UV Filter: effect on image quality

[Alis]

I use a UV filter on all my lenses for protection, as I assume many of you do. The only time I take them off is when I want to use another filter. Assuming that the quality of the lens is good, is there any effect on the quality of the image? Is there any special circumstances that you would take the UV filter off temporarily?

For polarizer filter I know that only a circular type filter will not interfere with the autofocus. Is there a type that interferes with the focusing mechanisms of the camera/lens in case of UV filters?

[Colin Southern]

I use a UV filter on all my lenses for protection, as I assume many of you do. The only time I take them off is when I want to use another filter. Assuming that the quality of the lens is good, is there any effect on the quality of the image? Is there any special circumstances that you would take the UV filter off temporarily?

I'm assuming that you mean "assuming that the quality of the FILTER (not lens) is good"?

Lowering image quality ("IQ") is one of the old wives tales that gets dragged out of the closet everytime there's a discussion of the topic. All I'm going to say (at this stage) is that - with the exception of extreme contrast situations (eg shooting bright lights at night) any "degradation" of image quality when using a high-quality filter will be undetectable (in much the same way as dropping a pebble into the ocean will cause an undetectable rise in the water level; it's something that may happen in theory, but in practice it's just not an issue). Like you, I leave mine on all the time - unless I have a reason to take it off. And I believe it's saved a lens from damage on more than 1 occasion.

For polarizer filter I know that only a circular type filter will not interfere with the autofocus. Is there a type that interferes with the focusing mechanisms of the camera/lens in case of UV filters?

Not that I'm aware of.

[Henry Peach]

I agree that the flaring and image quality problems with filters tend to be exaggerated in these sorts of conversations, but it does happen. I would say it depends a lot on what you shoot and your shooting style as to whether it's a problem worth worrying about or not (and your clumsiness). I like to shoot back lit subject and into light sources for some of my stuff, and it's usually pretty easy to see the difference between filter on and filter off in these sorts of situations. My biggest issue is that it's often subtle enough that I don't see it through the viewfinder or on the LCD; I notice it when I'm home working in the darkroom or on the computer.

I stopped using UV filters on my lenses over 15 years ago. At first I was nervous as I had been assured (by the UV filter salesman) that leaving my lenses vulnerable would be the end of them. All this time later I've yet to damage a single lens in a way in which a filter would've helped. I'm not doing anything I would consider extreme with my cameras, but they do get hauled all over, banged about, and used every single day without much regard to their well being.

If I were going into a situation where I thought my lens needed protection I'd put one on in a second. But my standard operating procedure is to skip the protective filter, and count on the lens hood for protection. It's worked so far.

Here is an article at SmartShooter.com demonstrating the possible effects (at least it happened to their cameras) of filter on/off shooting subjects against a fairly evenly lit, white background. You have to register to get access, but it's free, and they won't spam you.

http://www.shootsmarter.com/index.ph...id=167&acat=16

One thing they found was that while center image quality wasn't degraded much by a filter, their edge image quality suffered much more. Their test should be easy to try on your own to see if it's a problem with your gear.

[David]

Hi Sedali

This is an issue that I have looked at recently by taking some test shots with and without a skylight filter. Looking at the resultant histograms, I could see small changes in distribution of pixels and, with the eye of faith, there might have been changes in the actual images. However, I would not say that the image had deteriorated in any way. However, the shots I took were in benign conditions and there may be greater effects with more extreme lighting.

Cheers


David

[McQ]

I would challenge anyone to notice any difference between a new, multicoated UV filter and a lens with no filter. That being said, from a strictly technical perspective, a UV filter certainly cannot increase image quality, since it introduces an additional layer of glass between your camera's sensor and the subject.

However, to me it comes down to this: is any potential reduction in image quality reason alone to forgo the protection it can provide, and the potential long-term increase in image quality as a result?. Depending on the shooting conditions, I often find situations where I'm repeatedly wiping the front of my lens with a cleaning cloth. Over time this can leave micro-abrasions on the glass, especially if the cleaning cloth has to be used in the field, and is not perfectly dust/particle free. One can then just periodically switch out the UV filter for a new one, leaving nothing but pristine glass between your sensor and subject. Further, if there's no UV filter in place, I often find that I am much more reluctant to wipe/clean the front lens element in the field. In a sense, this means that I am potentially reducing my image quality by NOT using the protective filter.

I've also personally had two times where my tripod tipped over with the camera on it, resulting in a cracked/scratched/shattered UV filter. Everything else was fine. Just imagine if that was the front lens element...!

[Colin Southern]

I would challenge anyone to notice any difference between a new, multicoated UV filter and a lens with no filter. That being said, from a strictly technical perspective, a UV filter certainly cannot increase image quality, since it introduces an additional layer of glass between your camera's sensor and the subject.

Since it's a usually a UV filter that used - and although most UV light doesn't make it as far as the sensor anyway, I've wondered if the further removal of any remaining UV component by the filter, even though it's an additional layer of glass, could actually produce a theoretical improvement in IQ?

[McQ]

One could certainly purchase a "clear filter" to avoid any UV filtration, but I am not aware of any multicoated clear filters. Particularly for night photography (with many concentrated light sources), multicoating is essential to prevent internal reflection off the surface of the filter. My understanding is that there's no additional glass layer when comparing a UV filter to a regular clear filter, and that the UV filtration comes from a glass coating. However, one could argue that a glass coating is still effectively another layer of material...

[Davey]

Something colin said made me realise something. Most UV doesn't pass through most glass well (hence UV lab gear tends to have fused quartz windows) and it's irrespective of coating so if it doesn't reach the sensor it shouldn't have any effect. I guess some might make it through but the amount in most scenes I'd imagine to be insignificant enough to ignore it.

I don't have experience of UV filters so can't say for sure, also I'm not sure of the composition of lens optics so maybe they let more uv through than your average glass does but most recipes don't let that much through at all. Less common additives can completely eliminate it so i image the coating on UV filters you mention is similar compound to the glass additive. Also I doubt silicon sensors are sensitive to UV closer to 380nm end of spectrum that will get through the glass easier.

I'd say if there is no physical irregularities that will interfer, and effects such as reflection issues and so on are negated by the coatings you shouldn't have any difference. In theory some filters might do more than filter UV and cut some violet. If they do then there will be a visible difference in some scenes related to filtering of visible light as well as UV. However if the filter is a true UV filter (ie. no visible wavelengths cut) in theory I'd say it wont affect the image at all (providing as already mentioned by various people that there is no none UV optical interference from scratches, opaque filter material etc).

I'd guess it only affected film due to it's high sensitivity to UV so small amounts creeping through the glass would affect the film, correct me if I'm wrong but nothing (what I know if sensor technology which is not much) about digital tech should be significantly affected by UV. The very most that could happen since it's not a chemical process is close UV wavelengths could be picked up as violet so the blue channel in the captured data would be ever so slightly higher (insignificantly so).

Also perhaps some filters cut out a tighter band so only affect UV where as others cut out wider range so affect the visible wavelengths around there (especially violets) and hence image quality so people percieve visible difference. This is just guessing but might be helpful. If it is indeed the case finding out what filters cut what and by how much would be useful from protective measure choice POV. Another thought is the more the filter is treated to filter UV (ie higher the amount it cuts out) the more likely it is to affect closer visible wavelengths. If anyone can correct me on any of this please do as it's a long long time since I studied physics and it's definately not my area then. However most of it is basic physics and common sense though so imagine it stands. Hope it helps. On an odd off topic note I'm now wondering whether my mouse would still work through a uv filter (if only I had one) since it's 320nm laser (I think maybe there is a malfunction in my mind since I often wonder about such stupid stuff, time for a brain declutter I think).

[Alis]

Again a great discussion. Thanks everyone. I am convinced that the effect on IQ is close to zero if not zero.

It was one of the things on my checklist of things that I can do to improve image quality/ sharpness. So, one more down...

[Radu Dinu Cordeanu]

I use a UV filter on all my lenses for protection, as I assume many of you do. The only time I take them off is when I want to use another filter. Assuming that the quality of the lens is good, is there any effect on the quality of the image? Is there any special circumstances that you would take the UV filter off temporarily?

For polarizer filter I know that only a circular type filter will not interfere with the autofocus. Is there a type that interferes with the focusing mechanisms of the camera/lens in case of UV filters?

Hi,in this afternoonI did two pitures for you,with a kind of pola:varicolor blue-yellow.I try to send you.One is with maximum blue effect and the other w,yellow maximum effectThere are many interpositions.It is a very subtle filter indeed.

[bockey]

Aside from any filter on your lens: the sensors on all digital cameras have a lump of very expensive glass in front of them. This glass filters out all undesired radiation including UV and IR. Some camera technicians can replace this glass with an IR filter which only allows infra red light to get to the sensor and you have an IR camera. Permanently. And you can see what the camera is pointed at which you can't do with an on-lens IR filter.
But it can yield wonderful monochrome effects. Two of my friends have them.
Again, each pixel sensor has a filter - red, green or blue - in front of it. It all gets more complex.
So I use a UV filter for protection and have shattered two over the years. The lenses escaped.

[Paul]

This is an interesting discussion. I have read somwhere that UV filters affect circ. polars. I don't know if this is true or not. I took mine off just in case. However there is another reason I removed it and that is I use a Cokin filter system( full and graduated filters) and sometimes get vignetting from the lens hood(s). Particularly with the lense on maximum wide angle. I would feel much more comfortable if I could put it back. Any recommendations?
Paul

[Henry Peach]

This is an interesting discussion. I have read somwhere that UV filters affect circ. polars.

I've never heard of that, but stacking filters can cause vignetting with some lenses. And of course every glass/air meeting increases refraction and the chances for flare or contrast/saturation reduction.

[Paul]

Further to Henry Peach's reply the problem with respect to the lense hood is interference at the extremities of the of the picture when the lense has been set to its maximum wide angle position. I took the UV filter off to bring the filter cage and consequently the rectangular lense hood closer to the camera lense. It has helped a little but I would feel much more comfortable if I could somehow eliminate the problem and put the UV filter back.
Regards, Paul

[Colin Southern]

Further to Henry Peach's reply the problem with respect to the lense hood is interference at the extremities of the of the picture when the lense has been set to its maximum wide angle position. I took the UV filter off to bring the filter cage and consequently the rectangular lense hood closer to the camera lense. It has helped a little but I would feel much more comfortable if I could somehow eliminate the problem and put the UV filter back.
Regards, Paul

With UWA lenses all you can really do (if you want to have a protection filter mounted) is go for a slim variety. Normally the worst you'll get is a little vignetting which is easily fixed in PP.

The real issues start with multiple stacked filters - then it's not so much an issue with "vignetting" as it is one of outright obstruction.

[shreds]

I have always believed in UV filters for protection and have shattered a filter but got away without lens damage.

My Nikon 14-24 mm design prevents any threaded filter being added. At first I was wary of this, but now don't worry too much as the lens has stood up to all the usual punishment without any problems (I do clean it very carefully though). Gorgeous lens either way.

The only time I find a filter can be a problem is at night (I do quite a bit of night shooting), when flare does become an issue. I normally just abandon my zooms in favour of a prime, to ensure there is less glass, as all my primes do have filters. I dont believe that they make much optical difference if using high quality lens/filter combinations. I never stack filters these days, no need, but have seen the vignetting effect this can cause.

[Henry Peach]

The only time I find a filter can be a problem is at night (I do quite a bit of night shooting), when flare does become an issue.

Night photography sets up the perfect conditions to really notice the effect of filter flare: very bright highlights and darker backgrounds in which the flare stands out, and is easily noticed (see the link to the LL article below). The exact same thing is occurring during the day, but it's harder to see in the photo. It'll just be a spot or area of slightly lower contrast and saturation which may be hardly noticeable, in which case I guess it doesn't really matter. The reason I avoid them is not because I'm worried other folks will see the flaw. In my experience they probably won't unless I point it out to them. But once I notice it I'll see it every time, and it'll drive me crazy!

I think that like any other tool UV/protective filters must be assessed on an individual basis. Just consider that the reason why UV filters are pushed on new photographers as a must have item is not just to protect the lens; it's to sell more accessories. Take a dozen random, beginning, enthusiastic photographers, and see how many use UV filters? I bet it's most; they were told it's a necessity. Now take a dozen photographers who've been at it for over a decade, and see how many of them are using UV filters? I bet it's less than half.

http://www.luminous-landscape.com/co...m-feb-05.shtml

[Colin Southern]

Night photography sets up the perfect conditions to really notice the effect of filter flare: very bright highlights and darker backgrounds in which the flare stands out, and is easily noticed (see the link to the LL article below). The exact same thing is occurring during the day, but it's harder to see in the photo.

If you're using a high-quality multi-coated filter then during the day (when you're not dealing with extreme contrast ratios like you are at night) then it shouldn't be visable at all; with reflectances in the order of 0.2% that portion of the light is outside the dynamic range of the sensor.

Just consider that the reason why UV filters are pushed on new photographers as a must have item is not just to protect the lens; it's to sell more accessories.

I don't doubt for a moment that that's the case, but although the store does get to make more $$$ from the transaction, the user also benefits; I see it as being like the store that asks "and would you like batteries with that" when you're buying a new torch; yes, they make more money, but they're also doing you a noble service in that the torch is useless without the batteries.

What I find so interesting with these debates is that it's generally accepted that under normal shooting conditions there is no detectable loss of IQ, and yet there are scores of people who will attest to have had a lens saved by using a filter (and I'm one of them) - not to mention less insidious damage caused by constant cleaning of a front element, expecially where the likes of salt crystals are concerned. And keeping in mind also that Canon state in the manual for many lenses that the weather sealing is NOT COMPLETE unless a filter is fitted. So there are many potential benefits and no "down-side" under normal shootig conditions.

I like to draw a parallel with wearing a seatbelt when driving; even today, many people don't wear one - but I'll bet those who's life has been saved by one do - and I'll bet those who have been seriously injured through not wearing one probably do now too; I'm willing to be that the 1/2 who don't use a filter for front element protection yet probably haven't damaged a front element yet, with the over-riding rational (just like with seatbelts) being "accidents only happen to other people".

[Henry Peach]

What I find so interesting with these debates is that it's generally accepted that under normal shooting conditions there is no detectable loss of IQ, and yet there are scores of people who will attest to have had a lens saved by using a filter (and I'm one of them) - not to mention less insidious damage caused by constant cleaning of a front element, expecially where the likes of salt crystals are concerned.

My experiences are the opposite of yours. I own over 150 cameras with at least one lens for each, and I've never bashed a lens in almost 20 years. I've never even scratched the supposedly fragile coatings on my vintage lenses (modern multicoatings are hardened, and shouldn't have any issues with proper cleaning). I've been to the ocean, climbed mountains, and shoot from the stage-front crowd at punk rock bars several times weekly, and my lenses are fine. On the other hand I have experienced filter flare that I did not notice in the field.

I'm not debating the issue; I'm presenting my experience. I don't care what other photographers do. As I said, filters are tools, and their value must be assessed individually. I'm not saying don't use them. I'm just saying that anything you stick on the end of a lens should be considered. Most folks are told by a salesperson when they buy their first SLR/DSLR that protective filters are a must, and then they never question it.

[Colin Southern]

My experiences are the opposite of yours. I own over 150 cameras with at least one lens for each, and I've never bashed a lens in almost 20 years.

Unfortunately, this is the "big unknown" when it comes to accidents and "insurance"; just because someone hasn't damaged a lens in 20 years doesn't mean that they won't damage one tomorrow - I'm a careful driver and I haven't had a serious accident that would have required the restraint of a seatbelt in over 30 years of driving ... but I still wear a seatbelt every time I go driving. Unfortunately, unlike my driving record, and your lens record, accidents do happen to a lot of other people.

I'm not debating the issue; I'm presenting my experience. I don't care what other photographers do. As I said, filters are tools, and their value must be assessed individually. I'm not saying don't use them. I'm just saying that anything you stick on the end of a lens should be considered. Most folks are told by a salesperson when they buy their first SLR/DSLR that protective filters are a must, and then they never question it.

I hear what you're saying - and I don't really disagree with any of it; it's just my opinion that in using filters for protection, generally the "cheap insurance" benefits outweigh the (arguably) theoretical loss of IQ. I accept that salesman sell them (and owners often accept the advice as being gospel), but in my mind (assuming a quality filter) then it's tantamount to being "good advice, but for all the wrong reasons" - but the protection benefits are realised all the same.

Like you, I don't get too hung up over it - I'm often shooting in very contrasty light so my UV's come off (and to reduce vignetting) - but for everyday "normal" shooting I leave them on - and feel far more relaxed knowing that they're there.