ND filters for drones photography, timelapses and videography (contains video)

[WithTheBirds]

A polarized filter will cut out specific wavelengths of light so I suppose you are correct that it will cut down the amount of light reaching the sensor. However, It’s cutting out “unwanted” light rather than light at all frequencies, what an ND filter is suppose to do. That doesn’t mean you need to decrease shutter speed to still get a proper exposure necessarily though. I personally think of averaged metering as a “ball park” figure. It would require a histogram to know if you really need to add more shutter, in most cases people probably shouldn’t.

Maybe color cast isn’t the correct term, I actually took that from what you said, but it does affect color particularly greens and blues. we all know that colors are a mixture of certain wavelengths so by removing a certain wavelength, for instance with a polarized filter, we would expect a change. Clearly there is a change when using a polarized filter whether that’s suppose to be just a luminance change and not a hue change I am sure.

View attachment 87288
View attachment 87289
View attachment 87290

Your assumption, at least with respect to a polarising filter attenuating or admitting light based on wavelength is wrong. The source you copied the images from should give you the explanation.

The reason a polarising filter can significantly reduce reflections from non metallic surfaces (allowing us to see below the surface of water, through glass etc) is because the reflected light is polarised and when the filter is correctly orientated the reflected light is largely blocked.

Despite what you seem to believe when a polariser is fitted to a lens some additional exposure will be required (over that with no filter). That is an inescapable fact.

 

[WithTheBirds]

well put @brett8883

The fact you like the sound of something doesn’t make it true.

 

[AMann]

Neutral density filters are really useful when shooting a landscape with a bright foreground (snow or light sand or rocks) against a dark background or sky. You can still get a good exposure of the background without blowing out the highlights of the ground, and that means loss of color and detail.

They allow one to shoot with slower shutter speeds which can smooth out water and other moving objects for effects, and to get more details of bright objects such as a full moon or brightly lit snow.

A Guide to Neutral Density Filters

This article is a comprehensive guide to using Format Hitech Firecrest neutral density filters in photography.

www.bhphotovideo.com

I really love to use graduated ND filters on my DSLR, they allow one to get multiple stop differences between the top and bottom of a scene, and are particularly good for shooting a dark landscape against a bright cloudy sky. This article has good examples of this. I have not tried any graduated ND filters on my M2P, but suspect the results of using one for orthagonal (sky and land combined straight ahead shots) landscape shots would give spectacular results.

Neutral Density Filters: Everything You Need to Know

What are neutral density filters, and how can you use them for beautiful results? Discover everything you need to know!

digital-photography-school.com

 

[WithTheBirds]

Neutral density filters are really useful when shooting a landscape with a bright foreground (snow or light sand or rocks) against a dark background or sky. You can still get a good exposure of the background without blowing out the highlights of the ground, and that means loss of color and detail.

An ND is of no benefit here. A graduated ND might be however you still have the limitation of the transition between clear and grad being fixed and might prove not to be optimum for your composition.

A better solution is to shoot multiple frames and blend exposures in post processing.

they also allow you to shoot with increased depth of field which means more in focus.

They allow one to shoot with slower shutter speeds which can smooth out water and other moving objects for effects, and to get more details of bright objects such as a full moon or brightly lit snow.

The focal length and typical lens to subject distances on the magics make DOF field considerations irrelevant.

Bright subjects are easily handled by adjusting shutter, aperture and ISO, no need of ran ND here.

Yes, longer exposures can be obtained to create motion blur. You can also get a similar effect without having to worry about camera shake with shooting multiple frames and performing a simple stack, align and blend in post.

 

[AMann]

An ND is of no benefit here. A graduated ND might be however you still have the limitation of the transition between clear and grad being fixed and might prove not to be optimum for your composition.

A better solution is to shoot multiple frames and blend exposures in post processing.



The focal length and typical lens to subject distances on the magics make DOF field considerations irrelevant.

Bright subjects are easily handled by adjusting shutter, aperture and ISO, no need of ran ND here.

Yes, longer exposures can be obtained to create motion blur. You can also get a similar effect without having to worry about camera shake with shooting multiple frames and performing a simple stack, align and blend in post.

HDR photography works really well, I use it a lot, but it doesn’t work for video or on moving objects. Windy days where trees move, moving people and cars or boats, a moving camera, all create weird affects in HDR images.

 

[WithTheBirds]

HDR photography works really well, I use it a lot, but it doesn’t work for video or on moving objects. Windy days where trees move, moving people and cars or boats, a moving camera, all create weird affects in HDR images.

Yes. The benefit of ND’s in shoring video are well demonstrated. They still won’t get you anywhere in balancing bright skies and dark foregrounds (compressing dynamic range).

I prefer to manually blend images rather automated HDR. I have seen it used to good effect. It is an art in itself to keep the result looking within the realms of natural.

 

[Meta4]

A polarized filter will cut out specific wavelengths of light so I suppose you are correct that it will cut down the amount of light reaching the sensor. However, It’s cutting out “unwanted” light rather than light at all frequencies, what an ND filter is suppose to do. That doesn’t mean you need to decrease shutter speed to still get a proper exposure necessarily though.

You only have to look at a Pol filter to see that it's going to cut the light level, both polarised and also some degree of all visible light.
They are dark like sunglasses.

Clearly there is a change when using a polarized filter whether that’s suppose to be just a luminance change and not a hue change I am sure.

Properly used, a Pol filter will reduce the glare of reflected light, allowing the true colour of an object to show.

The catch is that to properly use a Pol filter is awkward with a drone.
If you point the drone anywhere else after aligning the filter it's not working properly.

I think there is a question of what is meant when people keep trying to tell people in essence "what you're enjoying isn't true" and attach some sort of "the colors don't change" comment with it.. and a question of what is meant when people keep saying they are enjoying the benefits of using x y or z filter.

The situation is more like a few people who have a flawed understanding of how ND and Pol filters work, what they do and how they are used and others that have a good knowledge of photographic principles.
The problem comes when members of the first group ignore good advice and are convinced that their incorrect ideas are correct.

First, if people weren't noticing a difference in their photos and output, we probably wouldn't have responded to the opening post to say that yes, for us, we are using them and finding great benefit.
Secondly, if a pixel color value changes, then that photo's color is, indeed, affected by said filter, no matter how eloquently or forcefully you can say it isn't.

There are lots of factors at work and it's easy for an inexperienced (and sometimes even experienced) photographers to be convinced that they understand what they are observing but have it very wrong.
What a Pol or ND filter can do is well established and whether you accept it or not and no matter how forcefully you insist you understand things.

For what it's worth, I'm a fairly non-versed photographer, and the camera on my Mavic Pro 2 is the most amazing camera I've ever used, and I've found this community to be mostly helpful, though there are some who insist that others see things their way about this or that.. just ignore them, they'll eventually give up.

You come across as someone that doesn't want to understand and resent anyone that might try to clear up your misunderstandings.
If you aren't a very experienced photographer, it's quite possible that other members with 40 or more years of high-level photographic experience might actually have some valuable information to impart.

 

[brett8883]

Your assumption, at least with respect to a polarising filter attenuating or admitting light based on wavelength is wrong. The source you copied the images from should give you the explanation.

The reason a polarising filter can significantly reduce reflections from non metallic surfaces (allowing us to see below the surface of water, through glass etc) is because the reflected light is polarised and when the filter is correctly orientated the reflected light is largely blocked.

Despite what you seem to believe when a polariser is fitted to a lens some additional exposure will be required (over that with no filter). That is an inescapable fact.

I acknowledge your extensive knowledge of technical photography and defer to you in most cases. However, I’ve found in the past that sometimes you dismiss something as technically incorrect when for all intents and purposes the original statement was true for the purpose. For instance you once told me that my diagnosis of a problem with someone’s sensor was incorrect that it was instead the alignment of said sensor that was the issue. I assume your explanation pinpointed the issue more acutely but that didn’t mean my diagnosis was incorrect and the prescribed remedy was the same, replacement.

For this reason I must press you further to figure out if your issue with something is on some technicality or if it’s the underlying message altogether. Remember if we want to help people we must explain it on a level they will understand. Let’s not lose sight of the forest by looking too closely at the leaves.

So going back to the meaning of what I said rather than the technicalities, if someone likes the look of photos taken with a ND/PL filter over sans filter do you believe that it was a good suggestion or a bad suggestion that they use a polarized filter that does not have the ND component to improve the sharpness and and overall quality of their photos while retaining the color profile achieved with the ND/PL filter?

 

[WithTheBirds]

I acknowledge your extensive knowledge of technical photography and defer to you in most cases. However, I’ve found in the past that sometimes you dismiss something as technically incorrect when for all intents and purposes the original statement was true for the purpose. For instance you once told me that my diagnosis of a problem with someone’s sensor was incorrect that it was instead the alignment of said sensor that was the issue. I assume your explanation pinpointed the issue more acutely but that didn’t mean my diagnosis was incorrect and the prescribed remedy was the same, replacement.

For this reason I must press you further to figure out if your issue with something is on some technicality or if it’s the underlying message altogether. Remember if we want to help people we must explain it on a level they will understand. Let’s not lose sight of the forest by looking too closely at the leaves.

So going back to the meaning of what I said rather than the technicalities, if someone likes the look of photos taken with a ND/PL filter over sans filter do you believe that it was a good suggestion or a bad suggestion that they use a polarized filter that does not have the ND component to improve the sharpness and and overall quality of their photos while retaining the color profile achieved with the ND/PL filter?

I can't recall the interaction where you claim I was dismissive (the sensor issue)- do you have reason to believe I was factually wrong in what I said? Did you point it out? I have no issue in being wrong about something. It happens a lot. To all of us. That is how we increase our understanding of things. Why might you take issue with someone being more specific?

If I am pressed further, as seemed to be your intent as stated, I am not focussing purely on any technical aspect more than to question generally your underlying message. With respect to your current issue generally, if I have understood you correctly, the meaning of what is said vs technicalities I am struggling with what seems to be your point. That is, their seems to be no ambiguity or subjective considerations.

To consider the question you have put to me now directly, as you phrased it- do I believe that it was a good suggestion or a bad suggestion that someone might use a polarising filter that does not have the ND component to improve the sharpness and and overall quality of their photos while retaining the color profile achieved with the ND/PL filter?

Given I am faced, in the manner you have presented it to me, with a binary proposition (good vs bad suggestion) I have to say it was a bad suggestion. As I had already inferred, certain of your apparent understandings of the performance of the subject filters seem to be incorrect. Specifically you are mistaken to the extent you believe that a PL filter might improve sharpness and that a ND/PL might achieve a particular colour profile. An ND or PL that imparts a noticeable change to the colour profile should be considered faulty. You also seem not to appreciate that all PL filters act similarly to ND's in reducing the light admitted to the lens, that effect is well known to be 1-3 stops.

 

[brett8883]

I can't recall the interaction where you claim I was dismissive (the sensor issue)- do you have reason to believe I was factually wrong in what I said? Did you point it out? I have no issue in being wrong about something. It happens a lot. To all of us. That is how we increase our understanding of things. Why might you take issue with someone being more specific?

If I am pressed further, as seemed to be your intent as stated, I am not focussing purely on any technical aspect more than to question generally your underlying message. With respect to your current issue generally, if I have understood you correctly, the meaning of what is said vs technicalities I am struggling with what seems to be your point. That is, their seems to be no ambiguity or subjective considerations.

To consider the question you have put to me now directly, as you phrased it- do I believe that it was a good suggestion or a bad suggestion that someone might use a polarising filter that does not have the ND component to improve the sharpness and and overall quality of their photos while retaining the color profile achieved with the ND/PL filter?

Given I am faced, in the manner you have presented it to me, with a binary proposition (good vs bad suggestion) I have to say it was a bad suggestion. As I had already inferred, certain of your apparent understandings of the performance of the subject filters seem to be incorrect. Specifically you are mistaken to the extent you believe that a PL filter might improve sharpness and that a ND/PL might achieve a particular colour profile. An ND or PL that imparts a noticeable change to the colour profile should be considered faulty. You also seem not to appreciate that all PL filters act similarly to ND's in reducing the light admitted to the lens, that effect is well known to be 1-3 stops.

Issue with the sensor. Return it if it’s new enough or contact DJI support for a warrantee fix and be sure to show them these pictures. It’s happening in the same spot on certain ones. DJI has been conspicuously quick to replace them so they must know about the issue.

The issue is unlikely to be a sensor defect. They are a pass/fail proposition off the production line. Poor alignment of sensor and/or optical elements in assembly is almost certainly the culprit.

I don’t know if you were right or wrong but my point was sometimes you can dismiss something as being incorrect on some technical grounds that isn’t relevant to the context of the conversation. I mean this in the most polite way possible.

No big deal I’m not that sensitive. I like having these debates and it’s makes makes us all better for it. It just seems sometimes like the context of a statement as far as how it relates to what we were originally talking about Is forgotten.

I was saying that using a polarized filter will cause a sharper image OVER using an ND/PL filter because the amount of light let through an ND/PL filter is arbitrarily reduced necessitating a longer shutter all things equal than with a filter that is only polarized. I was NOT saying a polarized filter will cause a sharper image over no filter. Maybe this is my fault for not being more clear.

Now let’s talk about this color issue with polarized filters. I’m getting the impression from you that you are saying polarized filters do not change color in any way and are therefore useless? What is happening with the photos I linked above then? Or is your point that anything that can be done in post better than what can be done with a filter should be done in post rather than with a filter? In which case I can get behind that but not everyone has the knowledge or skill to do that and sometimes old school is cool even if it’s not as “good” as new school.

Last, can you please explain, in your opinion, what is the appropriate usage of a polarized filter if any.

 

[WithTheBirds]

I don’t know if you were right or wrong but my point was sometimes you can dismiss something as being incorrect on some technical grounds that isn’t relevant to the context of the conversation. I mean this in the most polite way possible.

No big deal I’m not that sensitive. I like having these debates and it’s makes makes us all better for it. It just seems sometimes like the context of a statement as far as how it relates to what we were originally talking about Is forgotten.

I was saying that using a polarized filter will cause a sharper image OVER using an ND/PL filter because the amount of light let through an ND/PL filter is arbitrarily reduced necessitating a longer shutter all things equal than with a filter that is only polarized. I was NOT saying a polarized filter will cause a sharper image over no filter. Maybe this is my fault for not being more clear.

Now let’s talk about this color issue with polarized filters. I’m getting the impression from you that you are saying polarized filters do not change color in any way and are therefore useless? What is happening with the photos I linked above then? Or is your point that anything that can be done in post better than what can be done with a filter should be done in post rather than with a filter? In which case I can get behind that but not everyone has the knowledge or skill to do that and sometimes old school is cool even if it’s not as “good” as new school.

Last, can you please explain, in your opinion, what is the appropriate usage of a polarized filter if any.

Actually, at least from my experience and extent of knowledge- a PL filter effect is one that can't be done in post processing.

The PL is exceptionally useful to remove, or at least largely reduce, the impact of light reflected from non metallic surfaces. It is very effective in revealing what lies below the water surface or behind glass. You can significantly eliminate distracting highlight reflections from vehicles and other objects. The sky becomes darker and most colours will appear more saturated.

I do find that a PL is more of a hinderance than a useful addition for drone photography however. Setting to the correct orientation prior to launch is inconvenient however you are also restricted in your orientation of flight for the effect to not vary during a take if you are shooting video or even to have optimum effect with stills.

The other issue you likely have experienced is that with the relatively wide FOV it is very easy to end up with a shot where the effect can be see to vary markedly over a large expanse of sky. It looks very artificial in those instances.

Edit: I didn’t recall that discussion about the sensor defect. I do recall a disagreement we had about dynamic range and in particular whether Lightroom could always be relied on to identify blown highlights with the over exposure warning feature (whatever the correct name for that might be). I think you accepted in the finish I was giving good advice on that occasion. There have been and will be many more instances where I will be wrong about things though. I would be disappointed if it wasn’t pointed out to me also. Far preferable than labouring on with a mistaken belief.

 

[brett8883]

Actually, at least from my experience and extent of knowledge- a PL filter effect is one that can't be done in post processing.

The PL is exceptionally useful to remove, or at least largely reduce, the impact of light reflected from non metallic surfaces. It is very effective in revealing what lies below the water surface or behind glass. You can significantly eliminate distracting highlight reflections from vehicles and other objects. The sky becomes darker and most colours will appear more saturated.

I do find that a PL is more of a hinderance than a useful addition for drone photography however. Setting to the correct orientation prior to launch is inconvenient however you are also restricted in your orientation of flight for the effect to not vary during a take if you are shooting video or even to have optimum effect with stills.

The other issue you likely have experienced is that with the relatively wide FOV it is very easy to end up with a shot where the effect can be see to vary markedly over a large expanse of sky. It looks very artificial in those instances.

Edit: I didn’t recall that discussion about the sensor defect. I do recall a disagreement we had about dynamic range and in particular whether Lightroom could always be relied on to identify blown highlights with the over exposure warning feature (whatever the correct name for that might be). I think you accepted in the finish I was giving good advice on that occasion. There have been and will be many more instances where I will be wrong about things though. I would be disappointed if it wasn’t pointed out to me also. Far preferable than labouring on with a mistaken belief.

Yes after that first run in I have found you to be extremely knowledgeable and realize your intentions are in the right place.

My question, remains and I for some reason can’t word this without it coming off like a challenge which isn’t my intention but looking at the photos above where the blue sky is darkened by the polarized filter and the green in the leaves is brightened. It also looks like, at least in the sky photo that the hue of blue has changed with some green maybe a touch of red removed. Couldn’t that be the color change/color cast/whatever is the correct term that people using ND/PL filters and swear it makes their photos look better could be seeing? Low quality filters is another possibility which is correct we have no way of knowing but it’s at least possible and is the reason for my suggestion.

I agree with you about the orientation of the filter being a hinderance with drone but the filter doesn’t need to at its maximum polarity to be working.

The only way I know of in post to mimic somewhat the effect of a polarized filter would be to use dehaze and the selective color tool in photoshop or luminosity sliders of an HSL tool. It of course won’t bring out details never recorded and it’s tough to use just the right amount of dehaze.

 

[WithTheBirds]

Yes after that first run in I have found you to be extremely knowledgeable and realize your intentions are in the right place.

My question, remains and I for some reason can’t word this without it coming off like a challenge which isn’t my intention but looking at the photos above where the blue sky is darkened by the polarized filter and the green in the leaves is brightened. It also looks like, at least in the sky photo that the hue of blue has changed with some green maybe a touch of red removed. Couldn’t that be the color change/color cast/whatever is the correct term that people using ND/PL filters and swear it makes their photos look better could be seeing? Low quality filters is another possibility which is correct we have no way of knowing but it’s at least possible and is the reason for my suggestion.

I agree with you about the orientation of the filter being a hinderance with drone but the filter doesn’t need to at its maximum polarity to be working.

The only way I know of in post to mimic somewhat the effect of a polarized filter would be to use dehaze and the selective color tool in photoshop or luminosity sliders of an HSL tool. It of course won’t bring out details never recorded and it’s tough to use just the right amount of dehaze.

I don’t mind being pressed- the answer is I don’t know for certain- I will remain confident though that it isn’t because the PL favours particular wavelengths in the visible spectrum. If I remember school science correctly (it’s a long time ago) the sky is blue because shorter wavelengths are scattered more effectively by the atmosphere -and- leaves are green because chlorophyll absorbs red and blue wavelengths. If the PL is effective in blocking the stray reflections we see more of the actual colour of the subject.

Perhaps you can check around for what is actually going on and confirm.

 

[brett8883]

I don’t mind being pressed- the answer is I don’t know for certain- I will remain confident though that it isn’t because the PL favours particular wavelengths in the visible spectrum. If I remember school science correctly (it’s a long time ago) the sky is blue because shorter wavelengths are scattered more effectively by the atmosphere -and- leaves are green because chlorophyll absorbs red and blue wavelengths. If the PL is effective in blocking the stray reflections we see more of the actual colour of the subject.

Perhaps you can check around for what is actually going on and confirm.

Ok thanks. And just for the record I didn’t mean it favors the same particular wavelength in all situations I just mean that, depending on the scene, it will affect some wavelengths more than others because some wavelengths will have more polarized light than others but what wavelength that is will change depends on the scene in contrast to a ND filter which always affects all wavelengths equally always.

 

[Pilgrim Pilot]

@WithTheBirds just doesn't want people using ND filters. we get that. the fact that he can't stop repeating himself in the context of a conversation where people are reporting very favorable results from ND and Pol filters is why I began ignoring him largely, but he has derailed this conversation so thoroughly, and it keeps coming up in my alerts, so I figured I'd agree with the observation that @WithTheBirds is missing the entire point of the OP and the reality that people are responding positively because they have achieved positive results.

It is laughable, except for how much time it has consumed.

 

[WithTheBirds]

@WithTheBirds just doesn't want people using ND filters. we get that. the fact that he can't stop repeating himself in the context of a conversation where people are reporting very favorable results from ND and Pol filters is why I began ignoring him largely, but he has derailed this conversation so thoroughly, and it keeps coming up in my alerts, so I figured I'd agree with the observation that @WithTheBirds is missing the entire point of the OP and the reality that people are responding positively because they have achieved positive results.

It is laughable, except for how much time it has consumed.

Thank you for the complete misrepresentation. I use ND filters, frequently in fact. And not just with drones or exclusively with video for that matter.

 

[hiflyer201]

I feel that I've lost the chance to get good sharp photos during the golden hour by having a ND filter on the lens, and It's caused me to question how much I'm really gaining from using filters for video. I'm thinkig that in most cases I'm just too distant from the subject to have a big effect from the filters.
One thing I'm certain of...if we're smart enough to just make sure to push the button...we're going to get some amazing shots that non dronie's can only dream about.

 

[Meta4]

I feel that I've lost the chance to get good sharp photos during the golden hour by having a ND filter on the lens, and It's caused me to question how much I'm really gaining from using filters for video. I'm thinkig that in most cases I'm just too distant from the subject to have a big effect from the filters.

ND filters for video are not essential at all. Unless you have moving subjects, the filter isn't really going to do much Nd if there is motion, it might still look fine anyway.

 

[vicvideopic]

You do need them for time lapse photography, if you want the shutter speed of around 1/4 to capture movement of cars/clouds, for example.

I use a ND 256 for daytime hyperlapses.

Yes, I agree,
and for timelapses my favourite is the 256 or 512

 

[hiflyer201]

Yes, I agree,
and for timelapses my favourite is the 256 or 512

VIc, I wish you would do one of your fine video's on that subject or post one you have already done. Please