I have tried a few different programs that let you increase depth of focus by stacking pictures that are shot at varying focus distances. Most of those programs will readily fail when the subject is too complex or the focus depth is too extreme. Focus stacking is mostly used in two different realms, namely landscapes and macro photography. Landscape photographers usually want maximum depth of field and maximum resolution, which can be had by stacking photos shot at the sharpest aperture and at multiple focus distances. Macro photographers know only too well that a single close-up can have paper-thin depth of focus; combining a dozen or more shots is often necessary to get sufficient depth of focus. I have had too much grief using Photoshop and Hugin tools, but a (free) program that works pretty well for me is called CombineZM by Alan Hadley. Stacking pictures requires a lack of image movement, so wind can mess up your plans. The pictures need consistent exposure, so manual exposure works best (at a constant aperture). For macro work, I like to use my Nikon PB-4 bellows (with its rack-and-pinion focus rail) to easily move the camera/lens combination from shot to shot, shifting focus by maybe half of a millimeter per shot. Good luck finding a Nikon PB-4 bellows. The default settings in CombineZM don’t always work the best for me, so I thought I’d share how I make it work for me. It bears mentioning that “macro” in CombineZM means “run a sequence of steps” and not anything to do with close-ups. My most successful recipe to stack pictures is this: Post process and convert your pictures into TIF format (16- bit with LZW compression is what I use) using your favorite image editor. Run CombineZM (I use it in Windows7 and Windows10, but it works in other operating systems, too). Select File | *New and then select the set of TIF pictures to stack (select in focus-order). Wait until the pictures are loaded. Select Macro | Do Weighted Average. I have less success using “Do Stack” or “Do Weighted Average Correction”. After it finishes, use your mouse to draw the diagonals of a rectangle around the ‘good’ part of the result. Select File | Save Rectangle As. I just save the result as JPG, with typically 95% quality. Here’s a sample finished shot, which is from a stack of 10 files: “Do Weighted Average” macro to create the stack I actually took even more shots in front and behind of what was used above. The software started to mess up with this many pictures, so I omitted some shots to achieve the result shown above. “Do Weighted Average” with even more pictures in the stack. Note the evil ‘ghosting’. “Do Stack” macro. Note strange artifacts using this option. What a focus stack looks like before you crop it. Typical single shot depth of focus (60mm f/10) Depending upon your subject, you may have to iterate on the selected options or perhaps how many pictures you can stack. Life is rarely simple… Make sure your subject is a bit smaller than the frame, because you will have to crop the edges of the “stack”. Conclusion Focus stacking is one of those techniques that takes a little tenacity. There are many different tools that can stack pictures, with varying degrees of success (or failure). This is one of those digital tricks that seems to defy optical physics. If you're willing to put in the effort, the results can be quite rewarding. #howto

All camera companies (with the exception of Sigma and Leica) publish MTF curves for their lenses that are “theoretical” and not actually measured. Should you care? Personally, I believe in the old President Reagan saying “trust but verify”. What follows is a dose of reality, compared to theory. I have chosen what most people would agree are among Nikon’s best pro lenses for this study, lest I get accused of measuring lenses that were manufactured using lesser standards. The MTF curves I’m referring to are the traditional mix of MTF10 (contrast) and MTF30 (sharpness). I used the “mtfmapper” software version 0.5.8 to create the following charts. Personally, I place much more stock in the 2-dimensional MTF50 plots that measure the whole camera sensor. Unfortunately, getting 2-D MTF50 plots is hard to come by outside of this site. The MTF charts are traditionally generated for a wide-open aperture, so that’s how mine are measured. It’s unknown what focus distance is used by Nikon; mine will be measured at the distance needed to photograph an “A0” resolution chart filling the frame. I took the measurements in shade on a clear sunny day. Light wavelengths can affect measurements; I like to test using the same lighting conditions that I normally shoot. 105mm f/2.8G ED‑IF AF‑S VR Micro Nikkor This lens is supposed to be optimized for “close” distances, but I’m measuring it at a more conventional distance. Nikon Theoretical Chart (from Nikon site) for 105mm Measured MTF10 and MTF30 for 105mm at f/2.8 I don’t want to appear cynical, but I was 99% sure that my measurements would show less sweetness and light than the Nikon claims. This is pretty much borne out by the measurements. Take a look at the edge of the lens, though. It actually performs better than theoretical! Measured MTF10 and MTF30 for 85mm at f/1.4 Again, not quite as good as theoretical. The edges have a few pleasant surprises, however. 85mm at f/4.0 Just for fun, I tried an f/4.0 test. It really cranks up the quality, doesn’t it? 24-70mm f/2.8E ED VR AF-S Nikkor The wide end of this lens looks dramatically different than theoretical. Again, this lens at 70mm looks quite a bit different than the claims. Conclusion It appears that Reagan had some good advice. Bear in mind that these lenses don’t represent the whole population; your mileage may vary. My biggest surprise is that the FX frame edges fared better than expected. Trust but verify. #review

This version of MTF Mapper has some new features and some changed features. This is the software that I use to evaluate both lens resolution and focus calibration. The author of this program is Frans van den Bergh. You can get this software here: https://sourceforge.net/projects/mtfmapper/ This new revision can still use the original resolution and focus chart designs, which is a real relief if you have invested time, effort, and money in printing/mounting large versions of the charts. If you print the newer charts, you get some new and welcome abilities. What’s changed? You may want to review my MTF Mapper Cliff’s Notes article, detailing the older version capabilities. New Resolution Chart The biggest change as far as I’m concerned is the switch from ‘relative’ measurements to ‘absolute’ measurements in the resolution charts (grid2d and grid3d). The chart scales of earlier MTF Mapper versions would only have a value range matching the actual measurements, but now the resolution range starts at zero. Another big change is the switch to monochrome color coding in the 2d and 3d charts, instead of the ‘rainbow’ color coding that would auto-scale to the entire measurement range. A small but welcome change is the addition of the photo name under the chart, so you know where the chart came from. Original resolution chart design. You can still use this chart. New resolution chart design, showing annotated photograph. New chart up close. Good edge measurements are blue; “iffy” ones are in yellow. 2D Chart with absolute scale for MTF50 lp/mm Older 2D Chart measurements with relative scale for MTF50 lp/mm. 3D new resolution chart showing the absolute (monochrome) scale MTF 10 and MTF 30 Graphs Camera companies have traditionally published MTF charts that show 10 lp/mm and 30 lp/mm “theoretical” values. I stress “theoretical” here, because those companies are merely blowing smoke. They don’t actually measure anything (at least Canon and Nikon don’t). MTF Mapper can now plot the real-deal MTF10 and MTF30 charts, based upon actual reality. What a concept. These graphs use the same chart design used for "grid2d" and "grid3d". MTF10 and MTF30 measurements using the new resolution chart Focus Chart The new software can use the original focus chart. What’s new is how the ‘annotated’ version of the chart displays the measurements. The measurements are in “cycles per pixel”. The measurements are no longer embedded inside boxes, which makes reading the values and seeing the edges much easier. Focus chart photo, showing the annotated edge measurements “Profile” option for focus chart The chart above shows a very slight focus error. The chart is oriented to make the left side farther from the camera than the right side. The ideal angle to shoot the chart is at 45 degrees relative to the vertical. The measurements above would indicate that the camera (or lens firmware) needs some “-” focus-tune adjustment, to pull the focus toward the camera. I always recommend, by the way, to look at the annotated focus chart measurements. There are occasions when the numbers give you a better idea of how to adjust focus. Also, repeat this test several times to avoid reacting to normal focus variations. Lastly, perform the tests in good light for optimal reliability. In the focus test above, the camera focus point was placed onto the right edge of the large central trapezoid. Because the chart is rotated, the trapezoid looks like a rectangle in the photograph. Measure Longitudinal Chromatic Aberration Focus chart with “fiducials” for measuring longitudinal chromatic aberration Chart zoomed in. Shows green channel focus error. The other major feature addition is the ability to analyze how a lens focuses in the red, blue, and green channels. When the different channel color focus measurements don’t coincide, then you have longitudinal chromatic aberration. To create the charts shown above, the MTF Mapper needs to be configured as shown in the following picture: Preferences dialog. Note the camera sensor “pixel size” must match your camera. Summary The new MTF Mapper version 0.5.8 brings many welcome additions. You might want to retain your older version, however, if you prefer the “relative” versus the “absolute” resolution measurements. Please visit the Frans van den Bergh site and give him some praise for going through all this effort. Frans, you’re the man! #review

Blue Angel Daredevils Photographers lust after that pro camera model with those high frame rates, so that they won’t miss that crucial shot. Guess again. Let’s say you just got that new D500 and you dialed in 10 frames per second. How could you miss now? Well, let’s look at some simple math. The above shot shows two jets which are cruising at about 500 miles per hour. Their closing speed is 1000 miles per hour, which is about 0.28 miles per second, which is equivalent to almost 1500 feet per second. That D500 taking a picture every tenth of a second captures those jets every 150 feet or so. Those aren’t very good odds to get jets right next to each other, are they? A similar scenario gets played out trying to capture the touchdown catch. So what to do? How about relying on your own reflexes? You can be quicker than you might think. Something I’ve noticed is that most photographers will close their left eye while looking through the viewfinder with their right eye. Stop that! Train yourself to observe what’s going on with your left eye. You need to be able to anticipate peak action, and you can’t do that if you can’t see it. Next, you need to learn to compensate for the slight delay when you squeeze the shutter release before your camera takes the shot, called “shutter release lag”. This lag is usually about 40 or 50 milliseconds, unless you have one of those point-and-shoots that can take eons to respond. There is no substitute for practice. No matter how much automation your camera has, it will never be able to replace your human intelligence or your anticipation of action. Learn your camera and lens; know which way to twist that zoom ring. Make it become second nature to you to zoom out until you locate your subject, then zoom in to properly frame the shot. I'm not claiming that pro camera features have no value. I'm just saying that you can't necessarily buy your way to getting that great shot. If this was all trivial, then where’s the fun and challenge in that? Go after that satisfaction of owning the shot that didn’t get away! Jet Smooch #howto

This article analyzes the anti-vibration (OS) algorithms available using firmware version 1.02 for the Sigma 150-600 mm Contemporary lens. Sigma hasn’t published any information on the relative effectiveness of the three available OS algorithms, so I took it upon myself to see if there is any difference. The three available OS algorithms are called “Dynamic View Mode”, “Standard”, and “Moderate View Mode”. The default setting, if you don’t program any customization, is “Standard”. Each of these modes is available regardless of selecting “OS1” or “OS2”. The “OS1” is the normal hand-held mode, while the “OS2” mode is used for horizontal panning, such as while mounted on a tripod. All tests reported here are using “OS1”. Sigma "OS" is the same as Nikon "VR". You must use Sigma’s “USB Dock” with their Optimization Pro software to program any customization into the lens. The dock also allows programming focus fine-tune adjustments (16 settings for 4 focal lengths and 4 distances) and focus limiter modifications. To perform the tests, I used a tripod to rest my fist on, and then I rested the lens on my fist. This arrangement afforded me some level of aiming control, while still letting the lens “wiggle around” to simulate hand-held. I shot about 10 frames of my “A0” size resolution chart at 55 feet for each lens switch setting. The resolution chart images were analyzed using the MTF Mapper program from Frans van den Burgh (see this link). The resolution measurements require that the chart images be pretty level and perpendicular to the lens axis, which is why I didn’t simply try hand-holding the lens while shooting the chart. I shot each resolution chart image at 400 mm and f/6.3, and I used back-button focus with “AF-C” continuous auto-focus. These are typical shooting conditions for me, which is why I chose them for the testing. Beyond 400 mm, accurate aiming just gets too difficult for reliable/repeatable testing measurements. My lens was programmed with the “C1” switch setting having “Fast AF Priority” focus speed and “Dynamic View Mode” for the OS setting. I have already made tests that show this “fast” focus mode is essentially as accurate as the “Standard” (default) focus mode, at least when using firmware 1.02. The “C2” switch was set up with “Standard” AF focus speed and “Moderate View Mode” for the OS setting. If the customization switch is turned off, then you get “Standard” AF focus speed and “Standard” OS as well. Tests such as these are 'statistical' in nature, since they involve taking measurements with a lens waving around. I have included some data below, to give you an idea of how the measurements vary. Yes, I could have made 1,000 measurements at each setting to raise the confidence level; I leave that as an exercise to the reader. High Shutter Speed Tests My previous testing has shown an insignificant difference in resolution when you leave OS active at higher shutter speeds (1/1000 and above). This statement is not valid for all lenses!! Any OS algorithm is equally effective (or ineffective, if you wish) at high shutter speeds (you need firmware 1.01 or newer to get this result, however). I only saw a decrease of about 1.0 lp/mm MTF50 by leaving OS active above 1/500 shutter speed. OS Setting Screen C1 Switch Settings That I’m Using Now Medium Shutter Speed Tests The following tests were conducted using a shutter speed of 1/250 second. For 400mm using an APS-C sensor (600mm equivalent), I consider this “medium”, and starting to get into the realm of needing anti-vibration. Some people would benefit with stabilization at this speed, and some wouldn’t. Dynamic View OS, High-speed AF MTF50 Measurements: 36, 40, 38, 30, 36, 40, 42, 42, 40, 36, 40. Average = 38.2 lp/mm Moderate View OS, Normal (Standard) AF MTF50 Measurements: 42,40,42,40,44,40,45,38,42,42,42,42. Average = 41.5 lp/mm Standard (default) OS, Normal (Standard) AF MTF50 Measurements: 40,45,42,38,42,42,45,42. Average 42.0 lp/mm OS Off, Normal (Standard) AF MTF50 Measurements: 44,42,40,42,40,34,38,42,38. Average 40.0 lp/mm Results here don’t show much difference with OS active or not. The “Standard” OS algorithm got the best results, but not enough to really matter. Low Shutter Speed Tests These tests used a shutter speed of 1/60, or a little more than 3 stops beyond the traditional limit of 1/600 for an equivalent of 600 mm (DX frame). This is roughly the rated effectiveness of OS for this lens. Dynamic View OS, High-speed AF MTF50 Measurements: 23,26,34,30,26,30,22,28,28,28. Average 27.5 lp/mm Moderate View OS, Normal (Standard) AF MTF50 Measurements: 30,24,28,30,30,28,30,30,30,28. Average 28.8 lp/mm Standard (default) OS, Normal (Standard) AF MTF50 Measurements: 24,23,24,26,32,22,24,18,28,24. Average 24.2 lp/mm Results here show that the “Moderate View” is the winner. I didn’t show the “OS Off” here, because the images were mostly blurred beyond recognition. Bear in mind that the MTF Mapper software is extremely picky, so the numbers here may lead you to believe that OS is not that helpful. Not true. The pictures are enormously helped by the OS system when you shoot at slower speeds, but there’s no substitute for high shutter speeds. Conclusion It appears that “Moderate View” wins, although not by a huge margin. Sigma (rather cryptically) describes the effect of how each OS algorithm “looks” through the viewfinder. To me, what counts more is which algorithm provides the best anti-vibration effect in the final picture. It seems to me that there is a different end result in your pictures, depending upon which OS algorithm you pick. I prefer the “look” of Nikkor VR to Sigma OS when looking through the viewfinder, but both companies seem to provide roughly equivalent results in the final shot. Newer lenses invariably provide better stabilization, though. Sigma has the advantage of future OS algorithm improvements, however, available through a new firmware update. I saw a definite improvement in the Sigma auto-focus system (speed and accuracy) after loading the firmware versions 1.01 and 1.02. I also saw an improvement in the ability to not “mess up” the shot when forgetting to turn off anti-vibration at high shutter speeds. Don’t be surprised if Sigma has more tricks up their firmware sleeves in the future. Recently, Tamron finally saw the light, and is now copying Sigma with the ability to reprogram lens firmware (in their new 150-600 mm offering), with an essentially identical set of features as Sigma. Nikon et al. hasn’t yet seen the light. #review

Sigma sells a USB dock that lets you update and customize their lens firmware. The (free) program used with the dock is called Sigma Optimization Pro. I have an article on it here: . Sigma has been providing firmware updates for my 150-600mm Contemporary (and also for the Sports version). The first update (1.01) improves the auto-focus speed (they claim up to 50%). The second update (1.02) fixes focus issues with the Nikon D500 used with a teleconverter. I bought the USB dock to enable in-lens focus fine-tune. This style of focus fine-tune goes way beyond any other lens manufacturers; it lets you fine-tune at 4 focal lengths and 4 distances per focal length, giving you a total of 16 fine-tune settings. This feature totally transformed my lens resolution from mediocre to stellar. Nikon’s (and Canon’s) focus calibration only lets you perform a simple global focus shift; this just doesn’t cut it for focus calibration. It is handy, though, when you mount the Sigma on another camera body, where the camera’s focus fine tune gets applied in addition to the Sigma in-lens focus calibration. My focus calibration fine-tune settings Auto-focus Firmware Changes I tried their auto-focus customization options (fast “Fast AF Priority”, medium “Standard AF”, or precise (slow)) when I first got the lens, but found the ‘fast’ algorithm wasn’t very accurate. I settled on the default “Standard AF” auto-focus speed, since I’m not willing to sacrifice resolution for speed. I didn’t notice any precision improvement trying their “precise” setting. It took me a long time, but I eventually got around to testing the new focus algorithms that were provided with the 1.01 version of the software. I noticed when I first loaded the new firmware that the default speed (Standard AF) was more responsive than it used to be, and have been happy shooting with that setting. It never occurred to me to re-try the “fast” auto-focus setting; big mistake. It’s great. Accuracy is now essentially the same as the medium “Standard AF” setting, and it is simply faster. It’s like I just got a new lens, but for free. C1 switch settings: focus speed, focus limits, viewfinder stabilization ‘effect’ C2 switch settings: focus speed, focus limits, viewfinder stabilization ‘effect’ Accuracy Comparison: High-Speed Auto-focus vs. Standard Auto-focus I used the MTF Mapper software to evaluate auto-focus accuracy. This test is a bit ‘statistical’ in nature, because it’s based upon a moving target. I lean on a tripod while hand-holding the lens at 600mm. This technique lets me get more reliable framing of my resolution target, but the lens is still “wiggling” quite a bit. I use back-button auto-focus and “AF-C” continuous focus. I also re-focus by pointing away from and then back onto the resolution target while AF is active. Note that the following MTF results aren’t as good as a firmly-mounted lens on a tripod and remote shutter release. With a long lens, even 1/2000 will have a tiny amount of motion blur when hand-holding. 1) MTF50 maximum results with “Standard” AF Speed, 600mm 1/2000s f/6.3 hand-held: 28, 28, 26, 28, 28, 26, 30, 32, 26, 24, 24, 30, 28, 28, 26, 26, 26, 28, 30 Average MTF50 maximum: 27.47 lp/mm 2) MTF50 maximum results with “Fast AF Priority” Speed, 600mm f/6.3 hand-held: 28, 30, 23, 24, 28, 22, 28, 26, 32 Average MTF50 maximum: 26.78 lp/mm Conclusion: There is no real focus accuracy difference using the “fast” auto-focus algorithm versus the “medium” auto-focus algorithm using the 1.01 or 1.02 firmware. So why wouldn’t you just leave it on “fast”? Sigma didn’t advertise that their accuracy got better with the new firmware, but I’m seeing a definite improvement in both speed and accuracy. Vibration Reduction “Optical Stabilization” Firmware Changes? The good news doesn’t end there. When I first got the lens, I experimented with using their vibration reduction (they call it “optical stabilization” or OS). They provide the usual ‘OS1’ for general hand-held use, and ‘OS2’ for panning use. I tested the lens using OS1 and shutter speeds beyond the normal VR upper-limit of 1/500. I found that the resolution was reduced when I tried 1/1000 by about 9%. As a result, I would turn off VR (OS) at high shutter speeds, just as I was taught to do for all lenses with VR. Using the new firmware (1.02) I’m not noticing any measurable degradation in resolution at high shutter speeds (all the way up to 1/8000)! This is just fantastic. I’ve always hated having to remember to turn VR on and off to accommodate my shutter speed changes. Now, I can just leave VR on and forget it. I’m going to have to re-test my other lenses to see if they really require me to turn VR off with higher shutter speeds or not. The moral of the story is don’t blindly believe the urban legend about always turning VR off at high shutter speeds. Test it first! VR Testing at High Shutter Speed Sample These tests were performed at 600mm f/6.3 using the “fast” auto-focus setting, hand-held, AF-C “back-button” focus. Shutter speed was 1/2000 throughout. Sigma lens Firmware version 1.02. Again, these MTF50 numbers are lower than when using a tripod with a remote shutter release; even high shutter speeds with a big lens aren’t as effective as a tripod for static subjects. OS1 Active MTF50 maximum: 28, 30, 23, 24, 28, 22, 28, 26, 32. MTF50 Average: 26.78 lp/mm OS1 OFF MTF50 maximum: 28, 26, 28, 28, 30, 22, 26, 24, 22. MTF50 Average: 26.0 lp/mm Conclusion: There is essentially no difference with stabilization active or not at this high shutter speed. I tried tests such as these all the way to 1/8000 shutter, without significant changes to MTF50 resolution when leaving vibration reduction active. Isn’t this great that Sigma comes out with these firmware updates? If only Nikon and Canon could catch up to these guys in making smarter lenses. #review

This article will show you how to use the mtf_mapper_gui.exe program to measure axial (longitudinal) chromatic aberrations. Longitudinal chromatic aberration (LoCA) is the optical problem of focusing different colors of light at different distances along the optical axis. My other article about MTF Mapper is located here . The other article is about MTF50 resolution measurement and focus calibration. Be aware that you need to use separate charts to make separate measurements (lens MTF50 resolution, focus calibration, or LoCA analysis). Also be aware that measurements depend upon the color of light used while photographing the targets (I used outdoor lighting with a clear sky in the sun). LoCA differs from lateral (transverse) chromatic aberration, which instead spreads out different colors perpendicular to the lens optical axis. Lateral chromatic aberration typically shows up as purple corners in the photograph; you won’t see it in the center of the image. Lateral chromatic aberration is simple to fix; most modern cameras can even automatically fix it in-camera. Longitudinal chromatic aberration is more difficult to correct, although programs such as Nikon Capture NX2 can largely mask its effects. Axial (longitudinal) aberration diagram, courtesy of Wikipedia.org LoCA can rob a lens of resolution, since some colors will be in-focus and some colors will be out of focus. You can recover the resolution by stopping the lens down until all visible wavelengths are in focus, but artistically this is often a poor option. A common visual effect of LoCA is seeing magenta instead of white in specular reflections on the eye. If you’re interested in evaluating how much LoCA a lens has, then the MTF Mapper program can help you measure it, and the program is free at the time of this writing. The MTF Mapper program author is Frans van den Bergh. His software and printable test charts are available here. Frans comes across as a ‘brainiac’, and his documents can take your breath away. They’re worth a read, though, even if you can’t grok 100% of their contents. More of his writings about image analysis topics can be found here. If you like his stuff as much as I do, please let him know! I’ll bet Frans has spent a gazillion hours working on this software, and he deserves all the praise we can give him. I’m using version 0.5.7 of mtf_mapper_gui.exe for these tests. You need to print the proper chart and use the correct program preferences to get the desired measurements. The chart I used is called “mfperspective_a3.pdf” (printed on A3 paper), which is also available at the same site that you download the program. You’re supposed to rotate the chart 45 degrees about the vertical, with the taller vertical targets farther from the camera than the shorter ones. Mfperspective_a3.pdf chart photo. Chart rotated 45 degrees. The MTF Mapper program takes advantage of the fact that camera sensors are separated into red, green, and blue-sensitive pixels. The program can independently analyze single-color pixels or all of them together. By the way, half of all of the camera sensor pixels are sensitive to green, while 25% are sensitive to red and the last 25% are sensitive to blue. It turns out that human eyes are most sensitive to green, so having 50% of the pixels being sensitive to green makes pictures look ‘correct’. This kind of sensor design is called “Bayer”, named after the inventor Bryce Bayer who used to work for Kodak. Different combinations of the R,G,B pixels can recreate all of the colors we see. MTF Mapper uses a program called “dcraw” (included in download) that knows most camera “raw” formats, and is regularly updated for new camera models. Axial R,G,B Focus Measurement In the following test, I’m using the Sigma 150-600mm zoom. Big lenses typically have the most trouble with axial chromatic aberration, and they also tend to exaggerate any focus errors. Take a photo of the chart, aligning the camera focus sensor on the middle of the chart. Don’t worry if the focus isn’t perfect; what counts is the difference between the red, green, and blue color channels in the same photo. A perfect lens would focus all three colors at exactly the same distance. The largest aberration errors will be seen with the lens aperture wide open. You’ll likely get different measurement results as you change focal lengths, too. Set your camera for RAW mode, so that there are no in-camera compensations for sharpening or chromatic aberrations. Remember to rotate the chart so that it’s at about 45 degrees from the sensor, with the tall side of the chart target images farther away than the short images. MTF Mapper Settings Preferences to measure axial aberrations After getting a photo of the test chart, the program needs to be configured for the color to be analyzed and for the camera pixel dimensions. The Nikon D610 has pixels of 5.95 microns, but the D7100 has 3.92 micron pixels, for instance. The program needs to be run against the same picture three times, each time selecting a different Bayer color channel. Blue Bayer Channel focus results show focus is 14mm in front of the chart center. Chart close-up (green). The orange arrows are at the chart center. Green Bayer Channel shows focus is 5mm beyond the chart center. Red Bayer Channel shows focus is 4mm beyond the chart center. The results above indicate that the red and green colors are focused at almost exactly the same distance, but the blue channel shows focus is nearly 19mm closer than the green and red channels (at a focus distance of about 6 meters). The test results also show the MTF50 resolution, measured in cycles per pixel. The green channel is sharpest at .211 c/p (peak sharpness). The red and blue channels both measure about 0.197 c/p. This MTF50 measurement is only valid for the peak focus location; you should be analyzing resolution using the resolution chart (see the other MTF Mapper Cliffs Notes article). Evaluating the MTF50 Math MTF50 lp/mm = (c/p)* pixels tall / height_mm Line pairs/pixel height (lp/ph) = MTF50 lp/mm * sensor height mm. D7100: 3.92 micron pixels 4000 X 6000 pixel sensor (24 MP) 15.6mm X 23.5 mm sensor Green MTF50 = 0.211 c/p = .211*4000/15.6 = 54 lp/mm, or 844 lp/ph Blue MTF50 = 0.197 c/p = .197*4000/15.6 = 50.5 lp/mm, or 788 lp/ph Red MTF50 = 0.197 c/p = .197*4000/15.6 = 50.5 lp/mm, or 788 lp/ph Conclusion The MTF Mapper program is a great way to evaluate axial chromatic aberrations. This optical aberration is typically a bit mysterious to get a handle on, but now there’s an easy way to characterize it. When you can analyze by the numbers, then you can actually compare it against other lenses in a meaningful way. And you can’t beat the price. #howto

This article is an evaluation of the old manual focus 20mm f/4.0 lens that has been AI-converted. The conversion makes it possible to get automatic exposure on the better Nikon digital cameras, including the D610 with an FX sensor. The 20mm f/4 was manufactured from 1974 through 1978, back when Nikon was the big man on campus in photography. Mine was purchased in 1975, and virtually never came off of my Nikon F2 when I was back-packing (unless I did some macro shots). Small, light, sharp, tough, elegant, and wide; almost exactly like the ideal woman, except perhaps for the ‘wide’ part. This 20mm lens is one of the smallest and lightest FX lenses Nikon ever made. At 7.4 ounces, you hardly notice it’s there. It's about 1.4 inches long, like a thick body cap. Cameras like the D7000 series and D610 allow aperture-priority auto-exposure after defining the “non-CPU lens data” for this lens. I absolutely love its field of view (94 degrees) on the D610. The lack of auto-focus using a lens like this isn’t a hardship, since everything is typically in focus all the time. You still get the 3-stage focus indicator inside the viewfinder while manually focusing with the better Nikons. The 20mm, of course, has the little red dot on the focus scale for infrared focus compensation. Back in the day, Nikon really paid attention to stuff like that. One of my main uses for this lens is infrared, but unfortunately cameras like the D7100 and D610 are essentially useless for infrared (see this article: ). My D7000, however, works perfectly for infrared with this lens (using the Hoya R72 52mm filter). Manual-focus lenses are actually superior to auto-focus lenses for shooting infrared with a filter like the Hoya R72, since you can’t see through the viewfinder. You frame and focus (and use the lens focus scale red-dot IR shift) before attaching the filter. Those of you who have gone through the pain of framing/pre-focusing a ‘G’ auto-focus lens and then mounting an IR filter know what I’m talking about. Focusing is still silky-smooth, unchanged since the day it was manufactured. The focus scale is a thing of beauty. I have every reason to believe that this lens will last not just a lifetime, but multiple lifetimes. Although it works without vignetting, be careful using a polarizer on this lens; the sky will look too un-even because of the wide field of view. 20mm f/4 Nikkor AI-converted on the D610. Sweet. Resolution Tests I test lenses using the MTF Mapper software and the recommended resolution charts (printed to A0 size and dry-mounted). The article here explains the software and its use. As you’ll see, you are going to want to stop down to f/8 or more to get the corners you want. The center is already very good at f/5.6, and only gets better as you stop down. Avoid going beyond f/16, because of diffraction. All tests were done using the D610, with 24 MP (5.95 micron pixels). I only shoot un-sharpened RAW for the resolution tests. To convert the MTF50 lp/mm measurements into LP/PH, simply multiply readings by 24.0. To convert into LW/PH, take the LP/PH values and multiply by 2. Sample Photos Sun just out of frame. Palm fronds near frame edge are sharp, D610. The “wavy” distortion is quite minimal, D610. Infrared, Hoya R72, D7000 Conclusion If you're willing to stop this lens down to f/8 or f/11, the results are about as good as any ultra-wide lens made today. You won't find a more compact lens anywhere. I love that it uses 52mm filters, too. This has become my go-to lens for infrared photography. My D610 and D7100, by the way, are basically useless for IR unless I put the DK-5 eyepiece cap over the viewfinder. My other cameras are fine without the cap, unless I switch to my 850nm IR filter, which requires exposures of 2 or 3 minutes; all cameras need an eyepiece cap when exposures get that long. I understand that this lens is a bit rare these days; I have no intentions of ever selling mine. #review

This review will concentrate on the lens MTF50 resolution performance and how well the lens auto-focuses. Repeating the Nikon specifications of the lens isn’t why you should be reading this, since that data is readily available in any number of other places. This lens is often referred to as the "wedding photographer's lens". On FX, it gets you 95% of the focal lengths you'd need at a typical wedding, in addition to the wide-enough aperture for dim venues and great bokeh. Personally, I'd pair this baby with the Nikkor 85mm f/1.4 and you're good to go. The usual disclaimer: this is looking at a single copy of the lens. Yours will be different, but hopefully ‘similar’. These tests were done using a Nikon D610 (24 MP, 5.95 micron pixel) with un-sharpened 14-bit compressed RAW format. Nikkor 24-70 f/2.8 VR with lens hood. Massive. Here is a link to get pretty good overall information on this lens. They reduce resolution measurement down to a single number for an f/stop setting. It’s not that simple; resolution is a 2-dimensional thing, and there are sagittal and meridional characteristics within those two dimensions. Calibration First things first: focus calibration. My lens copy needs a fine-tune adjustment of about +16 on my camera, which is vaguely alarming for such a high-cost “pro” lens. I say “about +16”, because this lens needs a different focus calibration setting at different focal lengths. I am using the MTF Mapper software to get focus calibration measurements and MTF50 measurements, which I explain at this link You should really be using tools that give you answers “by the numbers” to get the most out of your equipment. The focus target I use is an A0-sized print, oriented 45 degrees to the camera sensor plane. Since the camera focuses on the vertical edge of a giant trapezoid, there’s no ambiguity about where the camera is focused. See the picture of the target below (the trapezoid looks more like a rectangle, due to the perspective distortion). I use “AF-C” with a back-button focus for this test, since that’s exactly the way I shoot the camera. Too bad Nikon focus calibration is so primitive, compared to Sigma. This lens (read ‘most lenses’) desperately needs calibration for different focal lengths and distances. Oh, well. I had to split the difference for the fine-tune value, giving more bias to 70mm (since focus errors are more noticeable there). For a typical factory-calibrated lens, my D610 needs a fine-tune value of +6, so measuring a +16 (or a little higher) doesn’t inspire confidence in Nikon quality control. As long as it stays within +- 20, I suppose I shouldn’t care. In case it’s not obvious, you never use “Live View” for focus calibration tests. Only use “phase detect” focus. I always use “AF-C” and not “AF-S”, since that’s exactly how I always shoot (in addition to back-button focus). Fine-tune +16 not quite enough at 70mm. The peak should align with the blue line. Fine-tune +16 a little too much at 24mm. The blue line is where the focus sensor was placed on the target; the green peak is where actual focus was located. What the focus target looks like at 70mm. It’s rotated 45 degrees to the camera. The left side is farther from the camera than the right side. I taped it on top of my resolution target, which is why you see faint gray squares (it doesn’t affect the calibration). Focus Speed and Repeatability This lens focuses (phase detect) very quickly and was totally repeatable (in decent light), so no complaints there. By the way, you should always conduct focus tests in good light, or else you’re wasting your time with trying to calibrate against a moving target. EV 10 or brighter is a good light level to use while testing. Even in dim lighting, I didn’t experience any focus hunting. Vibration Reduction (VR) This lens has state of the art VR, which is one of the biggest upgrades over the previous version of the 24-70 lens. I got nearly 4 stops improvement. I’m guessing that this is the main reason somebody would opt for this lens versus getting the previous version. ‘E’ Electronic Aperture Maybe you have the D5 and are shooting at 12 fps. Maybe you can tell the difference in the electronic aperture versus the mechanical aperture. I honestly can’t tell the difference in shooting with a mechanical or electronic aperture. Infrared For those of you interested in shooting infrared, this lens produces a minor hotspot in the middle of the field of view. If you stick to f/5.6 or wider, the hotspot virtually disappears. Since I usually shoot at 24mm and f/8 with infrared, I use Lightroom with its "radial filter" to decrease the exposure in the middle of the shot by a little more than a third of a stop to fix the hotspot. I also have to shift focus to about 8 feet to get infinity in focus with my 850nm infrared filter (again at 24mm). Centering Test Centering: out-of-focus rings should look perfectly symmetric The out-of-focus rings look perfectly symmetric about the center, indicating that the optics are well centered. This also gives you some idea about the bokeh, which I consider to be excellent. Resolution Here’s the meat of this review. Although the data below is in terms of “MTF50 resolution”, keep in mind that camera sensors with larger pixels will actually produce worse results in terms of “MTF50 lp/mm” values for a lens. You need to combine the MTF50 values with the size of the camera sensor to arrive at the overall picture resolution, typically expressed in “line pairs per picture height” (LP/PH). It’s simple to convert, however: LP/PH = MTF50 * SensorHeight_mm. The D610 sensor height is 24mm, so lp/ph is just (MTF50 * 24.0). For instance, a lens that has an MTF50 lp/mm of 40 gets (40*24) or 960 lp/ph. Like I said, simple. I use an “A0”-sized resolution target, where all of the small squares have a little slant to them. The MTF Mapper software I use gets answers that are very similar to Imatest. The main difference is that MTF Mapper is designed to provide answers separated into “meridional” and “sagittal” directions, just like the MTF line charts that manufacturers distribute. As a side note, however, manufacturers only give you "theoretical" and not actual MTF curves. Since the software can produce full 2-dimensional results, you actually see how the whole sensor performs. I imagine you expect your camera to produce 2-dimensional photographic results, so why wouldn’t you want measurements done the same way? What the resolution target looks like. Mine is mounted ‘upside down’. Very good center results wide open at 24mm. Corners aren’t too bad, either. Note that the meridional and sagittal measurements aren’t that different, so that means that astigmatism is well-controlled, too. Sagittal-direction corners at this aperture are among the best I have seen; meridional-direction corners aren’t so hot. Center MTF50 lp/mm is 45, so this means 1080 lp/ph. Note poor meridional (tangential) and good sagittal corner resolution at f/2.8 Resolution Summary Stop down to get the sharpest corners; otherwise, just set the aperture to get the desired depth of field. Center performance is stellar, unless you go beyond f/11. It’s slightly less sharp at 70mm, but not enough to really notice. I have read complaints about this lens for sharpness, but my copy was able to get 1200 lp/ph (MTF50 = 50 lp/mm) in the center at just about every focal length. I got at least 720 lp/ph (MTF50 = 30 lp/mm) in the corners at all focal lengths for meridional (tangential), except around 35mm, where it topped out at 600 lp/ph. For sagittal direction, the corners were nearly as sharp as the center, which is just brilliant. The real key to getting sharp pictures is focus calibration. I’m astonished at how many people ignore this fact. It’s a pain to do the calibration, but you’re rewarded every time you take a shot after that. Don’t get lazy. I’ll continue to harp at Nikon about being so primitive with their focus fine-tune abilities. So is just about everybody else, except Sigma. Mirror-less manufacturers claim their on-sensor phase-detect solves this, but the focus speed just isn’t there (yet). Sample Pictures #review

I happened to be testing an old Nikkor 20mm f/4 (AI-converted). I thought I’d try some infrared shots, since this lens is supposed to be excellent shooting IR. I use the Hoya R72 IR filter, with the 52mm thread diameter. This 20mm lens is about the smallest and lightest FX lens Nikon ever made. 7.4 ounces light. The D7000, D7100, and D610 allow aperture-priority auto-exposure after defining the “non-CPU lens data” for this lens. I absolutely love its field of view (94 degrees) on the D610. You’d never miss auto-focus using a lens like this, since everything is typically in focus all the time. You still get the 3-stage focus indicator inside the viewfinder while manually focusing. Still, 20mm on a DX camera isn’t that wide; my Tokina 11-16mm f/2.8 has no reason to feel threatened here. The 20mm, of course, has the little red dot on the focus scale for infrared focus compensation. Back in the day, Nikon really paid attention to stuff like that. Manual-focus lenses are actually superior to auto-focus lenses for shooting infrared with a filter like the Hoya R72, since you can’t see through the viewfinder. You frame and focus (and use the lens focus scale red-dot IR shift) before attaching the filter. Those of you who have gone through the pain of framing/pre-focusing a ‘G’ auto-focus lens and then mounting an IR filter know what I’m talking about. After all of those digressions, back to the subject at hand: IR shooting comparisons. The D7000 IR results indeed look excellent. Absolutely nothing to complain about here, aside from the gripe about the narrower DX field of view. I was shooting at f/11.0, 15 seconds, ISO 250. (“Sunny 16” rule would have been 1/500 at f/11, ISO 250. IR needed 12 stops more light!) Nikon D7000 using Hoya R72 with Nikkor 20mm f/4 AI-converted lens. Excellent Now, for the D610 infrared results. How to describe what I got? Epic failure comes to mind. Totally unusable. It appears that the light baffling and anti-reflection coatings inside the D610 act more like a mirror in the infrared spectrum. In comparison, this 20mm lens is wonderful for regular-light photography on the D610, especially for landscapes. Nikon D610, Hoya R72, Nikkor 20mm f/4 AI-converted lens. Gross. Next, I head for my D7100. Terrible. Exact same light baffling and anti-reflection coating problem in infrared. Nikon D7100, Hoya R72, Nikkor 20mm f/4 AI-converted lens. Gag me. Note the terrible horizontal glare across the entire frame for both the D610 and D7100. The Nikon D7100 misbehaves in a nearly identical way to the D610 when shooting infrared. My guess is that the camera internal baffling and anti-reflection coatings actually reflect instead of absorb infrared wavelengths. Ahh. I bet it's something wrong with the 20mm lens, you say. I bet the problem goes away with a different lens, you say. There's no way the D610 and D7100 could let me down this badly, you say. How could the D7000 possibly be superior to the D610 and D7100 in any way, you say. I tried using the 50mm f/1.8 AF-D with the Hoya R72 IR filter on the D610, since I’m apparently a glutton for punishment. Big nasty hot spot in the center of the picture, in addition to the terrible horizontal banding flare. Having used this lens in the past for infrared, I know it’s not the lens’ fault. I have to conclude that the D610 is useless for infrared photography. I didn’t have the heart to try this same lens on the D7100; I know the results would be the same. Now, the secret sauce to making the D7100 and D610 succeed with IR photography: you absolutely need to cover the viewfinder eyepiece with the little "DK-5" eyepiece blocker. Unlike the D7000, D60, D50, and D500 cameras I have tested, the light baffing in the D7100 and D610 seems to be inferior. You can clip the DK-5 onto your camera strap so you don't lose it, and you don't even need to take it off of the strap to slip it over your viewfinder! 50mm f/1.8 AF-D on the D610, HoyaR72 filter. Still gross and unacceptable. The Nikon D7000, as with all modern digital cameras, (that haven’t been converted to infrared) is very insensitive to infrared. The filter on top of the image sensor screens out almost all of the infrared wavelengths. Older camera sensor filters (like the D50 and D60) were much better at passing IR (a few stops better, at least). Aside from long exposure times, the D7000 provides top-notch IR results. Just keep in mind that the Bayer sensor only has a quarter of the photo sites sensitive to red, so your camera resolution is essentially divided by 4 as well. Nikkor 20mm f/4.0 AI-converted, on a D610. It's only a little bigger than a body cap. Moral of the story: don’t ditch that D7000 if you do infrared photography. For the D610 and D7100 (and probably the d7200), always use the little DK-5 viewfinder eyepiece blocker. By the way, I typically use Nikon Capture NX2 to convert my 'deep red' RAW shots into the samples you see above. I can't get the D7000 to succeed at measuring a scene to get "preset manual" white balance with infrared, although the preset measurement works for a D50 and D60. I use the Capture NX2 "Camera Settings", "White Balance", "Set Gray Point", "Marquee Sample", "Start", then rectangle-mouse-select the whole picture, then click inside the selection. This is a quick and easy way to get the picture really close to what the in-camera "preset manual" white balance procedure achieves for regular photography. Once the editing steps are entered, just save those steps as a batch process. The batch process can be run on a whole folder of IR shots, to quickly get everything converted. #review

You may have already read my article about using manual mode with your external flash, which allows you to shoot in “manual” mode but get automatic exposure via the flash. That article can be found here. That particular mode of operation is for when you have a fixed ISO setting, which is the normal case while using "manual" mode. What about manual mode without flash? It turns out that you can still get ‘manual mode’ to provide automatic exposure. You make this happen by selecting “Auto ISO sensitivity ON” in the “Shooting” menu (on Nikons, of course). You can see this mode at work in Manual by adjusting the shutter speed or aperture and observing the ISO indicator value changing to keep up with your selections. When you attempt to set your exposure to go beyond your maximum ISO limit you have set, you’ll see the exposure indicator show "under exposure" (and by how much) since it won’t go beyond your specified maximum ISO limit. Why on earth would you want to automate Manual Mode? If you’re using a long lens, it’s handy to be able to set a specific aperture and also force a (specific) high shutter speed. If you hate to give up automatic exposure just because you want to specify both an aperture and shutter at the same time, this is the secret sauce that lets you have your cake and eat it, too. I guess I must be getting hungry; that’s too many food references to ever put into a paragraph, let alone a single sentence. If you’re worried about getting quality results (and you should be) then you may not want to use this technique unless you have a camera with a sensor that can handle pretty high ISO values without getting excessive noise. I have had far more shots ruined by motion blur than I have lost due to image noise. If you have a long lens (e.g. 500mm or beyond) then there’s basically no such thing as too fast of a shutter speed to capture fast-moving subjects. To calibrate your “Auto ISO”, do some testing up front to determine what’s a reasonable upper-limit ISO. Two key points to consider here are color noise and loss of dynamic range. I always assume that I’m going to be doing some post-processing of my (RAW) shots anyway, so a small amount of color noise in the un-processed shot is completely acceptable. The upper limit of color noise should be where you can no longer clean it up in your image editor without excessive loss of resolution. You should consider noise reduction that only operates on the colors and not the ‘luminance’ channel; you will retain resolution but the image will have a slightly sandy-grained look. A camera like the D610 is still coasting with values like ISO 4000, but a camera like my D7000 is already a little short of breath at ISO 1600. Dynamic range goes out the window with high ISO; don’t use a technique like the one presented here for something like landscape shots, unless you’re forcing a particularly slow shutter speed. Some camera models offer “Minimum shutter speed = Auto”, where the camera will select the “1/focal length” in the “ISO sensitivity settings” menu (with additional adjustment options to get values like 1/(focal length * 2) ). I find that this type of shutter speed control is fine for static subjects, but is often unsuitable for fast subjects like flying birds or action sports. I’d rather crank up the shutter speed to cryogenic levels and leave it at subject-freezing speeds. This technique is, of course, not appropriate in all situations. As always, pick the right tool for the right job. #howto

It’s well-known that the Tokina 11-16mm f/2.8 DX will work on an FX camera. What isn’t well-known is the corner resolution performance on FX. This discussion is valid for both the original “Pro DX” and the “Pro DX II” AF-S versions, which have the same optical formula. See my original review of this lens (on DX, of course) here. Did you know that you get nearly zero vignetting at 16mm on FX, or that this is actually a wider angle of view (16mm FX) than you can get on your DX camera at 11mm? Your DX camera only sees the equivalent of 16.5mm on an FX camera while zoomed to 11mm on DX (a Nikon DX camera, that is). The Nikon DX has a 1.5X crop factor, so (1.5 * 11) = 16.5. Many people have said that the Tokina is usable in the 15mm-16mm range, but that’s not what my testing shows. 15mm starts to show vignetting, so I personally wouldn’t use the 15mm focal length. I just treat the Tokina as a “16mm prime” when mounted on an FX body. This lens isn’t sensitive to the focus distance causing a change in the amount of vignetting, so you don’t have to worry about that. I’m even able to use a “thin” UV filter and the lens hood while at 16mm on FX. Tokina 11-16mm f/2.8 DX lens mounted on the Nikon D610 FX camera First, let’s use an A0-size resolution target to evaluate resolution. I use the MTF Mapper software to evaluate resolution, and this is a picture of the resolution target designed for it. Tokina un-corrected resolution taget at 16mm f/2.8 shows minimal vignetting. Tokina corrected resolution taget at 16mm f/2.8 shows minimal distortion. Tokina at 16mm f/2.8 center resolution is already good, but corners are well behind. f/4.0 center is excellent, but corners still aren’t much improved f/5.6 shows pro-level center performance. Corners are still a bit weak. \ f/8.0 This is the best overall performance aperture setting f/11.0 Still very good overall performance. Some diffraction is setting in. Tokina at 16mm, focused at infinity. Corners have no vignetting problems Tokina at 16mm. Excellent edge-to-edge. Distortion is minimal. Conclusion Honestly, you wouldn’t know that Tokina didn’t make this lens for FX, as long as you park it on 16mm. Vignetting, resolution, and distortion are all well-controlled. I can say without any hesitation that this lens is fantastic for use on both DX and FX cameras. #review