Lately, there has been an increasing interest of Anti-Reflection (AR) coatings in the infrared photography community.
One company has boldly stated that their internal AR coated IR filters can reduce or even eliminate lens hotspots. To be fair, years ago we also experimented with AR coatings on internal filters but for a different purpose, to improve the light sensitivity by having more light pass through the filter instead of being reflected off of it’s surface. The end goal being shorter exposure times, not hotspot reduction. More about that below.
Since the competitor’s claims have been making enough waves that our customers have been asking us about the validity of AR coatings for this purpose we decided to put it to the test. The goal of the tests was to determine conclusively whether or not internal Infrared filters with AR coatings offered any advantage for hotspots, as is claimed by the competitor. The methods and results are to follow but first a quick refresher on IR hotspots:
IR Hotspots can appear at the center of an image and have a circular area that is usually lighter and of less contrast, occasionally also having some color shift. These hotspots tend to be less obvious at a wide aperture and progressively get worse as you stop down. The only good solution is to use lenses that don’t suffer from this problem. Some people find they can reduce the issue in post processing software like Photoshop, but this is much more difficult and would need to be done to every image, whereas changing lenses solves the problem instantly for all images captured.
Lenses are designed for normal, visible light photography and when used for a purpose for which they were not designed, issues can arise, like the dreaded IR hotspot. Unfortunately, the true cause of this issue isn’t entirely known as multiple factors could be playing a contributing role but the consensus is that the issue is caused by internal reflections of IR light within the lens itself, be it from the coating on the lens barrel being IR reflective or IR light interaction between the lens elements that make up the lens or the coatings on the lens elements themselves or perhaps a combination of all of these. Regardless of the cause, we know definitively that the issue occurs within the lens itself.
When we heard of claims being made that AR coatings on internal infrared filters that are installed over the sensor inside the camera can reduce or eliminate IR hotspots, we were naturally quite skeptical. The claims made no sense to us, why would more light reaching the sensor help in spots that are already formed in the lens BEFORE reaching the sensor? Nonetheless,
So we set out to test the claims. For our test, we used 3 Sony A6000 cameras:
- The first camera was converted with our uncoated Deep BW IR (830nm) filter.
- The second camera was converted with our own AR coated Deep BW IR (830nm) filter, we used the best of the stock we have.
- The third camera was purchased directly from and converted by the competitor with their AR coated 850nm filter.
For the test, we took the same shot with all 3 cameras with a series of both Sony E/FE lenses known to have hot spots and also with several Canon lenses used with an adapter.
The following lenses were tested as they are known to have hotspots:
Sony 16-35mm f4, Sony 16-50mm f3.5-5.6 (kit lens), Sony 16-70mm f4, Sony 24-70mm f2.8, Canon 16-35mm f2.8 II, Canon 24-70mm f2.8 II, Canon 50mm f1.4
These two sets of images are from all 3 of our cameras with the Canon 50mm f1.4 lens.
The first set of images are with the same exact setting straight out of the camera, the second set are the same images with exposure adjusted to look identical.
Life Pixel’s uncoated filter Canon 50mm f1.4 lens 50mm 1/30sec f22 |
Our AR coated filter Canon 50mm f1.4 lens 50mm 1/30sec f22 |
Competitor’s “hotspot reducing AR coating” Canon 50mm f1.4 lens 50mm 1/30sec f22 |
Life Pixel’s uncoated filter Canon 50mm f1.4 lens 50mm 1/30sec f22 – exposure adjusted |
Our AR coated filter Canon 50mm f1.4 lens 50mm 1/30sec f22 – exposure adjusted |
Competitor’s “hotspot reducing AR coating” Canon 50mm f1.4 lens 50mm 1/30sec f22 – exposure adjusted |
As you can see from the test images, we were blown away to find that our uncoated filter actually looked better than both our AR coated filter and the competitor’s AR coated filter. It was very shocking to see that the AR coating was making the hot spot worse and more exasperated. We also discovered that the coating was somehow reducing light sensitivity. We are certain that we took the same shot with all 3 cameras with the same aperture, ISO, and shutter speed seconds apart on the same tripod and light remained consistent as it was a clear sunny day and yet all the AR coated images were noticeably darker. So to conclude, our testing showed that AR coatings actually did not help with IR hotspots in any way and actually made the hotspots worse as well as losing significant amount of light.
We are confident in the tests performed and are glad to finally have some concrete data for our customers to review and to help them make informed decisions about their IR camera conversion options. While we will always continue to seek materials and methods to improve our conversions wherever possible, including ways to reduce hot spots, LifePixel will not be offering internal AR coated filters for the purpose of reducing IR hotspots as it does not work. Based on the test results, the data is just not there in support of these claims and frankly seems to be just another gimmick aimed at profit gains.
Below we will leave more examples from our testing with all lenses, we will follow the same pattern to do the first set of images straight out of the camera and the second set with exposures adjusted. Left to right: our uncoated filter, our AR coated filter, competitor’s coated filter.
Sony 16-70mm f4
 |
 |
 |
 |
 |
 |
Sony 24-70mm f2.8
 |
 |
 |
 |
 |
 |
Canon 16-35mm f2.8
 |
 |
 |
 |
 |
 |
We have many more images taken at various apertures that show this same conclusion, if you are interested in seeing more of our test shots, feel free to leave a comment below or email!
Latest articles by Noah Ashurov (see all articles)
- The Truth about AR Coated Infrared Filters - February 7, 2018
- The 4 Best Canon UV Conversion Set-Ups for All Budgets - January 3, 2017
- How To Get a FREE Camera Conversion - December 2, 2016
Noah Ashurov
All LifePixel customer service inquires should be directed to service@lifepixel.com or by phone at 866-610-1710.
See All Articles by Noah Ashurov
I’ll approach the hot spot issue a different way. Is there a Photoshop action to correct this, or a method that works in Photoshop to correct this issue?
Thanks,
Brian
Hi Brian,
Some hotspots can be minimized by removing the color in the hotspot area, usually red, and then slightly darkening the area to match the rest of the image. That is a best-case scenario. With strong hotspots, some detail is gone, and you cannot fix that in post.
As this competitor offers this coating as an option at $100 a pop, I think for a C note I could go a long way towards buying another lens, used perhaps, or using something I already own.
Anyway, thanks for another informative article.
I like the pictures
If hotspots come from the lens then it actually makes sense that an AR coating would transmit the effect more clearly. It’s not that the hotspot changes from AR coating but is simply revealed “better.” This would be analagous to the way you get better transmission from a coated filter on the front of your lens. It’s been a long time since I studied the physics of optics but it makes sense from a practical standpoint. It would seem that scatter and/or flare may be obscuring the effect slightly without the coating,
The real question is if there’s any noticeable difference in overall quality with lenses that do not have hot spots. Have you tried this comparison? I’d be curious to see the results.
Personally I’d rather have the utmost acutance and overall quality and just use lenses that don’t have hot spots. Having to choose the right lenses is just life in the infrared world.
Hi Joel,
Very informational comment, thank you. You are 100% correct in that it makes sense for the hot spot to get worse as there is better transmission with the AR coating.
We have done testing on lenses that experience no hot spot and found no difference in quality, other than the fact that you lost about a stop of light with the coated filter.
We also did testing on lenses that had a very mild hot spot when stopped all the way down to f22 (where hot spots are always the worst) and those did look significantly worse with the AR coated filters.
Not being able to use some lenses is something that is very difficult for some folks to accept as we get attached to things easily. Especially when it’s a lens that performs great in visible light. But as you said, it is just life in the infrared world. Some folks hear “hot spot reducing” and they jump on the train in hopes of a magic fix to an issue we all encounter. That is why we chose to make this information public as we have heard a lot of complaints about how this didn’t help their hotspot prone lenses.
Thanks again for the informative comment.
Thanks Noah.
Do you have any theories as to why there would be a loss of light with AR coating? It must be the particular coating blocks some infrared…maybe you already mentioned this (?) If so I would presume the filters closer to the visible spectrum have less of this problem? Just curious.
Your guess is as good as mine! Our initial thought was actually what you mentioned in that this particular coating must be blocking some infrared instead of allowing the IR light to pass to the sensor. Hypothetically speaking, we could’ve toyed around with the coatings more to find something that a better result (as you can see our AR coating is still losing some light but not as much as the competitors) but it still makes the hotspot a bit worse which is the reason everyone was so excited about this in the first place.
I have the Canon 16-35 f4 (the 2.8 was too pricey for me). Here’s hoping it will do as well as the f2.8 version.
Note I too ran into the hotspot issue on my Canon 24-70 f2.8 II from f4 up through f22; just could not seem to get rid of it.
Here’s hoping my 16-35 does better! :0)
The most shocking revelation of this article is that LifePixel has a competitor. My converted D70 has served me well for years, but I’ll be shopping for an upgrade from you in the near future.
My IR camera is a Fuji XE1 conversion to 590nm but mostly used with 720nm and 850nm lens filters on the Fuji 14mm f2.8 lens (best ever IR lens IMO) but even that sometimes begins to show some central lightness at f8 depending on the light’s direction.
Since Kolarivision state their AR filter is used to reflect the light which is causing the hot spots would it not make sense to coat a lens filter with the AR to prevent this problem light entering the lens in the first place.
What actually causes hotspots? Is it the lens elements or their grouping or the aperture blades maybe?
Hi Vic,
The Fuji 16mm f1.4, 23mm f/2 and 35mm f/2 are also great for IR! They seemed to have really nailed it with their prime lenses!
Believe it or not, all of our external filters actually have an AR coating on them, they always have, and always will. Not because it reduces the hot spot but because it just makes sense externally. You’re adding an extra layer of protection from scratches and brings up the overall hardness of the glass to reduce chance of damage.
Technically what you are saying SHOULD be correct. If we use these filters externally, we could be reflecting this light from ever entering the lens in the first place. Unfortunately, we don’t know exactly what is causing it and that is the bigger fish to fry. Hot spots can be caused for a variety of reasons, the most common of which is the coating on the inside of the lens barrel being reflective in IR light. Another frequent reason is light interaction between the coatings of the lens elements. A less common reason is light interaction between the lens elements and the imaging sensor (usually the micro-lenses) in the camera. In this case using the lens on one camera model may have no hot spot issues and on another model there could be prominent hot spots.
So it’s definitely a bit more complicated to fix this issue than throwing a coating on the internal IR filter.
Thanks for the comment!
I use my Canon 24-70mm f2.8 on my Life Pixel converted 5D Camera all the time without experience any centre hotspots. Could you be wrong about this lens?
Hi Steve,
There are two versions of the 24-70. There is the original and a newer Roman numeral II version. The older version is great for IR but the new one has a little bit of a hotspot. Visit this page for our testing with the lens. Also depends on what filter you are using as the Deep B&W is the worst offender in terms of hotspots.
Thanks
Interesting study, Noah. I own the 24-70 II and my filter is the Deep B&W. I rarely shoot wide with this lens and use it for zoom purposes which is maybe why I’m not seeing the hotspot. My usual go-to lens is the EF 17-40 f/4 as calibrated by LifePixel.
Zooming in definitely helps with this particular lens!
Very interesting! I’ll stick with you folks!
Btw, while the AR filters may prove “exasperated“, I think you meant “exacerbated”. Dyac?
Was not an error but I guess both words work! Maybe you should be writing our articles :).
This test is not valid. Kolari Vision claims their AR coating makes lenses more usable in the way that there is one or two f-stops gain before hotspots start to appear. i.e. if without AR coating hotspots become visible at f/5.6, with AR coating they become visible at say f/8 or f/11. So, I would really like to see this test repeated. Use a lens known for producing hotspot, find an aperture where hotspot becomes visible, then compare with AR filter converted camera at the same aperture.
Hi Krešimir,
We actually did the test at multiple apertures, including when the hot spot is just beginning to appear. We only put the f22 images up here because this is where you can most clearly see what is happening. This same trend followed whatever aperture we shot at and what actually ended up happening is the exact opposite of what you stated. Where as the hot spot usually beginning at f8 or f11 with an uncoated filter, you could start to really see if begin at a higher aperture with the AR coated filter. If you would like to see more images from our testing, feel free to email me. Thank you for your time.
Thank you for your reply. If that is the case, good thing I have ordered IR conversion from another of your competitors here in Europe who doesn’t offer AR coatings. I’ll just stick with the lenses that work good in IR. Good thing you have ”drop ship from a retailer” option which I might use next time.
You can definitely drop ship cameras to us from any store, we even sell both new and used cameras if that makes it any easier. Have a great day.