UV or Polarizing Filters

A

atomcrusher

Expert Member
Joined
Jun 27, 2006
Messages
4,208
Reaction score
5
Location
Western Cape
I have a Canon 550D DSLR, and yesterday while taking some photos of the snow on the Langeberg mountains, I noticed that the photos appeared 'washed out'.

The green canola fields appeared greyish, instead of the vivid green they actually were, and the snow & distant mountains seemed faded.

I've been meaning to get 58mm filters for both my Canon 'kit' lenses, a 18 - 50mm, and a 55 - 250mm, but I would like advice of if I should get UV, or Polarizing, filters for the lenses.

I guess the prime function would be lens protection, and I assume the plain UV filters would be OK, but reading websites like dpreview.com it seems there are pros & cons to both UV and Polarizing filters.

So, what's the advice on what filters I should buy? A local seller, Camera Warehouse ( http://www.camerawarehouse.co.za/product_info.php?cPath=23_34_70&products_id=276 ) has the following:

(Camerawarehouse.co.za ) offers 2 x Kenko 58mm UV filters, as below:

Kenko 58mm UV Filter
http://www.camerawarehouse.co.za/product_info.php?cPath=23_34_70&products_id=264

@ R119.00

and ...
Kenko 58mm PRO UV Filter
http://www.camerawarehouse.co.za/product_info.php?cPath=23_34_70&products_id=276

@ R 349.00

These are both UV filters, but why the big difference in price?

Which should I buy :confused: Money is tight in retirement!

I'd appreciate some advice please? I would have to buy online, as my Dorp has no camera dealer where I could buy them
 
Sorry, don't know much about filters, but here is what seems like an informative link -

http://www.mat.uc.pt/~rps/photos/filters_uv_pol/

I never use UV filters, sometimes a CPL depending on shooting conditions. No idea how they impact snow shooting.

PS is your WB and exposure correct? Easy to under-expose when shooting snow.

PSS you probably have to over expose by 1.5 to 2 stop to get the snow proper white
 
I dont think there is a one filter fits all scenario.
My wife uses a circular polarizing filter, plain UV, neutral density and neutral density graduated filters.
Each one has a specific function but the polarizing filter is normally used around water depending on the effect required.
 
undesign said:
PS is your WB and exposure correct? Easy to under-expose when shooting snow.

PSS you probably have to over expose by 1.5 to 2 stop to get the snow proper white
Click to expand...

Thanks for the link, and the WB heads-up. I was using the 550D in full-auto mode so maybe that's the problem
EDIT: that link you provided re: UV & Polarising filters is great .. detailed, but easy to understand for a novice such as me.
 
Last edited:
I don't advocate using UV filters for any reason (there's certainly no need to filter UV light on dSLRs because they're already filtered) apart from the most adverse weather conditions - I'm thinking sandstorm. An an additional - and unnecessary - piece of glass can cause a host of problems. CPLs and ND filters on the other hand can be very useful.

When you're shooting in the snow you need to bump up you EV.
 
http://www.lenstip.com/115.4-article-Polarizing_filters_test_Results_and_summary.html
http://www.lenstip.com/119.1-article-Polarizing_filters_test_-_supplement.html

http://www.lenstip.com/113.1-article-UV_filters_test_Introduction.html
http://www.lenstip.com/120.1-article-UV_filters_test_-_supplement.html

UV filters serve some practical purposes. The filtration of UV light is the least of these, as DSLRs' sensors are not overly sensitive to UV light (rather than being filtered against it). They can still pick up a bit of UV, but not enough that a UV filter will have a remarkable impact except under extreme scenarios where you're shooting landscapes with an insanely long viewing distance and a lot of haze between your position and backdrop items such as mountains.

Their more common purpose is to complete weather sealing on lenses that are weather sealed, which yours aren't. Even if you're getting one for such a purpose, some filters will still leak water between the rings and glass in the filter, so testing the filter and/or sealing it with some silicone mirror glue or the like is necessary anyway The reason there's such a difference in price between those two Kenko UV filters is largely because the cheaper one has only one anti-reflective/glare coating while the latter has multiple. More and better coatings mean less chance of filter-induced flare, ghosting, internal reflections and coma.

As for the grayish appearance in the canola, chances are good you were dealing with glare on the plants as well as snow - that's where a CPL would come in handy. I'm too tired to dig into my photos right now, but I have some examples here somewhere of canola here in the Overberg taken with and without a CPL (or I should still have).

It's very important to note that a CPL is a polarizer, so it tends to only cancel out light with a particular polarization. Because of this, trying to use it to get rid of glare from light coming from a diffused light source, such as that shining through a cloud, won't work nearly as well as using it to cancel out glare coming from a direct light source, ie clear sunlight. It's also important to keep in mind that the polarization is strongest when the CPL is appropriately oriented along with the film plane to be at a 90 degree angle to the sun or other light source. To determine how that would work, you can do this; stand outside and point at something with your index finger, your thumb pointing up. If it were mid-day and the sun were directly above your thumb, you'd have the greatest polarization effect possible. By twisting your arm so that your index finger keeps pointing where you'd shoot but your thumb rotates around the provided axis, you can figure out where the sun could otherwise be located for the CPL to still have a noticeable effect.

A CPL won't do much to snow in general, being that snow will be scattering the light throughout and thus causing a change in orientation of the light's polarization much like clouds would, but it will still help to reduce glare on anything not covered in snow and help to darken the sky - this would apply primarily to if you were actually in a snow-heavy area as opposed to trying to take a photo including a strip of snow somewhere.

Anyway, here's an image that provides an example of the effect a polarizer has on foliage, the grayish trees being the ones not covered by the polarizer's effect.
 
nanonyous said:
Their more common purpose is to complete weather sealing on lenses that are weather sealed, which yours aren't. Even if you're getting one for such a purpose, some filters will still leak water between the rings and glass in the filter, so testing the filter and/or sealing it with some silicone mirror glue or the like is necessary anyway The reason there's such a difference in price between those two Kenko UV filters is largely because the cheaper one has only one anti-reflective/glare coating while the latter has multiple. More and better coatings mean less chance of filter-induced flare, ghosting, internal reflections and coma.
Click to expand...
More accurately that should read some weather sealed L lenses. Some being the 16-35mm's, the 17-40mm, and iirc the 50mm f/1.2. The 16-35s and 17-40 have that front element that slides inside the barrel of the lens.
 
It's not limited to those, as there are many models which have skims in which the front element is screwed in underneath the taped-on 'ring' around the front that contains the lens information.

http://www.youtube.com/watch?v=G9y-Bt-gv90 <- in this video at 2:10 the guy removes the ring in question.

It mightn't seem like it, but the double-sided tape on those rings can at times come loose at any given point for no good reason and create small gaps behind which water/dust could get in. It's one of the points of entry for dust into the 24-105mm lens, among other models. As far as I'm aware, Canon's supertelephoto lenses don't have this issue as the skims are taped underneath a shell which is attached over the top, that has a rubber gasket to prevent moisture/dust from getting in there.

http://www.youtube.com/watch?v=ovxtgj4SsiI&feature=player_detailpage&t=547

Page with pictures of the EF-S 17-55mm f/2.8

Obviously not an L lens and so not even advertised as being weather-sealed/resistant, but it shows what's essentially going on under the ring on lenses like the 24-105mm as well.
 
Last edited:
nanonyous said:
It's not limited to those, as there are many models which have skims in which the front element is screwed in underneath the taped-on 'ring' around the front that contains the lens information.

http://www.youtube.com/watch?v=G9y-Bt-gv90 <- in this video at 2:10 the guy removes the ring in question.

It mightn't seem like it, but the double-sided tape on those rings can at times come loose at any given point for no good reason and create small gaps behind which water/dust could get in. It's one of the points of entry for dust into the 24-105mm lens, among other models. As far as I'm aware, Canon's supertelephoto lenses don't have this issue as the skims are taped underneath a shell which is attached over the top, that has a rubber gasket to prevent moisture/dust from getting in there.

http://www.youtube.com/watch?v=ovxtgj4SsiI&feature=player_detailpage&t=547

Page with pictures of the EF-S 17-55mm f/2.8

Obviously not an L lens and so not even advertised as being weather-sealed/resistant, but it shows what's essentially going on under the ring on lenses like the 24-105mm as well.
Click to expand...
It's only those four weather-sealed L series lenses that canon recommends in the manual the use of a Canon "Protect" filter in order "to ensure dust- and water resistant performance"
 
They may only recommend it for those lenses, but given I know there's a reasonable chance water could penetrate at the front of the lens, especially if using a lens hood and pointing the lens 'up' enough that significant amounts of water may suddenly enter the hood, I'd personally rather use a filter to ensure the water can't get in at the front of the lens and has to instead drain out at the sides of the hood.
 
atomcrusher said:
I have a Canon 550D DSLR, and yesterday while taking some photos of the snow on the Langeberg mountains, I noticed that the photos appeared 'washed out'.
Click to expand...

An effective way of dealing with this would be to use HDR. Generally, the "washed out" look is caused by saturation of the photosites, which is indicative of a scene that has a larger dynamic range than what you can capture in a single shot. Shooting in raw helps to squeeze out a bit more dynamic range. Some cameras, e.g., Nikon D5100, or D7000, will allow you to squeeze out at least three additional stops of dynamic in a single shot. Even with those cameras, it does not hurt to shoot a few "underexposed" shots while you are there (a tripod will help).

You can indeed use HDR without ending up with an artificial-looking photo --- you just have to learn how to use your particular HDR software. Everyone has their favourite, but I quite like "Luminance HDR" --- it is free :)

A polariser will produce a different effect, but since HDR is gratis .... why not give it a try?
 
Thanks to all above for the detailed and informative posts. I've taken in the hints & tips, and I've learned a lot (even if I don't understand some of it :o)

Here's a (resized much smaller from original) image of one photo I took, which led to my opening questions

IMG_2694.resized.jpg
 
Last edited:
Here's one of my first attempts at photo-stitching to create a panoramic image .. this was 4 x images stitched together using the Microsoft ICE application.

resized.and.cropped.jpg
 
atomcrusher, First thing I noticed is that the contrast is very low on this image. The camera has obviously taken a conservative approach to the exposure to maintain the dynamic range between the bright snow and relatively bright hills. Best thing might be to play with tone curve in your editing software. A subtle s-shaped curve will boost the contrast, and put more emphasis on the dark/midtone region of the curve and not the highlight area.
 
An S curve actually tends to increases the highlights.

The 'gray' on the horizon is haze, which a polarizer would cancel out to varying degrees. One can mess with curves to try and lessen its impact, but what you're ultimately doing is nothing more than trying to extract dynamic range from an area of the image that simply lacks it.

Images, in order: raw as LR interprets it, before adjustments, changed to camera neutral calibration to de-compress the tonal ranges, input values compressed and an S-curve applied. Notice how, in the last version, the dark areas become significantly darker and the bright areas significantly brighter. Also note the mountain in the background. While it may appear as though more visual detail (or contrast) is being gained in the mountain, it doesn't technically do so. All that's happening is your brain is being tricked by the harsher transition from light to dark, which gets interpreted as detail.

600px-8784-WM.jpg 600px-8784-WM-2.jpg 600px-8784-WM-3.jpg
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X