A Clear or UV Filter for Lens Protection or Not??

A filter has a mirror like surface that is parallel to the sensor. This can set up an infinity mirror effect.

When using filters, digital shooters need to be more careful than film shooters, about light sources (or bright objects) in (or near) the field of view.
What??? You forget something between the filter and the sensor/film, called the "lens".

Also, "infinity mirrors" depend on almost-100% reflection properties. Camera filters transmit most of the light energy and reflect only a small portion (especially if they are coated). So your "infinity mirror" would attenuate below detectable level after one or two reflections.
Even one or two visible reflections might be a problem in an image.
Add-on filter ghosting occurs between the filter and the top surface of the first element of the lens.
According to Canon, digital cameras do introduce issues with reflections between the sensor and a flat filter.
No. The Canon text you quoted referred separately to ghosting between the filter and the first lens element, and ghosting between the last element and the film. Two different ghosting artifacts. No visible ghosting occurs through an intermediate lens element.
 
Actually, this is from Canon's excellent book "EF Lens Work II The Eyes of EOS" (September 2006, Eighth edition). This book contains discussions of the challenges facing lens designers, and how they address them.

According to Canon: "The reflective characteristics of the image sensors in a digital camera differ from those of film in that they possess a higher reflectivity as well as a characteristic known as regular or “mirror” reflection, which has the effect of creating flaring and ghosting inside the lens when light from a bright source enters the lens and reflects back to the image sensor."

and "Players in a stadium or racecars whizzing round the circuit. All are lit up by the bright lighting in the stands, or the headlights on the cars, creating numerous bright light sources. Ordinary super-telephoto lenses have protective glass in front of the first lens unit. If this glass is flat, any light entering the lens from a bright light source will be reflected off the image sensor and back onto the inside of the protective element, causing spot-shaped ghosting.* To prevent this, meniscus lenses are used as the protective glass on all of Canon’s largeaperture IS super-telephoto lenses. Meniscus lenses are spherical lenses which have the same curvature on both sides of the lens. By using these lenses as the protective glass, the light reflected off the image sensor forms an image in front of the image sensor and then disperse."

and "*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."

But you do bring up a good point. Many people are quick to dismiss facts if they don't match their world view. I should have included a reference in my previous post.

If you search around on the web, you may be able to find a PDF of the Lens Work book. If you are interested in issues surrounding the design and use of lenses, it really is an interesting read.
Everything in those quotes from Canon is correct. However, you have conflated two separate quotes and ended up with a false conclusion.

The first quote talks about the reflective surface of the digital sensor increasing the flares and ghosts inside the lens. There is no mention of filters in this quote.

The second quote talks about reflections from a flat filter in front of the lens and it says "reflected off the image sensor". However, nowhere do they say that a matte surface sensor would cure this problem. I am sure that Canon would acknowledge that the problem with reflections off the back of a filter can occur equally with a matte surface film as with a reflective surface sensor.

Any good scientist must be careful not to read into a text things that are not there and end up jumping to false conclusions.

If you have a film camera, I'm sure you can easily set up an experiment to demonstrate ghost images from a filter in front of the lens. Set the lens wide open and if it is a zoom, use the long end of the zoom. Make sure there is a very bright light somewhere near the centre of the frame (but not exactly in the centre). Take the same shot with and without a filter.

Experiment is always the best way to resolve disagreement in science.
 
Actually, this is from Canon's excellent book "EF Lens Work II The Eyes of EOS" (September 2006, Eighth edition). This book contains discussions of the challenges facing lens designers, and how they address them.

According to Canon: "The reflective characteristics of the image sensors in a digital camera differ from those of film in that they possess a higher reflectivity as well as a characteristic known as regular or “mirror” reflection, which has the effect of creating flaring and ghosting inside the lens when light from a bright source enters the lens and reflects back to the image sensor."

and "Players in a stadium or racecars whizzing round the circuit. All are lit up by the bright lighting in the stands, or the headlights on the cars, creating numerous bright light sources. Ordinary super-telephoto lenses have protective glass in front of the first lens unit. If this glass is flat, any light entering the lens from a bright light source will be reflected off the image sensor and back onto the inside of the protective element, causing spot-shaped ghosting.* To prevent this, meniscus lenses are used as the protective glass on all of Canon’s largeaperture IS super-telephoto lenses. Meniscus lenses are spherical lenses which have the same curvature on both sides of the lens. By using these lenses as the protective glass, the light reflected off the image sensor forms an image in front of the image sensor and then disperse."

and "*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."

But you do bring up a good point. Many people are quick to dismiss facts if they don't match their world view. I should have included a reference in my previous post.

If you search around on the web, you may be able to find a PDF of the Lens Work book. If you are interested in issues surrounding the design and use of lenses, it really is an interesting read.
Everything in those quotes from Canon is correct. However, you have conflated two separate quotes and ended up with a false conclusion.

The first quote talks about the reflective surface of the digital sensor increasing the flares and ghosts inside the lens. There is no mention of filters in this quote.

The second quote talks about reflections from a flat filter in front of the lens and it says "reflected off the image sensor". However, nowhere do they say that a matte surface sensor would cure this problem. I am sure that Canon would acknowledge that the problem with reflections off the back of a filter can occur equally with a matte surface film as with a reflective surface sensor.
I urge to go read the entire "How to deal with flaring and ghosting particular to digital photography" section from the book.

I am fairly certain that it deals with how flare issues increase with digital cameras.

It then goes on to specifically mentions issues with flat protective lenses, and how light can "be reflected off the image sensor and back onto the inside of the protective element". Again the context of this statement is on the differences between film and digital.
Any good scientist must be careful not to read into a text things that are not there and end up jumping to false conclusions.

If you have a film camera, I'm sure you can easily set up an experiment to demonstrate ghost images from a filter in front of the lens. Set the lens wide open and if it is a zoom, use the long end of the zoom. Make sure there is a very bright light somewhere near the centre of the frame (but not exactly in the centre). Take the same shot with and without a filter.

Experiment is always the best way to resolve disagreement in science.
Yes, experiments are good.

Take the consider the following image:

With filter
With filter

The image was taken with a filter. It has no obvious image degradation from filters. There is no light source in the field of view. Yes, the contrast isn't great, but it was a consumer lens, not a pro level lens.

Now let's take the filter off and try another shot:

Without filter
Without filter

Even though the image with the filter had no obvious issues, there are improvements in the no-filter image. The contrast has gone up. The reflection on the glass has gone (apparently it was flare, not a reflection).

To be fair, there was a light source on the ceiling above the frame, but that was mrely near, not in, the field of view.

.

Obviously, not all tests will give results like this. It's going to depend on the specifics of the testing situation. I could adjust me lighting, and repeat this test with the same filter, lens and body, and get no noticeable difference.

That's a big problem here. In order to get a meaningful test, you really do need to run the test under your actual shooting conditions, with a specific lens, filter and body. Unfortunately, your results are only applicable to that particular configuration.
 
Last edited:
A filter has a mirror like surface that is parallel to the sensor. This can set up an infinity mirror effect.

When using filters, digital shooters need to be more careful than film shooters, about light sources (or bright objects) in (or near) the field of view.
What??? You forget something between the filter and the sensor/film, called the "lens".

Also, "infinity mirrors" depend on almost-100% reflection properties. Camera filters transmit most of the light energy and reflect only a small portion (especially if they are coated). So your "infinity mirror" would attenuate below detectable level after one or two reflections.
Even one or two visible reflections might be a problem in an image.
Add-on filter ghosting occurs between the filter and the top surface of the first element of the lens.
According to Canon, digital cameras do introduce issues with reflections between the sensor and a flat filter.
No. The Canon text you quoted referred separately to ghosting between the filter and the first lens element, and ghosting between the last element and the film. Two different ghosting artifacts. No visible ghosting occurs through an intermediate lens element.
But Canon does suggest that ghosting to the point of being a problem with a strong light source is possible when using a filter.

"*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."
 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
Every lens surface reflects some of the incoming light. If there are point light sources, you may see ghost images on the image due to reflections off the lens-side surface of the filter. Under some conditions these can be very noticeable. Most of the time not.
One of the differences between film and digital cameras is that undeveloped film has a matte surface, and a digital sensor has a mirror like surface.

A filter has a mirror like surface that is parallel to the sensor. This can set up an infinity mirror effect.

When using filters, digital shooters need to be more careful than film shooters, about light sources (or bright objects) in (or near) the field of view.
For example, the old Tamron Adaptall 90mm f/2.5 macro is a wonderful and sharp lens but the rear element is slightly concave and under certain circumstances, usually bright rear lighting, will reflect light back onto the sensor, resulting on a round patch with a slight colour shift to purple in the centre of the image.
 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
UV filters were designed for film and can actually cause a color shift with digital. Instead, use a high-quality plain glass filter. If you use a quality one I doubt there would be a noticeable degradation in IQ. I personally never use such filters and instead depend on a lens hood for protection. In 55 years I have never had a front element damaged so I don't feel it's necessary to spend the money on a clear filter.
 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
Petapixel.com has a rather disturbing video on this subject. Not for the faint hearted...

 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
Petapixel.com has a rather disturbing video on this subject. Not for the faint hearted...

https://petapixel.com/2022/06/08/ti...le-to-scratch-their-camera-lenses-with-rocks/
It would have been much cheaper for her to put a clear filter on the lens and scratch the filter instead of the lens. :-D
 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
Petapixel.com has a rather disturbing video on this subject. Not for the faint hearted...

https://petapixel.com/2022/06/08/ti...le-to-scratch-their-camera-lenses-with-rocks/
It would have been much cheaper for her to put a clear filter on the lens and scratch the filter instead of the lens. :-D
Yes agreed but that's the point of the video, it's confusing
 
Im out in the dust, dirt and elements with my camera a great deal. I always keep a good UV filter on all of my lenses except the 14-24 and the 200mm. While I try to be careful and do things like try not to change lenses in the field and keep my cleaning gear with my camera gear as well as a can of compressed air.

I always use HOYA or B+W Filters.

Will this protection hurt image quality or not?
I wouldn't use a filter unless I really needed an ND or polarizer for specific shot. With all the coatings, special glass, etc.. I think the petal hood offers plenty of protection and dust on outer element doesn't really show anyways.
Precisely...also, do not use filters when shooting Aurora Borealis or certain night time shots where filter coating rings will be revealed from stray light sources.

-M
 
Last edited:
A filter has a mirror like surface that is parallel to the sensor. This can set up an infinity mirror effect.

When using filters, digital shooters need to be more careful than film shooters, about light sources (or bright objects) in (or near) the field of view.
What??? You forget something between the filter and the sensor/film, called the "lens".

Also, "infinity mirrors" depend on almost-100% reflection properties. Camera filters transmit most of the light energy and reflect only a small portion (especially if they are coated). So your "infinity mirror" would attenuate below detectable level after one or two reflections.
Even one or two visible reflections might be a problem in an image.
Add-on filter ghosting occurs between the filter and the top surface of the first element of the lens.
According to Canon, digital cameras do introduce issues with reflections between the sensor and a flat filter.
No. The Canon text you quoted referred separately to ghosting between the filter and the first lens element, and ghosting between the last element and the film. Two different ghosting artifacts. No visible ghosting occurs through an intermediate lens element.
But Canon does suggest that ghosting to the point of being a problem with a strong light source is possible when using a filter.

"*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."
That was my point. Unrelated to the reflectivity of the sensor or film, as the internal reflection in this case is from the lens element directly below the filter.
 
there are improvements in the no-filter image. The contrast has gone up. The reflection on the glass has gone (apparently it was flare, not a reflection).

To be fair, there was a light source on the ceiling above the frame, but that was mrely near, not in, the field of view.
It is often the case that the source of the flare is a strong light source outside the sensor frame but whose light still impinges on the filter surface and is refracted and then reflected into the part of the signal path that hits the sensor. That is easy to confirm, as you just did.
 
A filter has a mirror like surface that is parallel to the sensor. This can set up an infinity mirror effect.

When using filters, digital shooters need to be more careful than film shooters, about light sources (or bright objects) in (or near) the field of view.
What??? You forget something between the filter and the sensor/film, called the "lens".

Also, "infinity mirrors" depend on almost-100% reflection properties. Camera filters transmit most of the light energy and reflect only a small portion (especially if they are coated). So your "infinity mirror" would attenuate below detectable level after one or two reflections.
Even one or two visible reflections might be a problem in an image.
Add-on filter ghosting occurs between the filter and the top surface of the first element of the lens.
According to Canon, digital cameras do introduce issues with reflections between the sensor and a flat filter.
No. The Canon text you quoted referred separately to ghosting between the filter and the first lens element, and ghosting between the last element and the film. Two different ghosting artifacts. No visible ghosting occurs through an intermediate lens element.
But Canon does suggest that ghosting to the point of being a problem with a strong light source is possible when using a filter.

"*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."
That was my point. Unrelated to the reflectivity of the sensor or film, as the internal reflection in this case is from the lens element directly below the filter.
Every time you add more glass you add more opportunities for reflections. That's why it is important to use high quality glass with high quality coatings.

According to Canon, the mirror like surface of the sensor adds additional opportunities for reflections, that are not present with film photography. Furthermore, Canon worries that there are even more opportunities when there is a flat element parallel to the sensor (such as a filter).

Canon goes on to say, that they strive to make sure the front element in a lens is affordable to replace. In cases where it would not be affordable, they add a sacrificial protective element. However, they are careful to avoid flat protective elements, and use curved protective elements.

Obviously, there are very many situations where filters don't cause any issues whatsoever. The trick is that there are situations where even a good filter can cause problems. The key is to know when a filter might be an issue, and consider removing it those situations.

One of life's ironies, is that the situations where a filter is most likely to cause an issue (such as the sun setting behind the model at a beach) are frequently the situations where you think you need the most protection.
 
there are improvements in the no-filter image. The contrast has gone up. The reflection on the glass has gone (apparently it was flare, not a reflection).

To be fair, there was a light source on the ceiling above the frame, but that was mrely near, not in, the field of view.
It is often the case that the source of the flare is a strong light source outside the sensor frame but whose light still impinges on the filter surface and is refracted and then reflected into the part of the signal path that hits the sensor. That is easy to confirm, as you just did.
Yes.

Had I relied on a test in a controlled studio environment, I might have incorrectly concluded that this filter never caused image problems.

That's a big problem. If you want to know if a filter is degrading your image, you really need to test under your actual shooting circumstances. This can be a problem for people who shoot in the field in a variety of changing circumstances.

An interesting test would be to put a quality filter on a pro lens and shoot towards a light source. Then try the same shot with a consumer lens, and no filter.
 
Should you use a UV, clear, or haze filter on your camera lens, or is it just an unnecessary nuisance that may degrade the quality of your photos?


Good video and worth watching. It is what I have done for years.
Very sensible video. The short piece of amateur psychology at the end really made me smile. I think he hit the nail on the head!
 
Wet-cleaning a lens is seldom advisable, and even though lens coatings are tough, micro scratches from small grit particles will abrade the coating over time after repeated cleanings.
This scratch was far worse than micro scratches you might get from cleaning.

 
But Canon does suggest that ghosting to the point of being a problem with a strong light source is possible when using a filter.

"*When a filter is mounted on a regular lens, ghosting can occur above the same spot where there is a strong light source inside the frame. When this occurs, remove the filter to photograph."
That was my point. Unrelated to the reflectivity of the sensor or film, as the internal reflection in this case is from the lens element directly below the filter.
But the filter can cause a reflectivity problem which after all matters more than the technical details of why the problem occurs. Sometimes we lose sight of the big picture while getting caught up in the minutia. I've been guilty of it myself.
 
Should you use a UV, clear, or haze filter on your camera lens, or is it just an unnecessary nuisance that may degrade the quality of your photos?


Good video and worth watching. It is what I have done for years.
While it sounds like a well reasoned discussion, the logic expressed in the video has some flaws, and some serious bias.

First of all, it implies that the issue against filters is a bunch of vocal people who have a strong bias against adding extra glass to the optical path. Essentially painting them as religious who take the position that if god (or the lens designer) had wanted that piece of glass there, then he would have put it there.

This is clearly not the case. As has been pointed out, there actually can be image quality issues when using filters. In fact the video does admit this when he points out that any time you really do need to get a good image you should remove the filter.

His next issue is that he claims that cleaning causes micro scratches. Therefore you should clean a filter, and not your lens. This is mostly an emotional argument. He obviously doesn't believe that this issue is significant. If he did, he would need to constantly replace his filters due to micro scratches. The fact that he shoots through his filters, even though they have micro scratches, indicates that he doesn't think micro scratches are really a problem.

Next he tries the scare tactic of implying that a filter can save you from thousands of dollars of damage. Imagine that you have a $3,000 lens, and the filter prevents front element damage. He implies that the filter can save you from a $3,000 loss. This is, of course, nonsense. If your $3,000 lens has a damage front element, you don't throw out the lens, you get the front element replaced.

He then told the story of a time when he got something on the coating of his filter, and the filter was ruined. Let's ignore the possibility that he simply had a filter with a cheap coating.

But let's assume this was a situation where the filter saved his lens. His claim is that a good filter should cost about 10% of the price of the lens. A front element repair should cost far less than half the cost of the lens. He has one reported issue. Assuming he has more than 5 lenses, he would have saved money by not buying all those filters, and just repairing that lens. But as he pointed out, people have a negativity bias. One bad apple spoils the barrel. If you read between the lines, his filter use really is costing him money in the long run.

He also goes on to claim that filters are particularly important when shooting in dusty environments or at the beach. This implies that in these circumstances a filter protects your lens. This ignores the real danger in these environments of dust or contamination getting into the internals of your lens. Filters don't help there. Of course, he does admit that if you are shooting at the beach with the sun setting behind the model, you should remove the filter.

But I do absolutely agree that there are some very vocal people who really want to convince others that their position on filters is the correct one. Some even go as far as to make YouTube videos.

He is also correct that most people's decisions on filters are based on emotion rather than reason.

The bottom line remains that the biggest difference a filter makes is usually emotional. If you feel better using one, then go ahead and use one. If they make you uncomfortable, then avoid them. Do keep in mind that whatever your preference, there are a few situations where you really should use them, and a few situations where you should avoid them. For the vast majority of situations it doesn't really matter, do what makes you feel better.
 

Keyboard shortcuts

Back
Top