Equipment

My Last Circular Polarizer Post

By
Published October 11, 2017
Testing Circular Polarizers

I posted an article about circular polarizing filters recently, letting you see some data I obtained for Lensrentals when they were reviewing which of the polarizing filters were among the best. I concluded that they all polarized light really well. I also said that none of the CPs I tested seemed to cause optical problems at 200mm and under, that some transmitted more light than others, and that you could tell that just by looking at them.

Which brings me to Roger’s First Rule of This Blog.

Rule #1: I am not responsible for what someone else says I said. If I say something wrong, I’ll correct it. If they say I said something I didn’t say to drive some click-bait, THEY said it. Not me.

A bunch of people wanted hardness tests, scratch resistance tests, light scattering tests, torque tests, and a 5-year failure rate. I want a pony to ride on my yacht. Sadly, none of the above is happening.

A whole lot of other people asked me to test something out of the price range I was examining; mainly they wanted to see how inexpensive filters did. That was reasonable, so I bought a couple of inexpensive CP filters and repeated the tests on them. You will need to look at the last article on the $100-$200 range filters to get some perspective before looking at this.

Today’s Contestants Are:

  • Tiffen 77mm Circular Polarizer      $35
  • Hoya 77mm HRT Circular Polarizer UV  $45

These are the lowest price CP filters I could find. Both are apparently uncoated, other than the CP films. The Hoya states it is a high-transparency filter, like the Marumi and B&W filters we tested previously.

Polarization

Just like the more expensive filters, these two extinguished at least 99.9% of polarized light, which is the limit of our measuring capability. (Actually, we can measure another decimal point or two, but I refuse to.)

About the Glass

Just like the more expensive filters, the two low price filters caused no distortion of our 5-micron pinhole at 200mm focal length. I have some question as to whether this test is valid for CP filters, but it’s the best I have.

Light Transmission

For our ‘better’ filters, light transmission in the nonpolarized position was 88% to 91% for the high-transmission filters; from 55% to 68% for the other filters.

  • Tiffen CP   38%
  • Hoya HRT  53%

So primarily, the low-end filters transmit significantly less light than the more expensive ones. Since they are uncoated, I would assume that difference is reflection rather than absorption.

Spectrometry

Please remember we aren’t doing ‘absolute transmission’ via spectrometry here. We’re simply looking at the relative transmission of various colors of light. The graphs are a bit noisier today because I set the software to average a smaller number of runs by accident. But it’s still sufficient to show you the curves.

Hoya HRT UV-CP

The transmission curve of the Hoya is similar to the high-transmission filters we tested last time. It is blocking some violet, as well as UV light and should have a somewhat warm appearance.

Olaf Optical Testing, 2017

Tiffen CP

The Tiffen curve is similar to the other lower transmission CP filters, but more accentuated in the red-yellow dip and probably has a cooler colored appearance.

Olaf Optica Testing, 2017

 

 

Conclusions

This will conclude my CP testing, I’ve found out what I need to know, I’ve added what seems a reasonable bit of extra information, and I’m pretty much over it now.

But there’s not much question that as far as polarizing light, cheap CP filters do it very well. They also, as you would expect from uncoated or partially coated filters, reflect a LOT more light. This is a significant issue on a clear or UV protection filter. I’m honestly not confident if it’s as big of a deal for a polarizing filter.

There could still be reflections, ghosting, and probably will be a loss of contrast. Whether somewhat is a little or a lot, I don’t know. We could calculate it for a clear filter, but I think polarization would reduce the effects somewhat. I would expect, though, that you’d be more likely to see it in the ‘nonpolarized’ position, and if you don’t need polarization, you can certainly take the filter off.

 

Roger Cicala and Aaron Closz

Lensrentals.com

September, 2017

 

Author: Roger Cicala

I’m Roger and I am the founder of Lensrentals.com. Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.

Tweet
Posted in Equipment
  • Impulse_Vigil

    To a multi million dollar contract with Amazon? 😛

  • Brandon Dube

    A 0.01 degree bending from a nominally flat surface would be an error of hundreds of waves in its fabrication. It would look like a fun house mirror to the eye.

    The focal length of the lens doesn’t have any impact on loss of image quality from the filter, only the entrance pupil diameter.

  • Thomas

    If everything is diffraction limited perhaps. I have the idea that in practive at long FL, some curvature in the filter matters more. Why: on a 400 mm lens, the angle of view is 6 degrees. On a 24 mm, it is 84 degrees. If a filter would bend part of the rays by 0.01 degree, the 400 mm image would be severely degraded (the blur being 1/600th of image size). For the 24mm lens, the blur would be 1/8400th of the image size, which might not be visible at all.

    In other words: long lenses turn a smaller difference in angle into a larger image on the sensor.

  • Devil’s Advocate

    I often use a CPL for this ‘highlight attenuation’ trick too – it will often allow an extra stop of exposure or so to bring the shadows up – very useful when shooting m43 like me.

  • Carsten

    Oops, this was supposed to be a reply to Roger’s reply with respect to variable NDs.

  • Carsten

    I would be much more interested in regular NDs, especially 6 stop and 10
    stop NDs. Besides testing for optical issues and vigneting, the most
    interesting aspect would be to assess their filter spectrograms, since
    most of them exhibit a more or less strong color cast.

  • DrJon

    Actually another interesting filter type to play with might be Variable-NDs, which are also polarising filters and have a fair number of nasties, especially in resolution at longer focal lengths, plus are essential to many video shooters…

    Sorry if you wanted to get away from filter testing, but while I might not be able to get you a Pony for your boat (as Animal protection rules get in the way) I might manage a Unicorn – being imaginary they aren’t covered by said rules – of course if it makes a load of holes in stuff that’s your problem, and they poop a lot…

  • Lee

    The CPL tests have been interesting. I did not expect to hear that the actual polarization didn’t vary very much.

  • That’s the other project we’re working on. And it’s pretty complex too, but we’re making progress.

  • Samuel H

    So you have some nice equipment for measuring color transmission in pieces of glass but you’re not making any more CPL filter articles. Lens color transmission analysis, then? Just suggesting ideas, in case you still have to justify all those new toys were a good buy. XD

  • A very good point. And I agree with your assessment of the research 🙂

  • I’m going to work on that, actually, but it’s way more complex so I’ll probably take a month or more to work on methods.

  • I don’t generally use them either but I just did find a use for them. While shooting a brightly lit and reflective stream with beautiful, darker plants on the opposite bank, there simply was too much brightness range for may D750 to handle. I was going to use a GND filter but another photographer recommended that I use a polarizing filter to subdue the reflection off the water which reduces the range from dark to light. Perfect.

  • Not THAT Ross Cameron

    At the risk of raising his ire, and bringing down the Wrath of Roger, has LR considered extending the CPL testing to variable ND filters?

  • Ruy Penalva

    Such a huge reduced transmission should be warned in the filter box. It is clear that they are not for non luminous lens and even for a luminous lens. Garbage. I had a Tiffen 77mm protector filter that makes a circular halo in my 24-105L IS Canon. Simply because full attached it touched the from elements of the lens making contact. Tiffen never more. Hoya only top with over 98% transmission.

  • Harold Mayo

    Look where that got the hosts of Top Gear?

  • Athanasius Kirchner

    Hurray for contrast-detect AF 😀

  • Arthur Meursault

    I am beginning to suspect that Roger would be fun to have some drinks with. How about a contest… ‘Drink with Roger and then juggle cinema and Otus lenses’? What could possibly go wrong?

  • Roger, you can now move to *linear* polarizing filters…

  • Brandon Dube

    In terms of aberrations from irregularity of the filter windows/plates, all you need to do is get a lens with an aperture that is as big as the filter. For a 200mm f/2.8, you get 71.5mm which is horses for courses the full aperture of a 77mm thread filter.

    At 400mm there may be greater selection for flare/stray light, and a loss of contrast as a result. A pinhole test is inherently more or less immune to stray light.

  • Wade Tregaskis

    In my anecdotal experience (i.e. sample size of one of each filter), there can be a huge difference in actual results between different filters (and vs no filter, or a plain UV filter). Most of which is actually because of systematic autofocus errors that the CPLs introduce (all do it to at least some degree in my experience, but again minimal sample sizes). e.g. I wrote about this a few years ago w.r.t. the my daily experience with my D5200: https://blog.wadetregaskis.com/image-quality-vs-circular-polarising-filters/

    It would of course be awesome if someone with ridiculous patience and free time were to explore that aspect more rigorously. Though I pity whomever takes that on, if only because of all the complaints they’ll get that they didn’t test camera model Foo and how it would obviously be better because everyone know’s manufacturer Bar has worse autofocus. 😉

  • Bob B.

    “I want a pony to ride on my yacht”.
    I don’t use polarizing filters…but thank you so much for the laugh!!!!!!
    🙂

  • Marc P.

    I’d never use such a cheap polarizer. softness and color shifts are with you….B&W, Zeiss or the Hoya HD Series are fine for my needs. And Tiffen is really crap, lowcost brand all the way, when it comes to UV Filters. But the worst ever i’ve encountered was from Green-L, uncoated glass…better no filter than these kind of things.

  • Bryn

    Is there any reason to believe that 400mm would be substantially different (non linear) than 200mm for loss of sharpness through a polarizer? I remember the old singh ray vari nd and it seemed like quality fell off dramatically after zooming past 200mm (anecdotal faded memory from years ago)

    Thanks for the data. Very helpful for making purchase decisions and choosing which polarizers are best for use when you do want a loss of light (and not)

Follow on Feedly