Mavic 2 Pro Low Light Shot, "After glow in the City by the Bay"

[Bmroche]

Go full manual mode, ISO 100, F3.2-4.0 for sharpness, shutter speeds of .5-1 second depending on how bright your night subject is, white balance set manually to around 4000k but you should play with it to make sure colors come out the way it looks to the naked eye, shoot in RAW so that you can make adjustments post. Apply noise reduction in RAW to both luminance and color noise. I use around 50% for each. adjust saturation and color temperature to your liking.

Boom, what Steve said!

Look up some YouTube videos on the exposure triangle (iso, shutter speed, and apeture). These are the three setting you will tweak in manual mode, and always shoot in RAW (I use JPG and Raw, so the jpg for quick share but Raw will always look better for editing).

If you can't lower the setting low enough you probably need some ND Filters.

But for night shots like these, I would definitely recommend manual mode, gives you the most flexibility.

Not always, but almost always keep your iso as low as possible to prevent grain (can go up to ~1600 max at night) but I try to stay 100/200/400. And open up the apeture (smaller number) to let me light in. Then the third item to adjust is shutter speed, and that you can go long exposure if you want trails of smooth waves etc, but will just have to play around.

 

[SprayingMango]

Go full manual mode, ISO 100, F3.2-4.0 for sharpness, shutter speeds of .5-1 second depending on how bright your night subject is, white balance set manually to around 4000k but you should play with it to make sure colors come out the way it looks to the naked eye, shoot in RAW so that you can make adjustments post. Apply noise reduction in RAW to both luminance and color noise. I use around 50% for each. adjust saturation and color temperature to your liking.

Thank you!

 

[owg666]

Beautiful shots !!! Love night city pics. Well done !!!

 

[Meta4]

Wow these night shots are amazing! I am struggling with getting night shots. You guys are going full manual mode and setting it all yourself? I don't even know where to begin in some cases, its either so blown out or I can't adjust some settings low enough.

If you don't know your way around camera settings, it's easy to make a mess of things messing around in manual.
And it's so unnecessary when the camera has a perfectly good metering system that can do it for you.
There's nothing magic about shooting in manual.
A properly exposed image is a properly exposed image, whether you adjusted the settings manually or the camera worked them out for you.

If you are having trouble with settings, try using auto and get some good shots.
Then you can look at them and see what might be improved by varying the settings for the next time.

For night shooting set Aperture Priority and if the scene is dark, open up the aperture to f2.8.
If your scene is well lit you can set the ISO at 100 and still get good shutter speeds.
If things are very dark, you'll have to increase ISO but this will mean you get more noise in the images.

 

[SteveCA]

one from earlier this evening. 4 seconds at F5.0, ISO 100, shot in RAW, noise reduction applied, sharpened 100%.

 

[Bmroche]

one from earlier this evening. 4 seconds at F5.0, ISO 100, shot in RAW, noise reduction applied, sharpened 100%.View attachment 50985

Love the light trails on the bridge!

 

[SteveCA]

Love the light trails on the bridge!

Thanks, next time I’m going to use an ND 8 and shoot an 8 second exposure for longer light trails. Last night I shot some 8 second exposures but none were sharp using F8.

 

[JonnyD]

Is it my monitor, or is it the M2P? I find that most of these images are too warm ... ~700K too warm.

 

[CanadaDrone]

Go full manual mode, ISO 100, F3.2-4.0 for sharpness, shutter speeds of .5-1 second depending on how bright your night subject is, white balance set manually to around 4000k but you should play with it to make sure colors come out the way it looks to the naked eye, shoot in RAW so that you can make adjustments post. Apply noise reduction in RAW to both luminance and color noise. I use around 50% for each. adjust saturation and color temperature to your liking.

Just to clarify for the benefit of others, WB is not set in a RAW image - you can change it to whatever you want after the fact with a single click, so no need to fuss about WB if you don't want to at the time of the shot.

NR is completely unnecessary at ISO 100 if you expose correctly. All NR is going to do is degrade the image unnecessarily.

Luminance NR will destroy the image's detail, and it's better to have a slightly grainy but sharp image than a smeared one (especially if you ever print your photos), so try to stay away from Liminance NR as much as you can. At ISO 100 there is no need whatsoever for Chroma (color) or Luminance NR unless you massively screw up the shot and have to push the file so far in PP that is would be otherwise unusable. If you have to use it, apply it selectively to areas without detail such as the background or sky and not to the main subject or entire image. Also note that NR is not applied equally across programs, 50% in one might be completely different than 50% in another.

Is it my monitor, or is it the M2P? I find that most of these images are too warm ... ~700K too warm.

It's not your monitor, the image is very warm/saturated but it's probably just the look he's going for which looks OK for this type of image. It just comes down to personal taste.

 

[SteveCA]

Just to clarify for the benefit of others, WB is not set in a RAW image - you can change it to whatever you want after the fact with a single click, so no need to fuss about WB if you don't want to at the time of the shot.

NR is completely unnecessary at ISO 100 if you expose correctly. All NR is going to do is degrade the image unnecessarily.

Luminance NR will destroy the image's detail, and it's better to have a slightly grainy but sharp image than a smeared one (especially if you ever print your photos), so try to stay away from Liminance NR as much as you can. At ISO 100 there is no need whatsoever for Chroma (color) or Luminance NR unless you massively screw up the shot and have to push the file so far in PP that is would be otherwise unusable. If you have to use it, apply it selectively to areas without detail such as the background or sky and not to the main subject or entire image. Also note that NR is not applied equally across programs, 50% in one might be completely different than 50% in another.



It's not your monitor, the image is very warm/saturated but it's probably just the look he's going for which looks OK for this type of image. It just comes down to personal taste.

sorry but I beg to differ. If you shoot long exposure photos with this camera, you will find that even ISO 100 photos exhibit lots of noise. and NR done properly does not destroy detail or sharpness. If you believe that ISO 100 photos don’t need NR, you simply haven’t shot long exposure night shots. NR when taken too far absolutely destroys detail but not when it is done manually with the photo viewed at 100%.

The NR settings I referenced are for camera RAW in photoshop.

I can easily post pre and post NR 100% crops that would 1. prove the need for NR and 2. prove that NR both luminance and color, do not destroy detail when done properly in photoshop camera RAW.

The bottom line is this. Many theories and beliefs are sometimes proven wrong in actual experience.

Until you’ve actually gone out and done it, they are just that, theories.

 

[SteveCA]

here are some shots from last night which was quite windy.

 

[CanadaDrone]

sorry but I beg to differ. If you shoot long exposure photos with this camera, you will find that even ISO 100 photos exhibit lots of noise. and NR done properly does not destroy detail or sharpness. If you believe that ISO 100 photos don’t need NR, you simply haven’t shot long exposure night shots. NR when taken too far absolutely destroys detail but not when it is done manually with the photo viewed at 100%.

The NR settings I referenced are for camera RAW in photoshop.

I can easily post pre and post NR 100% crops that would 1. prove the need for NR and 2. prove that NR both luminance and color, do not destroy detail when done properly in photoshop camera RAW.

The bottom line is this. Many theories and beliefs are sometimes proven wrong in actual experience.

Until you’ve actually gone out and done it, they are just that, theories.

Something else must be going on then if your output is so bad that it requires heavy luminance NR - probably just a settings or technique issue. I have shot literal tens of thousands of long exposures at night with everything from a Sony RX100 (same sensor as M2P), numerous APS-C DSLRs, and numerous FF DSLRs. I am guessing I have more than enough experience in this area - so I would politely ask you not to tell me what I have and have not done Further to this, I shoot real estate photography professionally, which often involves long exposures in low lighting (much longer than the ones you are taking). If you expose the shot properly you should not be seeing noise at base ISO bad enough that you need NR. My comments are not based on 'theory' but on years of experience in the professional photography realm and the physics of digital imaging.

Can I see a 100% crop of one of the photos you posted above? I'd like to see how you managed to use Lumanance NR without destroying any detail. At least looking a those web-sized images, they already look like they have heavy NR applied unless they are just blurry from drone movement. I like most of your images by the way - please don't take any of this the wrong way as I am only trying to help.

Here is an example of a much longer exposure than anything above taken with an 8-year-old point & shoot camera, and there is virtually zero noise in the deepest shadows with no NR of any type:

 

[SteveCA]

Something else must be going on then if your output is so bad that it requires heavy luminance NR - probably just a settings or technique issue. I have shot literal tens of thousands of long exposures at night with everything from a Sony RX100 (same sensor as M2P), numerous APS-C DSLRs, and numerous FF DSLRs. I am guessing I have more than enough experience in this area - so I would politely ask you not to tell me what I have and have not done Further to this, I shoot real estate photography professionally, which often involves long exposures in low lighting (much longer than the ones you are taking). If you expose the shot properly you should not be seeing noise at base ISO bad enough that you need NR. My comments are not based on 'theory' but on years of experience in the professional photography realm and the physics of digital imaging.

Can I see a 100% crop of one of the photos you posted above? I'd like to see how you managed to use Lumanance NR without destroying any detail. At least looking a those web-sized images, they already look like they have heavy NR applied unless they are just blurry from drone movement. I like most of your images by the way - please don't take any of this the wrong way as I am only trying to help.

Here is an example of a much longer exposure than anything above taken with an 8-year-old point & shoot camera, and there is virtually zero noise in the deepest shadows with no NR of any type:

Something else must be going on then if your output is so bad that it requires heavy luminance NR - probably just a settings or technique issue. I have shot literal tens of thousands of long exposures at night with everything from a Sony RX100 (same sensor as M2P), numerous APS-C DSLRs, and numerous FF DSLRs. I am guessing I have more than enough experience in this area - so I would politely ask you not to tell me what I have and have not done Further to this, I shoot real estate photography professionally, which often involves long exposures in low lighting (much longer than the ones you are taking). If you expose the shot properly you should not be seeing noise at base ISO bad enough that you need NR. My comments are not based on 'theory' but on years of experience in the professional photography realm and the physics of digital imaging.

Can I see a 100% crop of one of the photos you posted above? I'd like to see how you managed to use Lumanance NR without destroying any detail. At least looking a those web-sized images, they already look like they have heavy NR applied unless they are just blurry from drone movement. I like most of your images by the way - please don't take any of this the wrong way as I am only trying to help.

Here is an example of a much longer exposure than anything above taken with an 8-year-old point & shoot camera, and there is virtually zero noise in the deepest shadows with no NR of any type:

I too am just trying to be helpful to others so lets keep it constructive. I too have shot thousands of low light, long exposure photo going back long before digital. I've owned and used everything from the venerable Nikon F all the way forward to the first digital worthy of low light work, my Nikon D3s, to the present with the nikon D800 series cameras.

Most ISO 100 photos exhibit little to no noise, but not the Mavic 2 pro. I was suggesting that you have not shot much long exposures with the Mavic 2 pro otherwise, you would see significant luminance noise in those shots, unless of course I got a bad copy of the camera, which is always possible.

In any event. I suggest any interested parties go out and shoot some low light long exposure shots and look at the photos on a good monitor at 100% and see what I'm talking about.

Later when I have the time, I will post 100% crops for all to see that prove my points. In the end, if there is no noise to your photos, obviously dont use NR, it would be pointless. If you do see noise and want to reduce it without destroying detail, it can be done manually and it will not destroy detail if you go gently with luminance and color NR.

Lastly, I don't buy blanket statements made by "photographers" as most such statements have and will continue to be proven wrong. I've been shooting long enough to learn that you never know for sure until you've gone out and done it. I am always learning and have an open mind to new techniques as well as theories. I've had a number of very strong beliefs in the photography world that have later been proven wrong or techniques learned that obviate earlier beliefs.

 

[CanadaDrone]

I too am just trying to be helpful to others so lets keep it constructive. I too have shot thousands of low light, long exposure photo going back long before digital. I've owned and used everything from the venerable Nikon F all the way forward to the first digital worthy of low light work, my Nikon D3s, to the present with the nikon D800 series cameras.

What about the D3/D700 made them unworthy of low light work?

Most ISO 100 photos exhibit little to no noise, but not the Mavic 2 pro. I was suggesting that you have not shot much long exposures with the Mavic 2 pro otherwise, you would see significant luminance noise in those shots, unless of course I got a bad copy of the camera, which is always possible.

I own a camera with the identical sensor - even if I shoot a poor quality 8bit JPEG with NR off I can get clean long exposures with the correct settings. Maybe something is indeed wrong with your copy? Not even the Mavic Air and it's crappy 1/2.3" sensor gets noisy enough to warrant that much NR at base ISO and a proper long exposure. Can you replicate that same exposure duration with a ND filter in the day time? That might be an interesting experiment.

If you do see noise and want to reduce it without destroying detail, it can be done manually and it will not destroy detail if you go gently with luminance and color NR.

You cannot apply luminance NR and not destroy any detail. I will try to remember to post an example of this later. I am guessing that your threshold for acceptable detail loss is simply above zero, which is fine, but it is not accurate to say that you can use that type of NR and not lose any detail.

Lastly, I don't buy blanket statements made by "photographers" as most such statements have and will continue to be proven wrong. I've been shooting long enough to learn that you never know for sure until you've gone out and done it. I am always learning and have an open mind to new techniques as well as theories. I've had a number of very strong beliefs in the photography world that have later been proven wrong or techniques learned that obviate earlier beliefs.

One of the best parts about photography is that the technical side of it is entirely based on physics with objectively measurable variables. That is in fact part of what got me interested in it in the first place. There is nothing arbitrary or theoretical about things you can objectively measure, such as SNR, PDR, well capacities, etc. especially when these things translate so directly into what we see with our eyes. The exact same principles apply whether the camera is 400ft up on a drone or 100ft under the ocean in a dive housing - again something that I think makes photography so interesting. These are not 'beliefs' or 'theories'.

Lastly, I don't buy blanket statements made by "photographers" as most such statements have and will continue to be proven wrong.

So much for keeping it constructive Kind of insulting don't you think?

Anyways you have me genuinely curious - what blanket statements surrounding the physics of light collection and digital sensor behavior have you found to be proven wrong? The principles of photography are very much rooted in science and there is a stark difference between believing a perpetuated myth regarding a setting or something, and the former.

 

[SteveCA]

What about the D3/D700 made them unworthy of low light work?



I own a camera with the identical sensor - even if I shoot a poor quality 8bit JPEG with NR off I can get clean long exposures with the correct settings. Maybe something is indeed wrong with your copy? Not even the Mavic Air and it's crappy 1/2.3" sensor gets noisy enough to warrant that much NR at base ISO and a proper long exposure. Can you replicate that same exposure duration with a ND filter in the day time? That might be an interesting experiment.



You cannot apply luminance NR and not destroy any detail. I will try to remember to post an example of this later. I am guessing that your threshold for acceptable detail loss is simply above zero, which is fine, but it is not accurate to say that you can use that type of NR and not lose any detail.



One of the best parts about photography is that the technical side of it is entirely based on physics with objectively measurable variables. That is in fact part of what got me interested in it in the first place. There is nothing arbitrary or theoretical about things you can objectively measure, such as SNR, PDR, well capacities, etc. especially when these things translate so directly into what we see with our eyes. The exact same principles apply whether the camera is 400ft up on a drone or 100ft under the ocean in a dive housing - again something that I think makes photography so interesting. These are not 'beliefs' or 'theories'.



So much for keeping it constructive Kind of insulting don't you think?

Anyways you have me genuinely curious - what blanket statements surrounding the physics of light collection and digital sensor behavior have you found to be proven wrong? The principles of photography are very much rooted in science and there is a stark difference between believing a perpetuated myth regarding a setting or something, and the former.

you’ve made a number of blanket statements in these forums. do you really want to be called out on them? would that be helpful to anyone reading them? I spend my time flying and capturing photos, but if I wanted to be a keyboard warrior, and spend my time arguing, well you get the idea.

 

[CanadaDrone]

you’ve made a number of blanket statements in these forums. do you really want to be called out on them? would that be helpful to anyone reading them? I spend my time flying and capturing photos, but if I wanted to be a keyboard warrior, and spend my time arguing, well you get the idea.

If I have made any statements that you don't simply disagree with but can demonstrate are factually untrue, yes I would like to know - that would benefit both myself and those reading. I assume you would appreciate the same courtesy. We are all here to learn together. Feel free to use PM if you prefer.

I thought we were having a friendly discussion - apologies if it came off as anything other than that. One post ago you made the comment "lets keep it constructive" which I interpreted as an invitation to continue the friendly dialogue - I think that was a reasonable assumption. I asked you questions I was genuinely curious about and suggested a test that may help determine if you have a defective camera unit or not - all within the spirit of these forums. If I am being honest, your insult caught me off guard and if my tone changed as a result I apologize for that also.

Just because I said I would, and I think a NR demonstration can benefit others as well, here is an image I took with and without 50% manually applied Luminance NR in Photoshop. RAW image straight out of the camera, only thing I did to it was either apply or not apply NR and save as JPEG for viewing. I purposely picked an image with fine detail to illustrate how destructive NR is:

 

[JonnyD]

I didn't have much luck with NR either. As seen (1:1) quite a bit of detail was lost. But if I don't pixel peep, it's good enough for web work.

 

[MavicPro2]

I didn't have much luck with NR either. As seen (1:1) quite a bit of detail was lost. But if I don't pixel peep, it's good enough for web work.
View attachment 51402View attachment 51403

Great shot but I was getting my snacks and beer ready for the big fight so hold off any more posts please

 

[SteveCA]

Here are two screen shots from a night shot taken at ISO 100, 3.2 second exposure. you can see the noise reduction settings on both and decide for yourself.

I am certain that NR can destroy detail, there is no argument there. But it can be done in a way to limit that to serious pixel peeping at serious magnifications.

In any event, I am not interested in arguing, just trying to be helpful and show that our Mavic 2 does indeed have noise at low ISO's and that it can be fixed without destroying detail (not zero mind you, but not enough to hurt the image)

Cheers

 

[SteveCA]

Here is the full image that these crops were taken from