenses. I might mention that the Z8/Z9 cameras also have focus ‘detection’ modes, while the D850 doesn’t.  I used bird-detection in the above shot, where the camera immediately finds the near eyeball of the birds. In addition, I no longer have to worry about autofocus fine-tune calibration, which is different for all of my DSLRs and different with/without a teleconverter, too. I decided to do a comparison of the Z8 (equivalent to the Z9) and the D850 in cloudy and then indoors conditions. As usual, I do my testing by using a 120-frames-per-second video while filming the lens focus distance scale. Test #1: 3 meters start, 100 meters finish, EV 11.5 The first test was to focus on a distant tree in cloudy conditions, so that the target had good contrast. I knew that both the Z8 and D850 could focus on this, but I was interested in both focus speed and any focus-hunting that might occur. I set the focus ring on 3 meters (10 feet), and took a video while the lens focused at about 100 meters. A sample photo of this target was 1/2000s, f/8, ISO 2200 (EV 11.5) The D850 took 52 video frames (0.433 seconds). I noticed that the D850 over-shot the target, backed up to 21 meters, and finally focused at about 100 meters. The Z8 took 41 video frames (0.341 seconds). This camera had the lens hesitate at 14 meters, and then finished focusing on the target at 100 meters. Even the Z8 (and therefore the Z9) could stand some improvement in how it acquires the target focus, but I can’t complain about the 0.341-second result. This amount of focus change is much more severe than you’d normally encounter in the field, where it would typically change focus  over a few meters nearly instantly. Test #2: 3 meters start, 8 meters finish, EV 6.6 (indoors) For a much tougher test, I shot a medium-contrast target indoors with only some window lighting. A photo check here indicated 1/200s, f/8, ISO 12,800 for an Exposure Value of 6.6. The room I tested in wasn’t 100 meters long, so I settled for a target at 8 meters away. The D850 totally failed to find focus, and just hopelessly cycled back and forth between minimum and maximum focus distance. No surprises here. And so much for that -4EV spec. The Z8 accomplished the 5-meter focus change in 32 frames, or 0.267 seconds. Really, really fast. There was no hesitation in focus during this test with the 700mm combination. Also, this camera doesn’t care where you place the focus point; it works the same everywhere. That’s very impressive indeed. Summary The Nikon Z8 and Z9 cameras are just amazing in their focus capabilities in poor lighting conditions. You don't even need to worry about lower-contrast subjects. The DSLRS just can’t compete. Honestly, I thought the contest would be closer than it turned out to be. Using my 500mm with a teleconverter is no longer a problem under any but the dimmest of lighting conditions! 700mm 1/2000s f/8 ISO 9051, EV 9.4

This lens is advertised as an APS-C lens, but I’ll let you be the judge. The TTArtisan lens designers knew that the fringes of this lens wouldn’t be stellar, so they gave it a huge image circle that extends all the way to cover an FX sensor. I’m reviewing the version made for Nikon with the Z mount, but you can get it in many different mounts. Personally, I haven’t been interested in buying DX lenses, since they waste half of my FX sensor. This lens is an exception, because it does in fact cover the FX sensor. If I don’t like the outer fringes of a shot, I just crop to it to suit; you can’t stretch it if you shoot in DX mode. Unless you want to spend $8000.00 for Nikon’s 58mm f/0.95 Noct or $13,000 for Leica’s 50mm f/0.95 Noctilux-M, you might want to take a look at this TTArtisan 50mm f/0.95 lens. All three of these lenses are manual focus, by the way, and the TTArtisan costs a bit less (less than 2 percent of the Noctilux). You could actually throw in a couple of Nikon camera bodies with this lens instead of getting the Noctilux with an adapter. Am I saying that this lens is just as good as the Nikkor or Leica? Heck no. But if you’re after a specialty lens for portraits with a melted background, then read on. This is the very definition of a niche lens. If you’re after distant landscape shots, then run away from this TTArtisan. Try to imagine you’re more of a painter than a photographer when shooting, and it will put you in a better mindset. And forget about sharp corners. You don’t get a lens hood with this lens, so check out something like Amazon for a cheap 58mm screw-on lens hood; some of them also have pinch snap-on lens caps for the hood. You’ll want a lens hood to minimize lens flare. I also bought a snap-on lens cap that fits onto the hood, instead of using the provided screw-on lens cap. This lens is 1/6 stop faster than an f/1.0 lens, or 1 1/6 stops faster than an f/1.4 lens. That’s 2/3 stops faster than an f/1.2 lens. Lens Specifications 50mm, f/0.95 through f/16.0 14.5 ounces/ 411 grams 58mm filter thread Minimum focus: 19.7 inches/ 50cm 8 elements in 6 groups 10 rounded aperture blades Half-click stops f/0.95 through f/4, then full-stops (skips f/11). Manual focus only Focus ring rotation range 135 degrees No electronic contacts Metal and glass, period. TTArtisan 50mm f/0.95 lens design This version of their lens doesn’t have any aspherics; just a pair of high-index glass elements. That’s 8 elements in 6 groups. Usually, the lack of aspheric lens elements means that the bokeh is smoother. TTArtisan 50mm at f/0.95 (on left) versus Nikkor 85mm AF-S at f/1.4 (on right) Compare the two shots above. Both pictures were taken at their widest apertures, and I tried to get the same image magnification in both shots.  The Nikkor has much better resolution, and it has a very different look to it than the TTArtisan does. For a subject like this, I prefer the TTArtisan’s look; it’s more like a painting and unconventional. Your opinion may differ. f/0.95 Outdoors The Nikon Z9 and Z8 have a 1/32,000 shutter speed, so sunshine and f/0.95 work just fine together. For slower cameras, you’ll need to get yourself a neutral density filter for those wide-open daytime shots. Set up the Non-CPU Lens Data The first thing to do is to set up the non-CPU lens data in the Setup menu. This lets the camera get proper exposure and correctly use the IBIS system for anti-vibration. The EXIF data will now include what lens is being used, and what its maximum aperture is. The EXIF data won’t indicate the actual f-stop in use. Use Focus-Peaking, Please My mirrorless Nikons have great focus-peaking, which allows for very quick focus confirmation with manual focus. The ‘low sensitivity’ (1) setting gets the most accurate focus. The viewfinder image magnification lets me really nail focus when the subject holds reasonably still. Both the Z8 and Z9 cameras have the “Half-Press to Cancel Zoom (MF)” feature, to instantly let you frame the subject after proper (magnified) focus. Just half-press the shutter button after you focus on your subject to see the whole frame. The focus direction is backwards, compared to Nikkor lenses. And speaking of focus, the focus action is very smooth, but I wish the rotation was a bit more than the 135 degrees it has. The focus ring has sculpted indents in the metal, instead of the traditional textured rubber ring. Repair Distortion, Vignette, and Color Fringes Since TTArtisan hasn’t provided any lens correction profile (yet), you’ll need to manually fix various lens issues in your editor. You can save the fixes into a ‘profile’, which you can apply to subsequent photographs.  At these prices, you didn’t think that you’d get away with no image editing, did you? Image Distortion Distortion is rather pronounced, but it can be made to virtually disappear with suitable image editors. A sample lens profile manual setting in Lightroom is +15 to fix barrel distortion. In Capture One, the “SHAPE” section has a ‘Distortion’ slider in the Lens Correction | Lens area; I used 52. f/0.95 barrel distortion and vignette, full FX frame f/0.95 repaired distortion and vignette in editor f/5.6 un-modified distortion and vignette Vignette Vignetting is severe when wide open, but not that much different from the Noctilux lens. Again, use an image editor to eliminate this when it’s a problem. Many photographers actually increase vignetting with their editing software, since it can really enhance image aesthetics. A sample lens profile manual setting in Lighroom is +100, to get rid of vignette. Capture One, in the “SHAPE” section Lens Correction | Lens area, there’s the “Light Falloff” slider to fix this (84). Vignette reduces significantly when you stop down. Lateral Chromatic Aberration (CA) Lateral chromatic aberration f/0.95 Lateral chromatic aberration f/5.6 The CA at f/0.95 peaks at about 8 microns, which equates to 1.8 pixels. At f/5.6, the CA reduces to 1.1 pixels. I doubt you’ll notice it. Fix this “purple fringing” problem using your image editor. In Lightroom, I use the ‘Manual Lens Correction’ Defringe eyedropper. In Capture One, the ‘Refine’ tab has the Defringe slider in the ‘Purple Fringing’ section, plus a smart analysis in its Lens Correction section. Lens Flare Yes, flare is thare, but it’s not too bad. I purchased a separate lens hood to shade the lens and minimize it. Into the sun at f/0.95, full FX frame Into the sun at f/16, full FX frame Bokeh In a word, this lens is about bokeh. That expensive Leica lens has strange “half-moon” out-of-focus lights at the image fringes, while both this TTArtisan and the Nikkor Noct have symmetric “cat’s eye” lights at the fringes. Interestingly, the quality of the background heavily depends upon focus distance. It’s best at ‘portrait’ focus distances. At medium distances, you get a strange increased sharpness around the FX edges, which isn’t typically pleasant. TTArtisan cat’s eye at f/0.95 Contrast Wide open, and particularly at close focus distance, image contrast is reduced. Use your favorite editor to increase contrast to help fix this.  Stopping down even a little will enhance the contrast, which is the same for any really fast lens. Coma Coma is just plain bad in the edges and corners, but can be drastically decreased by stopping down the aperture. Never, ever try shooting the stars with this lens; you’ll have a hard time trying to un-see that coma later. Sharper, Please I love to use my Topaz DeNoise AI to sharpen the shot and also rid any image noise. Shots using the TTArtisan really benefit from this sharpening treatment. MTF Contrast Claimed MTF Contrast, from TTArtisan (24mm wide DX sensor) What’s shown above is the TTArtisan prediction for a 24mm-wide DX sensor at both f/0.95 and f/5.6. Measured MTF contrast, f/0.95 The measured MTF50 contrast curves are generally lower than the theoretical curves for the DX sensor range. My measured curves show why the TTArtisan engineers chose to call this a DX lens instead of an FX lens, when you look beyond 12mm from the image center. Measured MTF contrast, f/5.6 This lens shows a pronounced sharpness increase near the edges of the FX frame when stopping down. This leads to some strange-looking distant shots, where edges unexpectedly sharpen. Field Curvature Before I talk lens resolution, a disscussion on field curvature is in order. My testing is all done treating this lens as if it was made for an FX sensor. Since TTArtisan considers it an APS-C lens, measurements outside that area are a bit unfair. I have found that these optics have extreme field curvature, where the edges have the focus "plane" move away from the center. This means that the resolution near the edges will look really bad, since those tests assume the focus plane to be flat. The shot that follows was enhanced using the Photoshop "Find Edges" feature, which effectively shows where the focus "plane" really is. The focus 'Plane' looks like a Greek omega symbol The shape of what's in focus is quite distorted. I drew a green rectangle around the approximate DX (APS-C) sensor boundary. I drew a red line along what's in focus at f/0.95. This shot is a rug, photographed at about a 30-degree angle down from the horizontal. If the lens was perfect (no field curvature), then the in-focus portion would be a simple horizontal line across the shot. Instead, the focus (in the portion outside the DX sensor boundary) curves away from the camera. This severe field curvature is the worst aspect of this lens when you treat it as if it was made for FX cameras. At least now you'll understand in the resolution measurements that follow why the edge measurements look so horrible. Resolution I shot the resolution test charts in FX mode, but I drew a rectangle around the DX sensor area, to show what TTArtisan wants you to use. I shot the test chart from a distance of 5 feet (1.5 m). The lens center is excellent by f/2.8 and peaks at f/4. The frame edges don’t become decent for FX until somewhere around f/11 (no aperture click stop or marking for f/11). The whole frame is acceptable around f/8 for the DX area. FX corners don’t ever quite fully sharpen, but they get close at f/16. The FX frame edge sharpening after the mid-frame resolution plunge causes strange-looking landscapes on the frame edges. This is why I wouldn’t recommend this lens for general-purpose distance shooting, at least when you go beyond DX. If you stop down a little beyond f/8, then distant FX shots start to look ‘normal’ again. You’d never buy this lens for its resolution. It sounds like an excuse, but this lens really is about its ‘look’ with its bokeh and narrow depth of focus. I suspect that not using aspheric elements in this design caused more sharpness issues across the field of view, but enhanced the look of the bokeh. f/0.95 MTF50 lp/mm resolution, DX frame in green f/1.1 MTF50 lp/mm resolution, DX frame in green f/1.4 MTF50 lp/mm resolution, DX frame in green f/2.0 MTF50 lp/mm resolution, DX frame in green f/2.8 MTF50 lp/mm resolution, DX frame in green f/4.0 MTF50 lp/mm resolution, DX frame in green f/5.6 MTF50 lp/mm resolution, DX frame in green f/8.0 MTF50 lp/mm resolution, DX frame in green f/16.0 MTF50 lp/mm resolution, DX frame in green Samples f/0.95 full FX frame f/0.95 full FX frame f/0.95 Corrected for distortion and vignette, FX frame f/0.95 full FX frame f/0.95 full FX frame f/0.95 full FX frame f/0.95 full FX frame f/5.6 full FX frame. Edges are okay, but not corners f/0.95 FX frame. Edges show unusual increased sharpness Summary Let’s assume you have the cash to splurge for that Noct or Noctilux, but you hesitate to take the plunge. You could get this lens and play with it for almost zero investment (relative to those expensive lenses). If you decide that you really enjoy this photographic genre, then go ahead and Noct-splurge. If, however, you quickly lose interest in f/0.95 and manual focus, then you’d know to stay away from the dark side. I find that I have lots of fun with this lens, although not as a steady diet. It’s more like a high-calorie treat. Yes, it’s rough to focus on close moving targets at f/0.95; it’s best to do at least 5 fps and throw away the out-of-focus shots later. You find yourself more in the moment, having to manually focus while framing subjects. It’s easier to make suitable subjects look like “art” with this lens at f/0.95. If you were to notice that most of your shots are taken at f/5.6, then this lens is a waste; get a regular nifty fifty instead.  Of course it would be nice to have more even sharpness away from the frame center, but at the same time it gives you a more unique character to the pictures. Focus on what this lens can do and not what it cannot do. For me, I’ve decided that this lens is definitely FX and not DX. Minor cropping is called for on some shots, but that’s no different than for any other lens. At least the FX edges are there to crop.

Does that older lens deserve a second look with a newer camera? Eight years ago, I did a review of my Nikkor AF-S 85mm f/1.4 G lens. Back then, my newest camera was the 16 megapixel Nikon D7000 DX camera, so I tested the lens on that model. It’s well known that a high-megapixel camera should give you better resolution with a given lens, compared to a low-megapixel camera. Since my D7000 camera days, I have obtained better gear and thought it would be interesting to re-evaluate that old 85mm f/1.4 Nikkor on my Nikon Z9 mirrorless camera. I’m using the FTZ-II adapter on the Z9, of course. Nikkor 85mm f/1.4 G mounted on Nikon Z9 As camera technology improves, you can get better results from a lens in more ways than just better resolution. Especially with the switch from DSLR to mirrorless, you get better focus accuracy without having to even bother with focus calibration. With fast lenses like this 85mm, focus shift problems with aperture changes are finally cured, since the mirrorless Nikons focus at the shooting aperture. Mirrorless cameras also focus better in dim light than DSLRs, since they don’t have focus sensors hidden behind partially-silvered mirrors. Another huge advantage is having in-body image stabilization (IBIS) on the mirrorless camera, since this lens doesn’t have VR. The Nikkor 85mm f/1.4 AF-S has been the highest-ranked F-mount Nikkor lens at the DXO web site since they first measured it about 13 years ago. It finally got beaten by some of the Z-mount Nikkors; the highest-ranked lens from any manufacturer is currently the Nikkor 85mm f/1.2 S, which costs about $2,600. My 85mm f/1.4 still sells for $1,600 (it cost $2,200 at introduction). As of this writing, this F-mount 85mm is most likely doomed: Nikon is probably done making any more F lenses. My primary interest in re-testing this lens is resolution. Since my original testing was done using a DX sensor, I’m particularly interested in how bad the edges look with a high-resolution FX sensor. Even though the Nikon Z9 has 45.7 megapixels, this lens isn’t quite good enough to reflect all of that additional sensor resolution. Nonetheless, the resolution measurements had better increase using this camera! I found that focus repeatability was indeed much better using the Nikon Z9 (and the Z8) compared to the old D7000, but it wasn’t infallible. I have found that single-point focus is the least reliable way to focus this lens, which seems opposite of what you’d expect. Focus speed is a little bit better on the Z9 and Z8 cameras than the D7000, but not by too much. The 85mm is fairly slow to focus, since it’s optimized for focus accuracy. There are better lens choices to evaluate focus speed than this one. If you want to try manual focusing, the focus-peaking feature on the Z9 and Z8 cameras is really nice. The ability to magnify the image while looking through the viewfinder to really nail focus is also worlds better than using Live View on the old D7000. Resolution Testing I re-ran my old Nikon D7000 raw-format resolution test chart photos, using the newest version of my MTFMapper software. These D7000 test shots were created using an older version of a resolution test chart, but that chart is the same physical size and created using the same printer, paper, and ink as my new resolution test chart. The chart dimensions are 40” X 56”, which provides a very useful focus distance of about 13 feet, or 4 meters using an FX sensor. I used my new-design resolution test chart for evaluating the 85mm on the Z9 camera. The MTFMapper program is happy using both the old and new chart formats, and the results are equal in precision. 85mm at f/1.4 MTF50 resolution on D7000 85mm at f/1.4 MTF50 resolution on Nikon Z9 As shown above, the lens resolution increased using the Nikon Z9 camera by about 14%. This is a lot less improvement than you might expect. What it shows is that the lens itself doesn’t have enough resolution to make a major impact on better resolution results with high-megapixel sensors. What’s interesting to note, however, is that the edges of this lens don’t suffer much from a loss of resolution when increasing from DX coverage to FX. The corners have the very slightest dip in resolution. I think that my copy of this lens has a slight lens tilt; the right-hand side always seems to have slightly better results than the left-hand side, at least when shot wide-open. I have always maintained that an MTF50 resolution of roughly 30 lp/mm looks “sharp” and I’m sticking with that position. Using programs like Topaz DeNoise AI really helps enhance the sharpness, too. This means that the 85mm f/1.4 can be used wide-open, if desired. You should be more concerned with background bokeh and depth of focus decisions; in other words, concentrate on composition and art instead of sharpness. MTF contrast plot (actual measurements) f/1.4 The plot above shows how much real astigmatism the lens has, starting from the lens center. Nikon doesn’t ever show you actual measurements, just ‘theory’. 85mm at f/2.0 MTF50 resolution on D7000 85mm at f/2.0 MTF50 resolution on Nikon Z9 At f/2.0, the Z9 managed to get about a 15% resolution improvement over the D7000 camera. You can see the corners take a resolution dip, since the FX sensor sees so much more of the lens fringes. The lens takes a significant resolution jump going from f/1.4 to f/2.0 of about 38%. Now, it’s getting closer to ‘modern’ lenses for resolution. 85mm at f/2.8 MTF50 resolution on D7000 85mm at f/2.8 MTF50 resolution on Nikon Z9 At f/2.8, the Z9 saw about an 18% resolution increase overall, compared to the D7000 camera. The corners and edges of the FX sensor are excellent here. You may have started to notice a trend; the mid-frame resolution performance always seems to be a bit better than the center. I have read that the Nikon engineers made a design decision to sacrifice some center performance to enhance the mid and edge performance. Nikon chose to use no exotic lens elements here; there aren’t any aspheric elements. The lack of aspheric elements means that there is a small sacrifice in what could have been done with smooth resolution, but that would also have probably meant worse bokeh. 85mm at f/4.0 MTF50 resolution on D7000 85mm at f/4.0 MTF50 resolution on Nikon Z9 The Z9 has an increase in resolution of about 22% overall, compared to the D7000 at f/4.0.  Resolution across the frame is really, really good. At this aperture, it’s now competitive with modern lenses. 85mm at f/5.6 MTF50 resolution on D7000 85mm at f/5.6 MTF50 resolution on Nikon Z9 This lens reaches peak performance at f/5.6, which is just a bit better than f/4.0.  Compared to the D7000, the Z9 is about 15% better. 85mm at f/8.0 MTF50 resolution on D7000 85mm at f/8.0 MTF50 resolution on Nikon Z9 The Z9 is roughly 14% better than the D7000 at f/8.0.  The resolution dip, due to diffraction, has begun. Summary Using this lens on a Nikon mirrorless, such as the Z9, does indeed improve the lens resolution results, although not as much as people probably expected. The biggest benefit is being able to nail focus far more often. Even after all these years, the Nikkor 85mm f/1.4 AF-S is a great lens. It’s a favorite of many portrait photographers for good reason. It’s pretty sharp, has great bokeh, and can provide very thin focus depth when needed. This lens has aged better than most. I feel that this lens’ biggest problem has always been focus shift when changing the aperture (spherical aberration). Now that the Nikon mirrorless cameras focus at the shooting aperture, that problem has been solved. This lens’ biggest strength is the overall edge-to-edge resolution balance, combined with really good bokeh. My Sigma 70-200 f/2.8 Sport lens, for instance, smokes this lens for resolution, but the 85mm beats it for bokeh and of course being able to get to f/1.4. I always missed having vibration reduction with this lens; with the Nikon mirrorless cameras having IBIS, that problem has disappeared. The new Nikkor 85mm f/1.2 S lens is admittedly better in every category (except price), but the performance of my 85mm f/1.4 is so good that I personally don’t see the need for an upgrade. Switching to mirrorless cameras has really been a delight. My D850 DSLR sensor is just as good as the Nikon Z9, but the advantages of mirrorless can’t be denied. It’s really amazing to see the progress in cameras since the D7000. There’s virtually no aspect of that camera that hasn’t been usurped. My old lenses seem more like new ones when I mount them on my Z8 or Z9 cameras.