UV filters are perhaps the most popular type of lens filter, with many photographers using them to protect the front element of an expensive lens. Unfortunately, there’s a huge range in the quality of UV filters out there. I wanted to put together a methodology to test UV filters, and other filters, more precisely. Today, I’ll be performing those tests on the K&F Concept Nano-X 67mm UV filter.
Table of Contents
Intro to the K&F Concept Nano-X UV Filter
The K&F Concept Nano-X UV filter is a thin UV filter that has some sort of 28-layer “titanium coating” feature that is supposed to reduce flare and maintain sharpness. When I was testing it, I certainly didn’t notice any loss of sharpness with my unaided eye.
This filter comes in a nice case with a non-slip rubber holder inside. This design also fits smaller filters securely, and the rubber holder prevents the filter from rattling around and scaring rare owls. That will come in handy if I ever find a rare owl.
Color Accuracy
The first and most noticeable problem with some filters is a warming or cooling effect. I test this using a relatively constant light source on a Calibrite Color Passport target, checking the white balance settings with and without the filter in multiple trials. From these trials, a 95% confidence interval can be generated for temperature and tint modification:
The way to interpret these graphs is this: the color temperature change was just on the edge of statistically significant, and the average change was a cooling effect of about 24 Kelvin. Basically in the real world, you will not notice much of a change, and any slight change will be easily correctable. Not bad! As for the tint, the change was effectively zero.
But we have to dig a little deeper. Even if a filter changes the white balance a little, correcting it can still result in color inaccuracy in the individual color channels, even if white tones are corrected back to white. (This is especially prevalent in JPEGs, but can also be visible in raw files if it is significant enough.)
So, for the next color accuracy test, I measured the distortion of RGB and CMY colors in degrees on an HSL scale, as well as on two skin tone axes. This test is slightly affected by the lens used. In this case, I tested it with the Nikon 50mm f/1.4 Z, a lens which I’ve found to have minimal color inaccuracies.
Again, these bars represent the 95% confidence interval of the difference between the bare lens and the filter. It’s good if the bars contain zero, bad otherwise, and the strength of the effect is represented by the numbers. The units are in degrees, and typically if the effect is less than 0.5, it will be unnoticeable.
Since all the bars contain zero, with very small effects besides, this filter is pretty much perfect for color preservation. Even the distortion in the first skin tone is well below the threshold of being a problem photographically.
Flare and Contrast
All filters interfere with the light coming into the lens, and it’s just a question of how much.
Unfortunately, it’s very difficult to come up with a measurement of flare that is comparable and meaningful in all situations. That’s because each lighting situation will be different, and each lens also is prone to different sorts of flare that can be even more exacerbated by the filter. Nevertheless, I can say that in a variety of situations, I found the flare to be well-controlled with this filter. In a worst-case scenario test, with the lens already flaring, I compared the before and after contrast reduction once the filter was added. Here are side-by-side images showing the visual difference:
It’s not too noticeable, but there is a slight effect if you look closely in the bottom right corner, which is where the light source was coming from. And here is the corresponding confidence interval of the difference between the lightest bottom square and the darkest:
So, there was a meaningful loss of contrast with a small effect of about 3.2 luminosity units in the HSL scale. Therefore, in the worst of situations, such as slightly off-axis light like sunlight blasting into your lens, taking off the filter will likely improve your contrast a little.
Finally, I did measure the vignetting, which was nonexistent at 50mm, although I did not have wider focal lengths to measure.
Conclusion
Overall, the K&F Concept Nano-X UV filter performed well in my tests. It has excellent color accuracy, no meaningful loss in image quality, and only a slight reduction in contrast in more severe backlit circumstances. It’s also a well-made and durable filter that comes in a nice case. Especially considering the low price, this is an easy filter to recommend. (Prices range from $32 to $53 depending on the size that you select.)
If you’re interested in this filter, consider buying it through one of our affiliate links support the rigorous testing that we do at Photography Life! And be sure to select the correct filter size to match your lens:
My biggest problem with this filter is that it’s backordered at most sizes. If you don’t want to wait for yours to arrive, you may want to check Amazon or Adorama to see if they have the size you need in stock:
I hope that you enjoyed this review! I’m looking forward to testing more filters soon, including with some new testing methodologies that I’m developing. What about you? Do you want to see more filter reviews or scientific comparisons like this one? Feel free to let me know what else you’d like me to test!
I think according to manufacturers, UV filters are supposted to improve contrast and sharpness in some “real world” situations, by eliminating stray light in the UV spectrum. Did you notice any of this in your tests? Or is this a moot point with today’s advanced lens coatings?
Unlike colour film, consumer digital camera sensors have a UV filter and an IR filter. That’s why I use Nikon neutral filters, not UV filters, as lens protectors.
Right, that’s in addition to the lens coatings. I wonder why people keep using UV filters for lens protection (and why manufacturers keep making them), if there’s no benefit compared to neutral filters.
Out of interest, I submitted to Google search:
Is it a good idea to protect my lens with a UV filter?
The first result, which is marked “AI Overview”:
“Yes, using a UV filter to protect your lens is generally a good idea, especially when shooting outdoors or in potentially harsh conditions. A UV filter acts as a barrier, shielding the lens from dust, scratches, and accidental bumps, and can help prevent damage that could otherwise affect image quality.”
And nowhere under its subheading “Reasons to use a UV filter:” does it mention a neutral filter.
Yet another example of AI generating ‘pseudo‑profound bullshit’.
Gordon Pennycook, et al. ‘On the reception and detection of pseudo-profound bullshit’. Judgment and Decision Making. 2015;10(6):549–563. doi:10.1017/S1930297500006999.
I wonder if it’s as simple as whatever coatings they were going to use already block UV light, so might as well market it as a UV filter. I will say that as a film photographer, I actually did seek out a UV filter specifically when I needed something to protect my lens in the desert! But for digital, zero need for them over a regular clear filter.
You might find this interesting:
Looking at Clear and UV Filter Spectrograms.
Roger Cicala. Lensrentals. 2017‑09‑22.
www.lensrentals.com/blog/…ctrograms/
Thanks Jason for the informative article, I use UV filters on all my lenses except the Nikon 14-24mm that filter is two piece and the size of a dinner plate. I can rationalize the cost of a filter over a front element and besides Photoshop AI can help you make those corrections if need in post work. Hope you do the article on ND filters since I use a lot of those to. Cheers!
You’re welcome! Yeah, more ND tests will be forthcoming for sure. They are more interesting I think because they have more color distortion!
I did my own visual test of UV filters many years ago. At that time I found that the better filters were almost as good as the bare lens, but I could visually see a reduction in sharpness and contrast with lesser filters.
But I’ve completely given them up now, why take the chance of flair, reduction in sharpness, etc.? In my 30 years I’ve never damaged a front element on my lenses; just use the lens hood.
True, typically I don’t use filters myself except ND filters for video. Otherwise, it’s bare lens always. The exception might be as someone else posted, when there could be a very high chance of splatter.
Or in the desert with high winds. I got all my front filters scratched on one trip to Oman. I am glad I could just replace the filters rather than having to send the lenses for repair.
I’d absolutely love to see some instrumented tests of Breakthrough filters, and I bet they’d dig it too. Worst case, I’ll send you several of mine.
As we make the system of testing more sophisticated, I think that would be a great idea to test those!
Let me know if I can help facilitate this in any way. Sometimes I have their ear.
I never invested in this item, which is useless to me as long as the lens hood is permanently mounted on the lens. I’ve never had any breakages and I prefer to have one element less.
If it’s just for general protection, I absolutely agree that a hood is superior. I’m actually more interested in ND filters since those are necessary for certain applications.
When I was a newbie I was told that Vivitar series 1 uv filters were the best. I didn’t know any better back then. When I eventually changed them out, I went with a brand that B&H sells called Chiaro, which comes in several series ranging from inexpensive to very expensive. They have differing characteristics, and you get to choose what you want in a uv filter. I use the T98 series, which are not cheap but don’t break the bank. I’ve never noticed any issues with them. I don’t think that they impair my photos in any way. Since you are planning to test many types of filters, I suggest that you test one or more of the Chiaro line.
That is a good idea, Elaine. It would be cool to test the cheap vivitar filter you mentioned too against some of the better ones.
I believe that company went out of business. But… I have a whole bunch of them sitting in my attic doing nothing.
Would love to see a comparison with the NiSi AIR UV’s, it’s a claimed 99,6% transmitance vs 99.9%. I’ve been using the NiSi’s with a Nikon Z8 and a Z 180-600mm and I can’t find any loss of detail while using the filter, even searching at 200%. Did you test for “sharpness” with the K&F UV filter? These X-series are quite good, and for a stupid low price.
If there’s interest, I should be able to test filters for sharpness in our usual lens-testing lab while Jason runs other tests on them! My expectation is that most modern filters won’t lose significant sharpness, but I’d be interested to see analytically.
It’s something I’ve been wondering about.
Getting it quantified would be interesting, especially comparing something you’ll pay good money for, like a good B+W or manufacturer branded clear filter to something like a Chiaro 95% that comes for free with the lens I just bought.
I have a collection of those Chiaro filters since I expect they’re garbage. So why do I get the bundle with the stupid filter? Wish I knew myself that well.
Like Spencer said, we could test for that if there’s interest! That being said, I tried to perform a sharpness test but I could not really notice any differences at all, it’s very hard to tell that sort of thing with any comparable numbers with the unaided eye. I think the differences will be much more subtle than the differences between different lenses.
The tale is that you can’t use filters on a telephoto lens due to loss of sharpness. I know that’s complete rubbish, you must be using a bottle bottom as a filter to lose sharpness nowadays, but I think, for that reason alone, it would be interesting to see lab results :)
Other insteresting tests would be, resistance to water, grease and scratches.
I thought too that good quality filters (in my case Hoya HMC, 77mm diameter) could be used without problem. Yet, of two identical filters, one altered the sharpness of my Canon EF 100-400 mm lens v2, at 400 mm focal length for subjects already as close as 15 meters. Surprisingly that filter worked fine with the same lens at 100 mm focal length, and with others lenses (RF 24-105, EF 17-40; I still use it with those lenses). The other filter is on my RF 100-500, and I have yet to find any problem with it (it was OK with the EF 100-400). So I guess there is some variability among the same kind of filters, so that your tests should also address it maybe with three filters of different production batches (easy to say, more time-consuming and costly to do! :-) ).
One reason to use a filter is to avoid cleaning the front element if taking pictures under harsh conditions (such as in a meadow with splashes of dirty water) to avoid scratching the front lens. Otherwise, the lens hood is preferable to protect physically the front element of course.
The biggest problem I’ve had with filters (including K&F) is the phenomenon where the cheaper ones get stuck to each other or step up and rings (I’ve read this is a metal thread contact problem, especially with aluminum). Usually time in the freezer and a wrench solves it but it’s a pain. I’ve never had this problem with B+W. So optically I think many filters are equivalent but there is something to be said about build too.
Good point. I don’t typically stack filters or use step-up rings, but your information leads me to believe that more rigorous physical tests would be useful in the future! I will see if I can incorporate some.
Yeah aluminium threads suck, the metal is to soft it might get stuck if you apply enough pressure. Top shelve filters use brassm which is more expensive, but much more durable.
Yeah, it’s worse when the filter spins. I got a K&F variable ND for shooting video and it got stuck on my step up ring. What really sucked is that I didn’t apply a lot of pressure to put it on the ring, but spinning it to adjust the ND strength got it stuck. That was a challenge to get off, to put it mildly.
I carry filter wrenches for those occasions.