I was recently lent a Nikon Coolpix B500, which has a 22.5 mm to 900 mm zoom (35 mm format-equivalent focal length). The lens is actually 4 mm f/3.0 to 160 mm f/6.5 and has a macro mode, as well. I haven’t ever seen a detailed resolution analysis of a super-zoom like this, so I thought I’d take up the job myself. I won’t dwell on the camera features too much, but I can’t help myself from at least making a few observations about the camera body as well.
Camera Body Highlights
The camera itself is purely amateur; it doesn’t support raw-format (just jpeg) or even manual exposure control. Its 16-megapixel sensor is breathtakingly small: 4.62 mm X 6.16 mm, with 1.34-micron pixels. The pixel count is 4612 X 3468. A modern smart phone sensor actually has bigger pixels than this (typically 1.4 micron). But the camera costs around $300.00 which is cheaper than those smart phones.
I’m a big fan of lenses that provide great out-of-focus backgrounds (bokeh). Because of the tiny sensor in this camera, depth of focus is huge until you get to long focal lengths. If you can manage to get the background out of focus, the results are actually quite pleasing.
For those of you that are interested, it has both Wi-fi and Bluetooth, and supports SnapBridge. You can do the usual remote control from your smart phone, if you wish. It uses 4 AA batteries (I used rechargeables).
The camera, incredibly, is capable of shooting 7.7 frames per second in “sports” mode at full resolution. It can only shoot 7 frames at this rate, however, so its buffer is tiny. Given the lack of a viewfinder, however, you’re pretty much out of luck tracking any action if you zoom in to any significant degree.
You can shoot HD video, of course. I’m personally not very interested in video, so I won’t discuss it any further.
The ISO range goes from 80 to 3200, but image quality would hover near zero at ISO 3200. The shutter range is from 1 second to 1/5000 second (faster than my D610!). This camera is very, very poor at focusing in dim light and low contrast; it gets ridiculously bad if you get anywhere maximum zoom. This is probably my biggest gripe about the camera.
The camera grip is really, really nice; it’s deep and has a very tactile rubberized surface. The camera is very light; a bit too light for my taste. I’m used to the weight of ‘real’ cameras, like the D500 and D610, and I even use battery grips with those. I prefer the inertia and balance that DX and FX camera bodies provide (but I might change my tune at the end of a ten-mile hike).
The camera’s lens cannot be manually focused or zoomed, either; you focus with the traditional half-press of the shutter button, and you zoom either by rotating the shutter collar or pressing a lever on the side of the lens. There is no viewfinder, either; only a 3-inch, 921k LCD screen. Good luck finding and tracking a subject in the sunshine. Its big brother, the B700, costs about 50% more; it sports a viewfinder and the usual PSAM controls, plus a 60 X zoom.
For about $90.00 you can get a Hoodman Loupe to cover and view the LCD screen in sun (and you can use it on your other camera screens in Live View). Not the most convenient solution, but it works.
B500 at 160 mm zoom
B500 top view. Really deep grip.
B500 articulating 921k 3-inch LCD. No viewfinder.
The lens has a minimum focus distance of 12 inches at 4 mm and a minimum focus of about 11 feet at 160 mm. In macro mode (only at 4 mm) it will focus down to about 0.4 inches! The lens has vibration-reduction (which can be turned off for tripod use), and it works amazingly well.
There are no filter threads, and there is no lens hood, either. You’ll need to shade the lens with your hand.
It’s theoretically easier to design a lens that only has a small image circle, and this camera sensor only needs a really small lens image circle to cover it. I think you’ll agree that this design theory is in fact borne out with this camera/lens combination when you see the finished results.
The focus speed is pretty lazy. The zoom is painfully slow, mushy, and approximate. Once the focus and zoom gets there, the shot usually turns out just fine. I realize that I have been spoiled with high-performance cameras and lenses, but my patience was sorely tested at longer focal lengths and in anything other than bright sunlight. I just have to keep telling myself how inexpensive this rig is.
A quarter with the macro setting (4 mm)
They weren’t kidding about the 0.4 inches lens-to-subject distance in macro mode. Lighting at this distance is truly a nightmare. I had to direct a light beam in at a ridiculously steep angle for the above shot. You can get as snarky as you want about the lighting here; I just wanted to see how close I could get.
I’d have to call macro mode largely a hoax. If you can’t illuminate it, you can’t photograph it. And it’s not really magnified that much, either. A bug with any sense of self-preservation would be long gone.
Flare resistance is pretty good. 5 mm f/6.4
Flare and Chromatic Aberration
Take a look at this sample photo. The lens showed remarkable resistance to flare, even though I pointed the lens right into the sun. Impressive.
160 mm (900 mm FX) f/6.5 chromatic aberration
Since I can’t shoot raw format, I can’t tell you if the level of color fringing is due to a great lens design or perhaps in-camera processing. There’s a little purple fringing in the bottom left corner, but no too bad. Shots like this really emphasize any lateral chromatic aberration.
People really obsess about the 900 mm FX-equivalent maximum zoom, but I was more impressed with the 4 mm (22.5 mm FX-equivalent) end of the zoom. This lens goes really wide (77.3 degrees horizontal), compared to typical kit zooms that only zoom to 27 mm (67.4 degrees horizontal) FX-equivalent. The more expensive B700 lens doesn’t go as wide at this lens does (73.7 degrees horizontal); I’d much rather have this wider angle ability than a longer zoom. Throughout the entire zoom range, there is essentially zero distortion! The pictures below will demonstrate this. The main weakness of this lens is meridional-direction resolution, which approaches what I’d describe as shameful at longer focal lengths; it’s actually not too bad at the wide end.
That 40 X Zoom
Wide 4 mm. You can barely see the buildings
Telephoto 160 mm
I added an arrow in the 4 mm shot to show where I zoomed in. There was a fair amount of atmospheric haze, so don’t mistake that for a flare or lens contrast problem. Now that’s a zoom.
The following resolution analysis was done using jpeg with default sharpening. If it was available, I would have shot in raw and performed the analysis without any sharpening. I use the (free) MTFmapper program, which I describe here.
I have a 41” X 60” resolution chart to analyze the lens, except at 160 mm. I would have had to be nearly 200 feet away from the chart at that 900 mm-equivalent focal length to photograph the whole thing; instead I used a small chart that’s only 7” X 10”, at about 35 feet.
41” X 60” resolution chart, 4 mm (22.5 mm FX equivalent)
Note in the shot of the resolution chart that there isn’t any observable distortion, judging by the chart edges. I didn’t note distortion (or perceptible vignetting) at any focal length.
Please be aware in the following measurements that my usual results, measured in “MTF50 line pairs per millimeter” are highly misleading. I’m accustomed to seeing a good “MTF50 lp/mm” lens measurement peak at 40 to 50; seeing measurements around 300 seems astonishing. But here’s the deal: the camera sensor has a lot fewer millimeters in it, so it’s much less impressive than it sounds at first blush. A better unit of measurement for resolution in this case is “line pairs per picture height”, which gives you the total available resolution in the picture.
I have decided to provide 4 different resolution measurement units, so that you can take your pick: “cycles per pixel”, “MTF50 lp/mm”, “line pairs per picture height”, and “lines per picture height”. Besides the 2-D plots across the camera sensor, I looked at the low-level data to provide the peak center and corner resolution measurements.
Another little fact to note: the “MTF10/30” contrast plots I have included are based on jpeg with in-camera processing. These plots should really be based upon un-sharpened “raw” files to be directly comparable to other lenses. Since I can’t shoot raw with this camera, I wasn’t given a choice. The contrast numbers look too good to be true, and they are. The plots are at least useful to compare center-to-edge differences and meridional-versus-sagittal performance, too.
In the MTF50 resolution plots that follow, “red” is good and “blue” is bad. Unlike most other web sites, these plots show nearly 100% of the camera sensor’s field of view. This lets you evaluate lens resolution across the whole field of view instead of just a slice or a single point. Recall that this camera sensor is about 4.6 mm X 6 mm.
The resolution measurements were made with the lens wide-open. Stopping down the aperture will make all of the readings even better. I turned the lens vibration reduction off for these shots, and I of course used a large tripod.
MTF50 results at 4 mm f/3.0
The MTF50 lp/mm resolution measurements at 4mm show how much better the lens is in the sagittal direction (think spokes of a wheel) than the meridional direction.
4 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.49 368 1699 3399
Corner 0.32 240 1110 2220
4 mm f/3.0 MTF10/30 chart
The MTF10/30 chart is misleading, since it’s based upon a jpeg image with default sharpening. Again, this makes the resolution and contrast look better than it really is, when compared to raw, unsharpened shots. I don’t have a choice here, however. The lens meridional direction is consistently worse than the sagittal direction at all focal lengths.
When the sagittal and meridional resolution differs by a significant amount, you get astigmatism. You will note that astigmatism starts to become a problem at about 2/3 of the way from the lens center when zoomed to 4 mm.
4 mm corner detail (cycles per pixel on each edge)
You can see a huge quality difference when comparing the edges that point toward the image center (sagittal) than the meridional edges. This is why it’s important to analyze the edge directions separately.
Notice in the picture above the minimal chromatic aberration, which gets emphasized in a high-contrast shot like this.
MTF50 results at 17.6 mm (99 mm equivalent) f/4.6
17.6 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.38 285 1318 2636
Corner 0.33 248 1144 2289
17.6 mm f/4.6 MTF10/30 chart
At 17.6 mm (99 mm equivalent FX) the resolution is just plain spectacular. Astigmatism is very well controlled.
MTF50 results at 35.9 mm (202 mm equivalent) f/5.4
35.9 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.41 308 1422 2844
Corner 0.34 255 1179 2358
35.9 mm f/5.4 MTF10/30 chart
Very, very good resolution at 35.9 mm.
MTF50 results at 52 mm (294 mm equivalent) f/5.7
52 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.41 308 1422 2844
Corner 0.34 255 1179 2358
52.2 mm f/5.7 MTF10/30 chart
Meridional resolution is taking a nosedive in the corners here, but sagittal resolution is excellent.
MTF50 results at 70 mm (394 mm equivalent) f/5.9
70 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.40 300 1387 2774
Corner 0.31 233 1075 2150
70.0 mm f/5.9 MTF10/30 chart
Again, meridional performance in the corners is pretty bad, but the sagittal performance is great.
MTF50 results at 160 mm f/6.5
160 mm cycles/pixel Mtf50 lp/mm lp/ph l/ph
Center 0.27 203 936 1873
Corner 0.21 158 728 1457
160 mm f/6.5 MTF10/30 chart
Performance takes a giant hit at maximum zoom. Be that as it may, check out the shot below of the moon at 160 mm; it may not compete with ‘pro’ monster lenses, but it still looks pretty good.
160 mm (900 mm equivalent) f/6.5 1/1000s ISO 125, un-cropped
Rufus hummer detail crop
The default in-camera noise reduction slightly smears fine details, even at low ISO and bright light. The majority of the ‘smearing’ is probably the meridional-direction weakness in the lens, however.
160 mm f/6.5 1/250s ISO 125, slightly cropped.
The moon shot was hand-held at maximum zoom. Kudos to the VR system in this lens! Considering the maximum zoom being used and that I hand-held the camera, these results are nothing short of fantastic.
5 mm f/3.2 1/125s ISO 125
There are certainly limitations with the B500 camera, but it is capable of very high quality photographs. Keep in mind that this camera/lens combination costs less than many DSLR kit lenses!
It’s hard to draw any blanket conclusion on this camera. It has a weird mix of really nice and really irritating features. Personally, I require more direct control (P, A, S, M) than this camera provides, and I really missed raw-format files. The lens, however, exceeded my expectations; with a bit of coaxing, it’s possible to take really nice pictures with the B500.