UV Filter or No UV Filter: Can You Tell the Difference?


There is an old saying amongst some photographers that using a UV filter will degrade the quality of your image. But is it really true? We put that to the test recently in the B&H Executive Offices. We'd like to know, in the comments below, if you can tell the difference between the two images, and tell us which one was shot with a filter and which one wasn't.

Gear Used

Olympus E-5

12-60mm F/2.8-4 SWD

B+W 72mm haze filter


Both images were shot in RAW at ISO 400, 23mm, F/3.4, and at 1/320th (which was perfectly balanced according to the E-5's meter). Both images also focused on the same point on top of the sugar dispenser's lid. Afterwards, they were put into Photoshop Elements and resized to 1MB for the web. Go right ahead and pixel-peep: let us know if you can tell the difference.

Picture 1 (Click here to view this photo larger)

Picture 2 (Click here to put your pixel peeping abilities to the test)

Again: One of these photos was shot with a UV filter and one wasn't. Can you tell which one is which? Let us know in the comments below.

The Answer:

Image one was shot without the UV Filter. Image two was shot with the UV filter attached to the lens.

Did you get the answer correct?

Discussion 38

Add new comment

Add comment Cancel

Seems as if though the top one has it on. It just seems that slight bit more defined. Very subtle

you can see the difference in the light reflection. the top image is sharper, but i'm not sure if it's due to the filter or not.

 My vote is picture #1

 Picture 1 has a filter

I'll say the first one, just because of the noise color on the bottom right of the pics ... but I can be so wrong ;-)

If you put a gun to my head (and please don't!) I'd say the first shot had the filter BUT I always have put a UV filter on my lenses primarily to protect the fornt element and secondly to (supposedly) help with haze where the eye doesn't see it but the camera does.

The top one is with the filter.  It cuts the glare a bit and darkens the darker shades because of it.

‎2nd one, just not quite as sharp - that being said there isnt any difference if i am totally honest!

Hi Chris,

I think that your comparison does not do justice. I would think that a UV filter would work best outdoors shooting into a haze (see, horizon and so on).

would love to see that comparison.\

personally, I use those as lens protectors more than as UV killers :D

On my screen, the top one has a slightly cooler / bluer tone to it, but I'm not sure if it's because of the filter. It's interesting to see a difference between the two images. Great test!

I am going against the grain here and say the UV filter was used for picture #2. The reason has to do with the fact that there is a slightly elevated bluish cast in picture #1.

I'd say it doesn't matter, because the photos were not taken in conditions where the addition of a UV filter would be most likely to affect the image negatively. Do the same test in various states of direct natural backlight, and let's view those results.

While it's obvious the subject matter of the photo is NOT what you'd use a UV filter for... the point of the exercise is if there is a noticeable difference in quality in the two images.

Personally I don't see any degradation of the image that would cause me to be concerned about using a UV filter.

I'd say number 1 has the filter based on the slightly modified lighting 

 The bottom photo has the filter. It is so obvious.

It's quite obvious from this test that UV filters don't degrade image quality,because no one seems to agree which one it is!

That's why I always have my UV (Multi coated B+W, Heliopan) on all my lens, all the time (except for when I'm using a CPL), it protects your lens and not so obvious from this test but improves quality in distant shots with strong daylight.

While I'm here, I might as well try my luck!

I really can't tell but, I say no 2.

The top one has the filter for sure

Why are folks pixel-peeping on a scene lit with artificial light? Where's the incoming UV?

A UV filter is supposed to help screen out UV light (usually part of sunlight) which allegedly interferes with exposure meters - although I suspect today's exposure meters are far less vulnerable than they used to be many years ago.

There's also the old concern about UV light causing a response on the film. But today's electronic sensors already have band-pass filters built right into the sensors array. This is in a Bayer pattern for most of today's sensors.

So I claim that there is and there should be No Meaningful Difference between the two shots.

Sorry to disappoint, but the physics just are not there. On two levels.

Oh - and as far as optics degradation due to using a filter:

I don't think you'll see one here. B+W makes some very good glass.

Try a filter that costs 2 bucks and comes in a cereal box ******* in cardboard. You might see something then.

One last comment - and that has to do with good filters being able to control internal reflections, and thus unwanted lens flare. Most good multi-coated filters do this very well.

The test shot above does not appear to provide an opportunity for observing flare, which is unfortunate.

A quick google search "what does multicoating do for filters" yields some good examples. I'd link here, except the spam filters prohobit that sort of stuff.

I am going to say that the #2 photo has the filter. To my eye (and on my monitor) it looks to have a pink cast in comparison to the #1 image which appears more blue, especially in the shadow cast to the bottom left of the object. In terms of electromagnetic frequency response, perhaps the UV attenuation rolls off gradually and bleeds into the visible blue light wavelengths a little bit....? Then again my eyes could be completely out of calibration ;)

Final answer ;)

I too say it was the second one, based solely on a slightly warmer cast to the color of background surface.

Well, I decided to take the two images and difference them in Photoshop. (Take both images, put each on a layer, then use the subtract filter.)

No flare that I could see.

No color difference eiher. JPEG noise swamped most everything.

There is a slight difference in rim lighting, which I think has to do with a light being moved slightly or being blocked by the camera operator. Look at the highlights on the near rim of the canister and you'll see the most pronounced difference between the  two images.

Thanks Chris for a great post. Now that we have the long-awaited answer, I think it's prudent to answer the first question (oddly, some refused to answer the "which one has a filter?" question even after several interesting but not pertinent comments): "There is an old saying amongst some photographers that using a UV filter will degrade the quality of your image. Is it really true?" The answer is no.

I always have a UV filter attached to my lenses, whether shooting indoors or out. That said, from practical experience, you don't always need one. One of my favorite photographs was taken on a sunlit beach without a UV filter on the lens (I cracked it just prior to the shoot while changing lenses). However, to refrain from using one due to a perceived degradation in image quality is clearly a baseless concern.

On another note, for such a nice lens, there sure is a lot of light falloff at the edges of the frame (which, because it is evident in both photos, has nothing to do with the filter)!

-Matthew Scampoli

I put filters on all my lenses. Ever been on a shoot where some jerk decides its funny to put fingerprints on the lens? Yeah, raise your hands...

Realisticaly you can't tell it unless its outdoors in some conditions. And the line about air and glass interface is pretty much people taking a physics course and being dangerous.

 The lid in the top image is much darker, i prefer it! 

This  "test" proves very little. The real challenge for filters is when AF is critical, or when dealing with flare. Expensive, multi-coated filters copes a bit better than the cheapos, but can also have flare issues and cause strange AF behavior with some lenses.

Here's a link to a test I did a couple of years ago, to convince myself to loose the filters:


A dark object on a dark background does not tell the full story.  Lets see an object against a white background. I would suspect if a UV filter would degrade the quality of the image it would be causing CA or color fringing on the edges because of the extra piece of glass.

On my screen, there was no difference between the two images, as long as I made sure to view them in the same spot.  In other words, any difference was caused by my eyes being slightly off axis.

As was said, this could be very different with a cheap filter. 

No difference! 

To be quite honest, I see no difference between the two images.

 Poor example photos to use in this debate.  

I stopped using UV filters as they're usually just a ploy from camera shops to sell "add-ons" (just look at what site this is posted on).  Modern digital camera sensors already filter out UV light so basically all you're doing is adding a piece of glass to the front of your lens which can lead to auto-focus searching, loss of contrast and lots of extra light bouncing around.  

I find a lens hood perfectly adequate protection for normal shooting conditions.  Now if you're shooting in blowing sand or sea spray then by all means use one if you feel you can't keep your lens from otherwise getting damaged.  The fact though is the outer element of your lens is the least expensive part to replace on most lenses, should it ever need to be replaced. 

Photography for me is first and foremost about getting the shot and getting it at it's highest quality.  Always using a UV filter has hindered that in the past, especially in lower light and bright daylight so I'll never use them again. 

Of course this all depends on your camera body and lens as to what difference you may or may not notice in image quality.  I actually saw a loss of color and detail when shooting daylight landscapes with a Canon 5D, 70-200 f/2.8L and a B+W Multi-Coated UV filter.  Everyone's results may vary but for me, I'll just use the money I used to waste on these filters to buy something I actually need. 

 P.S. Canon replaced the outer element on a Canon EF 17-40mm f/4L that I was given used with a large scratch on it for twenty more US dollars than a high quality B+W UV filter at 77mm would have cost.  True I was without the lens for only 14 days but people acting like they're going to be out a large sum of money or have to replace the entire lens because of a damaged front element have been sadly misinformed.  

I never buy a second hand lens that wasn't used with an UV filter most of the time. Often the front element looks horrible with permanent fingerprints, dust and dirt on the edge.

A UV filter can cause flare and veiling and that's just very ugly. Good UV filters flare less, but are expensive. Why pay so much more for your lens? Modern lenses have good scratch resistant coatings and pass very little UV light. Some lenses even have a recessed front element. Unless you are on a sandy beach, near volcanic ash or on a mountain top, UV light is not a problem most of the time.

same, and subtle difference

I got it. I noticed the glass dispenser comes out of the background slightly with the UV filter. Most of my shooting is of cars for a dealer's website, and while it's easy to beat almost everyone else for quality and pop (because almost no one else in that field has any idea what trhey're doing) I'm not above trying easy other things to improve here and there even more. Shooting interiors with flares and shadows and focus issues is tough enough at times, but I'd like to prevent issues as automatically as possible, and customers want to see details with crystal clarity. It looks worth it to at least give one a try. Depending on the subject and lighting, improvement may be more obvious.