[Filter Test] Blocking UV for Visible Photos Made with Full-Spectrum Camera: Part #1

[Andrea B.]

I'm out the door to begin an experiment discussed below.  Also the SigOth has a minor concert thingie (for friends) coming up, so I have to do the dreaded housework. (noooooooooo...... not housework again!!!!) I hope I'm back in a couple of days. If not, then please send rescue. Preferably one of those nice Rescue Dogs with the brandy cask around its neck. Altho, out here in the Wild West a brandy-laden Rescue Llama might be the way to go. Sadly, the Rescue Bobcat idea didn't work out. Turns out that Bobcats have a weakness for nips of the brandy and tend to lose their way.

---

ADDED 12 Dec 2021:  In retrospect, I think this was not a particularly good test. I'm going to try again here: LINKIE

 

 

16 Dec 2021:

Part #2 is here: LINKIE 

Part #3 SUMMARY is here:  LINKIE

 

 

 

---

 

 

[FILTER TEST] Blockiing UV for Visible Photos Made with Full-Spectrum Camera:  Part #1

 

Background

Explore UV-contamination and UV-blocking on a full-spectrum converted camera used for Visible photos.

Is UV-contamination a problem? If so, what is the effect?

If so, which UV-blockers are best for preventing it?

 

In the olden film days, UV-haze was a problem in distant landscape photos. The short-ray scattering caused a kind of veiling or fog which obscured details. UV-cut filters were very popular and were routinely left on a lens as a protective filter. (Lens hoods are better for that, but I digress.)

 

These days we know that our stock digital cameras have sufficient UV-blocking from internal filtration. But a full-spectrum converted camera needs both UV-blocking and IR-blocking when used for a Visible photograph. We have discussed IR-blocking and all agree

that S8612 glass works best in UV-pass filter stacks and that BG38 is good for IR-blocking in Visible work with a full-spectrum body.

 

But we have not had a lot of discussion about the effects of UV contamination, if any, on a Visible photo from a full-spectrum camera. Let's look more into this to answer the questions above.

 

 

---

Filters

 

I will start with the only filters I currently have that are useable for UV-blocking.  I will also try to get some more UV-blockers or UV/IR-blockers  in to continue the experiment after this initial work.

 

 

• Singleton IR-Blocker:  BG38 x 2.0 mm
  IR-block, passes UV and Visible light.
  Are there any effects from the UV light on the Visible photo?
  Here is an experiment using various blue-green IR blockers, including the BG38, for Visible work: LINKIE

 

• Singleton UV-Blocker:  Baader UV/IR-Cut
  Blocks both UV and IR light.
  This filter has sharp left & right shoulders. On a converted body it passes more violet/blue and orange/red than a stock camera. Does the Visible photo show that? How does this filter compare to the singleton BG38? Do we see a difference which could be attributed to UV-suppression?

 

• Stacked UV-Blocker:  BG38 x 2.0 + LP 400
  Blocks both UV and IR light.
  This is an obvious combo which, surprisingly perhaps, I have not yet tried. In theory, it should provide a sloped-shoulder Visible curve on both sides. How does the resulting Visible photo look in practice?

 

• Stacked UV-Blocker:  Baader UV/IR-Cut + BG38 x 2.00
  Blocks both UV and IR light.
  This last summer I experimented with adding  a BG38 to a Baader UV/IR-Cut to improve Visible color by giving some slope to the sharp right-hand shoulder of the Baader UV/IR-Cut. So this combo is an obvious choice to use again.

 

---

Test Scenes

 

As in film days, I would think that distant scenes would likely show the UV-haze problem in a Visible photo from a full-spectrum conversion. But to confirm that we probably need to also check medium range and close-up distances.

 

Long distance photo, distant mountain range for which UV-scattering is obviously evident to the naked eye.

Medium range photo, generic landscape, some shrubs, some sky.

Close-up photo, not macro, just a few feet

 

 

---

Gear

 

• Panasonic S1R, full-spectrum conversion, or
• D610 full-spectrum conversion
• Coastal Optics 60mm f/4.0, to maximize capture of UV light
• Filters, as above
• Sunlight
• Tripod, hat, sunscreen, lip chap stuff, sturdy shoes, notebook, pack llama to carry gear, rescue llama on standby, lots of patience

 

 

---

Other Potential Filters for Future Tests

 

I would like to try to get at least another one or two more UV-blocking filters to work with in later additions to this experiment. Here are some which have been suggested.

• Zeiss T*
• Tiffen 2A
• Tiffen 2E
• B+W 486

 

 

 

---

First Results

 

I photographed an afternoon scene facing north-northwest. The shrubby hills are about 3-4 miles away. Behind the hills are some mountains. Look to find snow-topped Old Baldy just peeking up behind the hills. In the foreground are some juniper shrubs and dried grasses. The roofs of local homes are visible just above the shrubs.

 

Side Note:  I should have made a photo of the scene with a stock camera. I'll do that for the next set.

 

For each filter and stack I photographed the Spectralon rectangle to use for white balance. I also photographed a Color Checker card for eventual color correction profiling. All exposures are:

• f/16
• ISO-100

 

 

---

Conversion

 

I could have easily used one converter, but I have been enjoying playing around with various converters recently. 

DxO Photo Lab for initial conversion and white balancing. Produce master TIF.

Photo Ninja for Details +15 on the TIF. Minor other tweaks not worth mentioning.

Luminar for cloning out YADDB (yet another dumb dust bunny) on the TIF.

Photo Mechanic for cropping, resizing and creating JPG for upload.

 

VERY IMPORTANT:  THERE IS NO Additional COLOR PROFILING beyond the default Panasonic Lumix S1R profile supplied by DxO Photo Lab. One of the follow-up experiments might be to determine whether a color correction profile made from the Color Checker card would make the following 4 results identical or nearly so. But for now I'm just trying to nail down the basic UV/IR-cutting properties.

 

 

---

Enough Talking, Let's Look

 

Here are the 4 results. 

For the best view, close all sidebars and expand your browser as wide as possible to undo the responsive resizing. Try clicking up the photo to its maximum 1200 pixels. Stand back a couple of feet from the monitor and look at the colors.

 

This is not about "likes". This is about color reproduction. This is about possible UV-contamination and its effects. Look for differences and try to explain them.

 

Photo 1

 

 

Photo 2

 

 

Photo 3

 

 

Photo 4

 

 

 

 

---

Supporting Data from Raw Histograms

 

The non-white-balanced raw composites are full of green and very boring. So let's just look at the raw histogram. Here's the lograthmic version.

 

Please ignore small differences in exposure. I was trying to pick the best photo to illustrate results. Sometimes one picks the file needing a minor half-stop tweak on the exposure slider over the slightly better exposure which had a bit of blur.

 

One histogram, the first, is quite obviously different from the other three. Histograms 3 and 4 are very similar.

 

Raw Histogram 1

 

Raw Histogram 2

 

Raw Histogram 3

 

Raw Histogram 4

 

 

 

 

 

 

[Andrea B.]

Sky Only Histograms

The sky areas was selected for the raw histogram. These are linear.

Look at the relationship between the green and blue masses.

 

 

1 Sky Only Raw Histogram

This tells me, I think, that There is a cyan sky. which is just a little bit more green than it should be for a good sky color. (Andrea, don't overinterpret.)

 

 

 

 

2 Sky Only Raw Histogram

This histogram indicates to me there is more blue and blue-violet in the sky area than in any of the other three photos. This is apparent in Photo 2.. Can we conclude that this filter passed some UV which was recorded in the blue channel?

Nah, I'm overinterpreting this. There's a bit more blue. Stop.

 

 

 

3 Sky Only Raw Histogram

Again, a cyan sky.

 

 

 

4 Sky Only Raw Histogram

This sky is cyan but moving slightly towards blue. There is less green in the sky.
Added: But that is not easy to see in the photo.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[dabateman]

Photo 2 has the most uv entering.  The mountains are hazy and the white peak isn't very visible. 

Photo 1 looks best to me, with photo 3 a close second or tied for first.

The sun or lightning changed between photo 3 and photo 4. So hard to make a judgment.  Photo 3 has darker hills than photo 4, but I think a cloud moved or the time is different between these shots.

 

[Andy Perrin]

I don't quite understand this test, because it seems multiple variables are being changed at the same time, specifically the red is being attenuated in some but not others, and simultaneously the UV is being cut in some but not others, sometimes both at once. So this is a mixed test. I don't know what we will be able to conclude.

 

I vote that the Baader alone be removed from the running here, or at least identified. It is not a fair comparison if we want to know the effect of various types of UV blocking.

[Cadmium]

Of the other list of filters, the Zeiss T* is definitely a winner for cutting UV off at the 400nm point.

Never tried it, but stacking it with BG38 might be something to try, and compare.

Get one of those, they are great for UVIVF also.

 

[Fandyus]

2 looks the most realistic to me

 

It's the most bland, but the sky is hardly ever as brilliant as we edit and desire it to be.

 

Edit: just realized I made this judgement while I had a blue light filter applied on my device as it's long past bedtime. I'll look at this tomorrow and make another judgement.

 

Edit 2: after having a better look, 3 appears the best, but 2 is still alright looking as well.

[Andrea B.]

I don't quite understand this test....

 

There are 3 UV-blocking filters in the test and one which does not block UV. I wanted to look at the results to determine whether we can see a difference between UV-blocked and non-UV-blocked photos. We cannot judge the efficacy of a UV-blocker if we don't know what a non-UV-blocked photo looks like in comparison.

 

I've never done this experiment before so I did not know how it would turn out. It took me a couple of hours to work through what I was seeing and to understand it. I think I've got it now. But I might have it wrong. We'll see.

 

.....just realized I made this judgement while I had a blue light filter applied on my device

 

That is a good point to mention because all color is viewed in context. And the human brain has its ways of maintaining color constancy in spite of what the eye is sending it.

 

 

 

 

[Andy Perrin]

37 minutes ago, Andrea B. said:

I don't quite understand this test....

 

There are 3 UV-blocking filters in the test and one which does not block UV. I wanted to look at the results to determine whether we can see a difference between UV-blocked and non-UV-blocked photos. We cannot judge the efficacy of a UV-blocker if we don't know what a non-UV-blocked photo looks like in comparison.

No, there are 3 UV blocking filters and one that doesn't block UV. But there are also three filters that apply a red slope and one that does not. The combinations are:

 

1) Singleton IR-Blocker:  BG38 x 2.0 mm   - does not block UV, red slope, blocks IR

2) Singleton UV-Blocker:  Baader UV/IR-Cut - blocks UV, no red slope, blocks IR

3) Stacked UV-Blocker:  BG38 x 2.0 + LP 400 - blocks UV, red slope, blocks IR

4) Stacked UV-Blocker:  Baader UV/IR-Cut + BG38 x 2.00 - blocks UV, red slope, blocks IR
 

I'm saying the proper comparison is between (1), (3) and (4) here, and (2) doesn't really belong because two variables are changing, not just one. If the intent had been to see the effect of the red slope, then (2) would make sense, but I thought this was about the UV-cutting.

[Andrea B.]

So, just ignore the photo which turns out to be from the Baader UV/IR-Cut. I must test that filter anyway because it is a UV-blocker used by many people.

 

If I get some more UV-blockers in to test, then later we can group them into the square-shouldered variety and the slope-shouldered variety.

---

 

I'm not sure it is the sky which tells the complete tale in this test. The Sky 1 is the most saturated while Sky 2 is the least saturated. Sky 3 and Sky 4 are about the same saturation. But it is not much of a saturation difference.

 

As for hue, Sky 1 is only about 5° different from Sky 2,3,4. So I think my earlier remarks about sky color based on looking at the histograms were a bit of over-interpretation.

 

But there is something about Photo 2 which seems anomalous. And similarly Photo 1. At this point everybody should have figured out which two photos are made with the singletons and which two are made with the stacks.

 

 

 

 

[Andrea B.]

Photo 2 has the most uv entering.  The mountains are hazy and the white peak isn't very visible. 

 

There you go! Dave nailed it.

 

---

 

I sometimes think the raw histograms are more confusing than useful. I crossed out some of my histogram remarks above because I think I was reading too much into the charts.

 

 

[Andy Perrin]

2 hours ago, Cadmium said:

Of the other list of filters, the Zeiss T* is definitely a winner for cutting UV off at the 400nm point.

Never tried it, but stacking it with BG38 might be something to try, and compare.

Get one of those, they are great for UVIVF also.

I just got my Zeiss T* in both 77mm and 52mm. I do indeed like them! 

 

That said, an interesting point to consider is that in the recent PetaPixel article on this subject, they found that the built-in filters cut at the following wavelengths for different brands:

 

Of these, Sony is an obvious outlier, and PetaPixel author's opinion was that people were standardizing on 420nm as the cutoff, especially since Sony got criticized for being too blue. So maybe the right filter is something like the Haze 2E, which is a 420nm cut, rather than the Zeiss 400nm.

[Andrea B.]

That is so interesting !! The violet is being cut. No wonder it is so difficult to photograph.

[Doug A]

416, 418, 420, and 422nm. Where are the camera companies getting such tightly spaced glass? Doubt Hoya and Schott are making so many different filters so close together. Who else makes this kind of glass?

Thanks,

Doug A

[Andy Perrin]

6 minutes ago, Doug A said:

416, 418, 420, and 422nm. Where are the camera companies getting such tightly spaced glass? Doubt Hoya and Schott are making so many different filters so close together. Who else makes this kind of glass?

Doug, you are taking those numbers as exact, but those are averages. If you click through and look at the graph  you can see that each manufacturer has a spread of values. The brands do cluster around the average but each number is like +/-5nm. That said, my guess is that they are coating the glass, so it's a mixture of absorption glass and dichroic coating.

[dabateman]

Yes if you look at the filters that are removed from our cameras during conversion, you can see they have some coatings to them.

 

Thats why a stock Olympus is the best right, allows light in all the way down to at least 380nm :)

A stock Panasonic is m43rds choice for less UV an little bit of violet. 

 

This test does indicate that with a full spectrum converted camera and a great UV passing lens, than a UV blocker is needed.

Or use a stock camera or a lens without any Uv transmission or both.

 

[Andrea B.]

•  

The more I look at the 4 photos the more confused-er I get !!

("more confused-er" - american slang).

 

I need to do more work on this UV-blocking area. I'm not sure why we haven't looked into this before because many of us use our converted cameras for Visible photos. Like when shooting UV or IR documentary, I always want an accompanying Visible reference photo made at the same time in the same light.

 

FILTERS

• Photo 1 Baader UV/IR Cut, singleton
• Photo 2 BG38, singleton
• Photo 3 BG38 + Baader UV/IR Cut, stack
• Photo 4 BG38 + GG400, stack

 

To my eye, it was very obvious that Photo 1 and Photo 2 were different from Photos 3 & 4 (which were most alike of the set). Also obvious that Photo 1 and Photo 2 were quite different from each other.

 

Anybody using the Baader UV/IR Cut has known for a long time that this filter tends to pass too much orange & red due to its square right shoulder. It also has a square left shoulder, but does not pass too much violet & blue in natural sunlight because there is less violet & blue to pass. I've found a color correction profile improves the color in most Visible photos made with the Buvircut but does not completely remedy the colors for some photos.

 

The BG38 singleton photo shows evidence of UV scattering at a distance. I can't quite put my finger on how Photo 2 differs from Photo 3 & 4 other than by some fairly minor color differences and maybe a touch less saturation in the sky. Also, as David noted, there does seem to be more haze on those mountains in Photo 2, more violet tones.

 

After noting the results from PetaPixel which Andy posted, I think that I need to test the BG38 with my longpass 420 mm. The camera makers put real money into deciding these cutoffs. So, even tho I happen to think, as a result, that violet is difficult to photograph, they must have some reason for excluding it near 400 nm. Probably they cut off after 400nm simply because when making long distance photos, there is some apparent UV-haze effect, as we have just noted.

 

To extend this experiment:

• Make some medium-distance and close photos.
• Test the BG38 with GG420. (I don't have anything between GG400 and GG420.)
• Add the Zeiss T* and perhaps others.

 

All comments, suggestions, corrections welcomed, as always.

 

 

[Cadmium]

Great, the filter names.  They all look so close to me, except the top one (Baader UV/IR-Cut) looks redder to me, I see it mostly in the grass.

That is my experience with the Baader UV/IR-Cut, redder than even the BG38.

 

Let's not forget that visual can be achieved even better on SOME CAMERAS with other BG filters (BG40, BG39, S8612).
USUALLY, BG38 or BG40 work best for most cameras, but someone sent me tests of all those and S8612 was absolutely the best for their camera.

I think, not sure, their camera was the T1 IR.

Andrea, the KV-418 might be another one to try. 418nm cut off (50% I think). Get one if you still can somewhere.

[Andy Perrin]

Andrea, regarding cutting at 420nm, the PetaPixel article suggested that it was because skin tones came out too cold with a 409nm average cutoff like Sony has. So it’s possible that YOU might actually be happier with a lower cutoff for flowers, since that was not the motivation the manufacturers probably had for going higher. 

[Andrea B.]

It seems obvious that UV-blocking filters are simply longpass filters, most likely coated. See supporting evidence in next link.

 

Looking at Clear and UV Filter Spectrograms
by Roger Cicala, 2017
LensRentals.com

 

Amongst the measured filters:

• B+W Digital Pro UV Haze
• Heliopan UV SH-PMC Multi
• Zeiss T*

 

 

---

 

 

It also seems obvious that internal UV/IR-cut filtration in a camera is a combination of a longpass filter together with a blue-green blocker. (I'm assuming that internal camera filters are non-dichroic.) See supporting evidence in next link which contains the chart of 82 internal filters from various camera brands.

 

UV Lens Filters: A Spectral Analysis

by Pat Nadolski, 2020

KolariVision.com

 

 

---

 

 

So therefore, why experiment with a Zeiss T*? How can that filter have a substrate which is anything other that Schott GG longpass glass or any similar longpass glass from Hoya or others? Yes, anti-reflective coatings were probably added to the Zeiss T*, but this is about cutting off UV.

 

ADDED LATER:  The AR coating!! That is why we need a Zeiss T*.

Possibly, coated BG38 will become available some day. OR coated GG400.

 

I think that the BG38 + Schott GG400 or BG38 + GG420 or some similar BG + GG stack represents the basic, non-dichroic UV/IR-cut filtration - the archetype of UV/IR-cutting.

 

Yes, there might be nuances to the choice of BG glass. As mentioned above maybe BG40 works slightly better for some cameras. (Not sure why that would be true, but then I don't test all cameras to know for sure.) And yes, there might be nuances to the choice of longpass cutoff anywhere between 400 - 420 nm. 

 

I think we should just recommend BG+GG to kill UV/IR in full spectrum cameras for those who want to use their full-spec cam for Visible photos. It would the the least expensive way to go because the dichroic filters are very pricey in comparison.

 

Any comments or suggestions or rebuttals or support on this?

 

 

 

[Andy Perrin]

Quote

I think we should just recommend BG+GG to kill UV/IR in full spectrum cameras for those who want to use their full-spec cam for Visible photos. It would the the least expensive way to go because the dichroic filters are very pricey in comparison.

 

Any comments or suggestions or rebuttals or support on this?

 

Quote

Yes, anti-reflective coatings were probably added to the Zeiss T*, but this is about cutting off UV.

 

Is it about cutting off UV or about getting the best possible visible light photo?

 

I think you may be right about most of them being a BG+GG combo at core, but I don't think your final conclusion is justified because it depends what you are trying to shoot. I use my camera to shoot a lot of things in visible. Lately I've been trying to shoot Christmas lights. Ghosts and other partial reflections are both an issue with light coming directly at the lens. Those coatings make a real difference to the quality of the resulting photos. If ALL a person wants to achieve is a spectrum approximating what the internal blocking filter gives, going with uncoated BG+GG is enough, but personally I do care about other things like AR capability. So maybe another round of tests looking at glare issues is needed.

 

Along those lines, a thing that might be fun to test is the Kolari Vision UV/IR Cut Hot Mirror Pro 2, the new version. If that's a cemented stack, it might be even better with regard to ghosts/reflections, which may be a good thing to test... the current test against a far horizon cutting haze is certainly one thing to look at, but it's not the only thing.

 

Do you think there's any way that they might LOAN you a review copy? 

[colinbm]

Andrea, do you have a Zeiss T* ?
As you can see, in the links that you provided, many cameras have different spectrums with there UV / IR cut filters.
With a converted camera we have the ability to try & match the original UV / IR cut filter to get a good visible image again.
Even if we can match the curves of the OEM, it can still not be enough to get a good visible image, so some fine adjustments of the filter stack will still be needed.
But the Zeiss T* has a deeper cut then the others to block more of the UVA leak into the visible light, especially with UVIVF.

[Andrea B.]

Andy:  Is it about cutting off UV or about getting the best possible visible light photo?

 

Andy, I'm gonna go with an 'and' and not an 'or'.

You comment was well made.

I'm getting ready to write a little summary which includes everything suggested.

---

 

 

Col, no, I don't yet have a Zeiss T* here. On order though.

 

---

 

[Andrea B.]

16 Dec 2021:  Most of this Initial Summary made it into Part #3 SUMMARY here: LINKIE

In the Final Summary, I delisted the Baader UV/IR-Cut as "recommended".

---

 

Here is my Initial Summary of this experiment.

 

• First, let me observe that the Zeiss T* UV-cut filter must have a substrate of Schott GG400 longpass or similar.
  Schott and Zeiss have a long history together. I do not know for certain, but what else could it be but GG400 or similar?
   
• Second, let me give due recognition to the fact that anti-reflective (AR) coating is a good thing.
  And we'll definitely take it when we can get it.

 

In order of increasing quality, here are some recommendations for UV/IR-Cut filter stacks when using a full-spectrum converted camera for Visible photographs. There are some name-brand UV/IR-Cut laminated filters or UV/IR-Cut dichroic filters which have not yet been tested here. 

 

• Singleton: Baader UV/IR-Cut
  • dichroic
  • good UV/IR-cutter, but passes too much red
  • too expensive
     
• Stack: Schott BG38 + Schott GG400 (or GG 420)
  • filters reusable in other stack types
  • no anti-reflective coating
  • uncoated BG glass needs routine maintenance
     
• Stack: Schott BG38 + Schott GG400 (or GG420)
  • where one of the filters has AR coating
  • AR coating adds to the expense
  • a cemented version of this stack is better
  • if not cemented, filters reusable in other stack types
     
• Stack: Schott BG38 + Zeiss T* UV-cut
  • name-brand filters like Zeiss are more expensive: (US)$67.00 for 52 mm.
  • filters reusable in other stack types
     
• NEW Singleton: Kolari Vision UV/IR-Cut Hot Mirror 2 --- Looks very promising!
• name-brand filters are more expensive: (US)$119 for 52 mm.
• appears to be a laminated, coated stack. Cool!!
• I went ahead and added it to the list based on its excellent UV/IR-cut chart.
  I have one on order to test and will update this summary when it arrives and I've taken it out for a spin.
• Notice in the Kolari write-up they mention "some variation between individual cameras and brands"?
  You can fix that with a Color Checker card (or similar) and a converter app which creates color profiles.

 

Important Footnotes:

1. Substitutions: Hoya also makes blue-green glass and longpass glass, as do the infinitude of Chinese ZWB and ZB filter makers
2. I've tested Schott BG38 for cutting IR with two recent cameras, Panasonic S1R and Nikon D610, both as full-spectrum conversions. My own recent tests showed that Schott BG40 is OK for those two cameras, but BG39 and S8612 were not as good for visible work. 
3. Color correction profiling improves all Visible work with a full-spectrum converted camera. The degree of improvement varies, of course, depending on subject, light and gear in use.
4. There are a lot of UV-cut filters on the market, but most of them go way past 400 nm. To reproduce the original UV/IR-cut filtration in a modified camera, we must stick with a UV-cut which is somewhere between 400-420 nm.

[Andrea B.]

I didn't want to write anything else in the summary post so that it could be referenced.

 

But I wanted to respond to Andy's suggestion of the Kolari Vision UV/IR Cut Pro Hot Mirror 2.

 

No, I'm not going to test that filter for now. Maybe later. That Kolari hot mirror is just coated BG38. It cuts in around 320 nm. For our purposes that seems to be quite a lot of UV passed. Indeed, I question the naming of this filter as an UV/IR Cut.

Correction: Crossed that out because I scrolled right past the Chart for the Kolari UV/IR-Cut Pro Hot Mirror 2

and thought the old chart below it was the one I was supposed to look at. Wrong!

The NEW Kolari UV/IR-Cut Pro Hot Mirror 2 is definitely worth testing, and I have ordered one.

 

I'll also note here that the Kolari UV/IR Cut (H-alpha pass) has the same square shoulders as the Baader UV/IR-Cut. So there's no point in testing that one. It is made with Schott B270 substrate.

[aphalo]

T* is the name Zeiss gives to the antireflection multicoating they use on objectives, binoculars and filters. The original T* from the 1970's has been improved since then according to Zeiss' web pages. The UV T* filter has a very sharp cut-in near 410 nm, which suggests a dichroic (interference) filter type. The data sheet is available online.