Leica v Heliopan IR filter anti-reflective coatings

[Guest tummydoc]

Advertisement (gone after registration)

I decided to bring lunch from home today, and found myself with a few moments of spare time so I did a little experiment I've been meaning to do for a while.

 

I removed the IR filters from two lenses (both E39, one Heliopan and one Leica) and held them parallel to the floor at chest level. Playing them about I could see an astonishing difference between the brightness and clarity of my reflection as well as that of the ceiling light fixture above me. The Leica filter acted almost as a mirror whilst I could barely detect a reflection off of the Heliopan. This was the case regardless of which side of either filter faced up. As we know, the more light reflected off rather than transmitted through a glass surface, the lower the contrast and the higher the propensity for flare. My experience has been quite a number of instances of flare with the Leica filters in place whilst it hasn't presented itself with the Heliopans. I recall Erwin Puts made a somewhat similar observation in regard to the Leica filters. Now I can see exactly why that is.

 

Mind you I am not disputing the advantage of the Leica filter vis a vis the firmware cyan corner correction which has been purposely calibrated to the Leica filters (or vice-versa). But considering the cost of the Leica filters, not to mention the lenses whose extraordinary cost we accept in expectation of their superior optical properties (indeed, the main reason many people buy Leicas), I'm a tad disappointed that Leica's filters don't achieve the same transmission/anti-reflection quality as the Heliopans. I'm sure someone will offer up an excuse on Leica's behalf why that "had to be" the case, or why it is "an advantage", not to mention a torrent of "I've never had flare with my Leica filters" but I would be curious for anyone who happens to have both brands at hand to try the same observation for themselves and comment...as well, anyone who has the B+W version (which I do not).

[jlm]

my first read is that if a filter acts more like a mirror to off axis light it would reduce flare and increase contrast

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Guest tummydoc]

I'm sure you're a crack mechanic John, judging from your expert milling work, but I have to disagree with your statement. Lens/filter coatings are called anti-reflective for a reason: because the less they reflect, the more light is transmitted through to the film or sensor, heightening contrast and reducing the scatter that results in ghost images and flare. If you were to compare an un-coated piece of filter glass with single- and multi-coated glass you would notice that in that order, they will give weaker and weaker reflections off their surfaces.

[farnz]

... As we know, the more light reflected off rather than transmitted through a glass surface, the lower the contrast and the higher the propensity for flare. ...

Surely this can't be correct can it, Vinay? By "light" you must be referring to non-image forming rays for them to cause flare. So the fewer non-image forming rays that enter the lens, ie that are reflected, the lower will be the probability of flare and therefore the higher will be be the image's contrast because of the higher ratio of image-forming to non-image forming rays reaching the sensor (or film).

 

I agree with John.

 

Pete.

[dpattinson]

I don't know about flare, but it makes sense that if the inner surface of the filter is more reflective, then any light reflecting off the front element is going to be more likely to ghost back into the image.

[jlm]

we may both be right. the antreflection coatings may increase transmitivity but that is not the same as reducing non-image forming light. on the other hand, internal refections would be reduced by the AR coating and that would improve contrast.

and i have never machined a crack.

[jaapv]

Advertisement (gone after registration)

Cracked a machine maybe? Seriously, I doubt if a test of this kind has any validity to the flare rendering of a filter. I'm sure there is a more controlled way of measuring this. ( Sean??)

[george +]

Tummydoc, I did the same "test". Saw no difference. Would you explain what your point was - by a bit more measurable means please?

[Guest Walt]

By coincidcence I noticed recently that the Heliopan filters, in actual photographs, were showing significantly stronger secondary reflections of light sources within the image than either the Leica or B+W filters. So, I stopped using the Heliopans. (Incidentally I use these IR filters for BW work and consider them essential for that, although the lens coding is not.)

 

Walt

[ho_co]

Slightly off topic, but recall also that most of the current Leica filters do not cut as much IR as Leica had originally planned.

 

That is, Sean's original example from Leica was a Schneider 486, but when they began shipping, they had changed to a 'weaker' design.

 

The exception to that is in the E60 and E67 sizes, which Leica says are stronger than the others. (No indication from what I've seen whether how they compare with the other brands.)

 

That may enter into Vinay's observations. But firmware is still squishy, so we can't make firm judgments yet about effectiveness.

 

--HC

[Guest tummydoc]

Seriously, I doubt if a test of this kind has any validity to the flare rendering of a filter.

 

Let's not forget that some of the most fundamental principles of science were discovered by observation. I believe it has significant validity for the following reason: if you observe the reflection from the surface of a piece of uncoated filter glass it is extremely strong, whereas a single-coated filter has a weaker reflection and a multicoated filter the weakest, almost invisible reflection. It is an accepted fact that an uncoated filter or lens element will reflect and scatter both off and on-axis light rays causing decreased contrast and increased flare; that a single-coated filter has less of this detrimental effect, and a multicoated filter least of all. Therefore if a Leica IR filter is observed to have a reflection from its surface comparable to a single-coated filter and a Heliopan's reflection is comparable to a multicoated filter, I believe it is a valid conclusion that the same relative effects on image quality will result. And evidence from my own experience using the two types of IR filter supports that conclusion.

 

I'm sure there is a more controlled way of measuring this. ( Sean??)

 

Perhaps so, if someone has the apparatus. This is the best I could do: Both filters, photographed with a 105mm lens on a 1.5x-crop DSLR, were propped at an angle such that they were both exhibiting maximum reflection of the overhead fixture (with my eyes I could see the detail in the diffusion panel of the ceiling fixture but to focus on it meant completely blurring the filters themselves as the reflection is "virtually" more than 2 metres whilst the filters were mere centimetres away). The Leica filter is on the left. The difference is beyond startling.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[jaapv]

Yes, that is true, Vinay, but we are not dealing with antireflective coatings here. These filters are multilayered interference filters and the top layer is likely to vary from brand to brand, with totally different reflective characteristics, so mere observation of the reflection per se will not be any measure for the way the filter performs.

[Guest tummydoc]

I believe we are dealing with antireflective coatings (or certainly should be ), because both filters exhibit weaker reflections compared to plain glass. The Leica filter simply does a lesser job of it. In fact I recall someone (possibly Sean?) mentioning they were told by one of the filter manufacturers (possibly B+W?) that only one side of the IR filters can (and has) antireflective coating. FWIW these 2 filters exhibit exactly the same tendency irrespective of which side faces out.

 

I believe, until someone who designs filters for a living corrects me, that the stronger a reflection is off the outer surface of a glass, the greater number of light rays are being reflected away rather than transmitted through that glass. And until someone who designs lenses for a living corrects me, I believe less light rays transmitted through the glass equates to lower contrast. Furthermore, that the more light rays reflected off the inner surface, the greater the propensity for "veiling glare" (as Mr. Puts would say), ghost images and other unwanted reflections.

[jaapv]

As far as I know antireflective coatings would throw the interference filtering out. The B&W filter probably had the interference filter (which consists of dozens of layers) on one side, antireflective on the other. The Leica filters and presumably the Heliopan ones can be used both front and back, so may have IR filtering on both sides.

[jlm]

"I believe less light rays transmitted through the glass equates to lower contrast"

 

not so sure I agree with this

 

less light rays equate to less light which is an exposure issue, not a contrast issue. it is the ratio of shadow to highlight that contributes to contrast. in this context, any non-image spill or flare light can fill the shadows, blowing out shadow detail, and that will reduce contrast.

[ptomsu]

The Leica filters are definitely different from the others, like Heliopan or B&W.

 

If this has any effect on the results I do not know, but the convincing thing for me to go with Leica filters was that they are less bulky than the others!

 

And price difference - come on, if you are jumping into this M8 adventure, you have to know from the beginning it is going to be very expensive!

 

No money, no fun!

[masjah]

I thought the anti-reflective property referred to multiple internal reflections within the filter (or lenses as appropriate) which would then eventually find themselves transmitted through to the image plane. So I'm not sure that the external reflection at the outermost surface is indicative of anything in particular. (I might have got this wrong though - I would be delighted to be put right by someone who knows better!)

[Guest tummydoc]

"I believe less light rays transmitted through the glass equates to lower contrast"

 

not so sure I agree with this

 

less light rays equate to less light which is an exposure issue, not a contrast issue. it is the ratio of shadow to highlight that contributes to contrast. in this context, any non-image spill or flare light can fill the shadows, blowing out shadow detail, and that will reduce contrast.

 

Perhaps I mis-stated it. Reflections and scatter cause loss of contrast, and antireflective coatings reduce reflections and scatter. When the filter manufacturers describe their coatings they often point to % transmission as a measure of effectiveness, hence my confusion.

 

All I know for certain, and these are accepted facts, is that uncoated filters and single coated filters show more surface reflection than multicoated filters, and uncoated and single coated filters cause more flare/ghosts/loss of contrast than multicoated filters. Whether that is a corrolation or a coincidence seems to be the key issue under debate here, and one which no one who has posted thus far (myself included) is qualified to answer with authority.

[1234]

I get different results doing the "maximum reflection" test and have a different interpretation of the results discussed.

 

Using Leitz UV/IR cut filters and B+W 486 filters, I get essentially the same significant reflection of a light bulb (or an outdoor scene through the window) although the B+W reflection has a magenta cast to it whereas the Leica is more neutral. Not that that means anything.

 

If you take a modern Leica lens and observe the reflections from the various elements of an incandescent bulb you will see a variety, maybe 4 or 5, different pastel colors of reflections from those inner lens elements. One presumes that is not a bad thing.

 

Even if you do see reflections OUTSIDE the lens/filter system, that cannot reduce contrast because the unwanted light is outside, not inside headed toward the sensor. It could reduce brightness some insignificant amount.

 

Lastly, the job these filters is really doing lies outside the human visible spectrum so it is hard to judge the performance of filters without technical equipment. When people talk about "stronger" or "weaker" filters, what do they mean? That the width of the passband is different? That the rejection outside of the passband is different? The undefined adjective has no meaning.

 

Harry

[Advertisement]

Hallo,

Look here: Leica M

[Guest tummydoc]

Thank you for checking with a B+W filter. I have only plain UV filters in B+W marque, the original single-coated variety as well as those marked "Multi Resistant"/MRC, but the reflection from the Leica IR filter is virtually identical to that of the single-coated B+W. The MRC B+W filter has the same extremely dim, almost indecipherable reflection as the Heliopan IR filter. I know from practical use that the single-coated B+W UV filters are much more prone to induce flare and ghosts than the MRC, so from this I would expect the B+W 486 to join the Leica IR filter in being more prone than the Heliopan to cause the kind of mischief filters can cause.

 

And to re-clarify, the issue of which filter performs best to cut IR, or best complements the M8's own weak IR filter is entirely outside the intended sphere of this discussion.