A good polarizing filter is probably the single most useful filter for color photgraphy, and is still useful in black and white work. If used properly, it can reduce reflections from shiny surfaces, heighten sky contrast, improve foliage contrast by removing flare and double as a 2x ND filter for some situations (and two stacked linear polarizers can also make an infinitely variable ND filter).

Polarizers come mostly in two types - linear and circular polarizers.

Most normal light has a mixture of polarizations that is mostly pretty random.

Linear polarizers selectively remove that part of the light that is not polarized in the same plane as the filter. What remains and passes through to the camera lens is mostly polarized in the same plane as the filter is oriented. (that is very similar to an red filter for B/W that removes mostly green light, passing primarily red light on to the lens).

Circular polarizers are really two filters stacked one behind the other. The filter furthest from the lens is a regular linear polarizer. Immediately behind that (in the same mount) is a "quarter wave plate" oriented at 45 degrees from the polarizer. That converts the "linear polarized" light coming back from the front polarizer to "circularly polarized" light that contains all polarizations again. Just how it does it is complicated, but I can post a more in-depth article if interested.

Reflected light is partially polarized in the process of being reflected, whether it is from an object in the image that you are trying to photograph, or reflected from the beam-splitting mirror in your DSLR that allows autofocus/metering. That means, in theory at least, that feeding the camera with linearly polarized light might interfere with light metering and/or autofocus, while feeding it with circular polarized light should not. Because of this, the normal dogma is that you need a circular polarizer on a DSLR to allow autofocus and metering to work properly. In actuality here, I have found little or no problem with linear polarizers (which are often better, cheaper and less prone to other artifacts including color casts). That has been true with istD, istDS, K100D and my current K20D.

The simplest test for whether a polarizing filter is really and truly a circular polarizer is very easy. Look through it into a mirror. This is a Hoya PL-CR filter that is first held in the normal orientation (male threads toward the camera lens or my eye). Light is reflected off the mirror and passes back through. There is some darkening (about 1-2 stops with each pass through the filter) but the camera is clearly visible in the reflection. There is very little color cast visible.



Now reverse the filter. This is the same Hoya PL-CR filter with the male threads pointed to the mirror, away from the camera. Now the mirror image of the filter is very dark, making it harder to read the logo and much harder to read the lens lettering..



(Just why that happens is fairly complicated. Again, if there is interest I can post a more in-depth explanation).

If you try the same test with a good linear polarizer, there is absolutely no difference on reversing it - the image looks identical, and with minimal color cast.


Now look at a cheap Visico CPL circular polarizer bought off ebay. This is in normal orientation, male threads to camera lens.



And this is the Visico CPL reversed, female thread to the lens.



The first thing that is obvious is the lack of darkening when the filter is reversed, but there is also a definite yellow color cast in both orientations. This "circular polarizer" is only circular in shape, not in type!

This holds true for other filters I have acquired over time, some with lenses, some on their own. The Hoya 67 and 77 mm PL-CR filters are true circular polarizers with minimal color cast. The Hoya 49 and 52 mm PL are true linear polarizers, again with very little color cast. In contrast, each of the cheap Asian filters from Visico or no-name from Korea fail this simple test, both obviously as true circular polarizers and also by introducing unwanted color casts.

Try it with your own filters and see which others pass the test!

[Iksobarg]

Using a hoya cpl I was getting a horribly blurred image at 320mm on FA80-320 while shooting my family at the beach on a sunny day. I was using a good hood too.
At ~140mm or less it wasn't there, the blurring.
Plus, I noticed the cpl changes from dark to light when rotated.
Please explain.

---

Last edited by Iksobarg; 08-03-2013 at 06:54 AM.

[Kokosh_Anelle]

Using a hoya cpl I was getting a horribly blurred image at 320mm on FA80-320 while shooting my family at the beach on a sunny day. I was using a good hood too.
At ~140mm or less it wasn't there, the blurring.
Plus, I noticed the cpl changes from dark to light when rotated.
Please explain.

I was just looking for some information about if there is any sense in buying a CPL for Tamron 18-250.
Got once a cheap one on eBay, the results were very bad but now I'm thinking about investing in some proper one, but still am not sure if it will work.

I would really appreciate any info.

[GR Jim]

I've been researching UV and CPL filters for a while now and K&F Concept filters have caught my eye. Does anyone have any experience with them? I'm particularly interested in the effect the green glass coatings have on the final image.

[pacerr]

but still am not sure if it will work.

One thing to keep in mind when using any PL filter; the polarizing effect varies with the angle relative to the sun/light source(s). Using a relatively wide angle lens perpendicular to the direction to the primary light will produce variable polarizing effects across the broad image.


Some folks bad-mouth PL's simply because they use them in adverse conditions. A little thoughtful experimentation with that filter will produce better, and more consistent, results.

Used properly, an economical non-CPL filter will often produce satisfactory results outdoors.
Having said that, a PL filter is one of the few effects you can't reasonably duplicate in post-processing. (The other is a neutral density filter's effect on shutter speed versus motion.)

[awscreo]

Got me a Sirui Nano CPL and ND filters. Optically they're pretty good imo. Not much worse than far more expensive Marumi ones.

[Wheatfield]

One thing to keep in mind when using any PL filter; the polarizing effect varies with the angle relative to the sun/light source(s). Using a relatively wide angle lens perpendicular to the direction to the primary light will produce variable polarizing effects across the broad image.


Some folks bad-mouth PL's simply because they use them in adverse conditions. A little thoughtful experimentation with that filter will produce better, and more consistent, results.

Used properly, an economical non-CPL filter will often produce satisfactory results outdoors.
Having said that, a PL filter is one of the few effects you can't reasonably duplicate in post-processing. (The other is a neutral density filter's effect on shutter speed versus motion.)

A linear polarizer has a tendency to blind the AF sensor via cross polarization at the half mirror, and can also knock exposure out of whack in cameras that have the metering sensor in the mirror box.
To the best of my knowledge, all Pentax cameras except the LX film camera meter off the focusing screen when using the OVF, so exposure accuracy shouldn't be an issue.
If you don't mind manual focusing, a good quality linear polarizer is fine, but I don't think anyone makes them any more, at least not at a price point that makes them more economical than a good circular polarizer.
OTOH, the linear polarizing filters made by B+W (for example) prior to the advent of autofocus and the resultant need for circular polarizers are excellent quality filters, far better than the cheaper circular polarizers being hawked these days.
A really good circular polarizer costs real money, and probably isn't worth it if one is using lower end consumer zooms or the knock off prime lenses coming out of China or Korea.
OTOH, if one has just spent the better part of $3k on a D FA* 85/1.4 lens, they would be introducing a very false economy by purchasing a cheap filter to put on the front of it.
A cheap filter looks OK on a cheap lens because it isn't necessarily the weakest link, a cheap filter on an expensive lens will tend to bring the lens down to it's level.

[pacerr]

...a good quality linear polarizer is fine, but...

A good summary and highlights the economy of high grade filters. A well chosen and intelligently used economy filter can make a notable difference in results and certainly beats not having the too-expensive, gold-plated model in hand.

Three investments one can make in filters (and all other gear) that are free of monetary cost.

1) Research the materials available so as to understand the inherent capabilities and limitations of the device -- RTFM. [ Such as 'The Photographer's Guide to Using Filters', Joseph Meehan. Amphoto Books ]

2) Spend a pleasant hour or more of well informed and dedicated experimentation with a filter involving its anticipated use. Instant review, histograms and comparison frames are wonderful learning tools.

3) Slow down. Spending an extra ten seconds specifically planning intended results and verifying the camera settings before shooting avoids most disappointments when working within the capabilities of one's gear. (That 6-P's thing.)

I don't recall ever snap-shooting something while using special filters or having a linear PL filter screw up focus or exposure while shooting mindfully with gear I understood.


To borrow sage aviation advice: Never let your camera go somewhere your mind hasn't already been. (Works for life too.)

---

Last edited by pacerr; 09-15-2020 at 02:23 PM.

[GR Jim]

I've been researching UV and CPL filters for a while now and K&F Concept filters have caught my eye. Does anyone have any experience with them? I'm particularly interested in the effect the green glass coatings have on the final image.

OK. So after some more research, I found that K&F Concept is a Chinese company and I suspect that the low prices will probably reflect low image quality, so I'm taking them out of the running. The Kenko Puro line is now my top choice for good quality filters.

[artrasa]

I've used a Hoya linear polarizer filter with a K10D camera. Never really noticed any problems regarding exposure or (auto)focusing.

[Wheatfield]

I've used a Hoya linear polarizer filter with a K10D camera. Never really noticed any problems regarding exposure or (auto)focusing.

Exactly. The only time you should find any problem with a linear polarizer on a Pentax camera (other than the LX film camera) is if you manage to cross polarize the AF sensor into blindness, in which case your AF will start to hunt, and may refuse to lock focus simply because it can no longer see.

[tibbitts]

OK. So after some more research, I found that K&F Concept is a Chinese company and I suspect that the low prices will probably reflect low image quality, so I'm taking them out of the running. The Kenko Puro line is now my top choice for good quality filters.

I have never tried a K&F filter, but I wouldn't rule out a filter due to where it was assembled. It wasn't that long ago that people wouldn't buy products from Japan or Germany, after all. You might at least test one, especially if you buy where you can get free (even with shipping provided) returns. I have had one filter, a Vivitar, that clearly affected image sharpness, and a few that don't handle reflections very well, but overall filters probably aren't the most difficult products to test for youself.

[Wheatfield]

I have never tried a K&F filter, but I wouldn't rule out a filter due to where it was assembled. It wasn't that long ago that people wouldn't buy products from Japan or Germany, after all. You might at least test one, especially if you buy where you can get free (even with shipping provided) returns. I have had one filter, a Vivitar, that clearly affected image sharpness, and a few that don't handle reflections very well, but overall filters probably aren't the most difficult products to test for youself.

Yup. Take a picture with the filter off, then take the same picture with the filter on. I do several pictures of each parameter and pick the best one for comparison.
Make sure to refocus in case the filter affects it.

[Serkevan]

I have a K&F reverse graduated filter (100x150mm) and it works pretty well - for single filter use I haven't seen any color cast or sharpness loss. It was cheaper than, say, Rollei or other staple names, but I still paid 60€ for it on sale (regular price was 90€ IIRC).

The problem with China-made stuff has more to do with the fact that a lot of factories do make very low end cheapo things, and that of course makes it difficult to get trust. There are good Chinese products, they just won't be super budget friendly... my coffee grinder is designed and made in Taiwan and it's fantastic - extremely well built, performance is great and it is a bit cheaper than the competition. But it still was 150€ instead of, say, 200. Laowa lenses are also Chinese and they seem to be very good, but they aren't exactly super cheap either.

[gofour3]

Exactly. The only time you should find any problem with a linear polarizer on a Pentax camera (other than the LX film camera) is if you manage to cross polarize the AF sensor into blindness, in which case your AF will start to hunt, and may refuse to lock focus simply because it can no longer see.

If you shoot the LX with the shutter dial in "AUTOMATIC" you can use either a LP or CPL, as the offending mirror flips up when the LX meters off the film plane. If you shoot the LX in "manual" then yes you can only use a CPL, as the offending mirror will mess up the metering.

Phil.

[Wheatfield]

If you shoot the LX with the shutter dial in "AUTOMATIC" you can use either a LP or CPL, as the offending mirror flips up when the LX meters off the film plane. If you shoot the LX in "manual" then yes you can only use a CPL, as the offending mirror will mess up the metering.

Phil.

That's only true for automatic shutter speeds longer thsn x-sync. At sync speed and above, the camera is setting its exposure via the reading it takes just prior to pushing the button.