DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Hugely underexposed in desert with ND8 on auto expose. How did this happen?

rickray

Active Member
Joined
Feb 13, 2017
Messages
32
Reactions
38
Age
64
Hi there,

I'm new to the ND world but just did a shoot under intense bright sunlight at midday in the desert near Palm Springs. Thinking I could compensate for the bright sunlight, I added a Polar Pro ND8 for the shoot, but left the exposure settings on automatic.

Normally I try to get a good exposure setting on auto expose and once I get up in the air, I find a happy medium between sky and land, then lock the exposure so I don't get any major shifts in exposure during a flight. I rarely want to be worrying about exposure while all the other factors involved in flying are at play.

I flew mostly land with a little sky and locked my exposure to that.

I was amazed to get home and see my footage is HORRIFIC. It's extremely dark, almost unusably so, and very very grainy. In the midday sun of Palm Springs!

I swear I never was in manual exposure mode, only auto, and was just using the "locking" icon in the far upper right corner of the DJI4 app to lock in exposure. I'm saddened that the desert was so bright that I could not see how dark it was turning out, even with a hood on.

But can anyone explain to me how in bright desert conditions, I got this result? Should I never used manual exposure with an ND filter? I'm normally a good pilot but this really threw me and the results are just awful. Shouldn't my problem be OVEREXPOSURE, not UNDEREXPOSURE? What mistake did I make? I don't want to repeat it again.
 
  • Like
Reactions: Chirp
a, turn on histogram so that you can keep your eye on exposure regardless of ambient conditions
b, you sure your EV value wasn't accidentally set to e.g -2.0EV?
 
  • Like
Reactions: truckstop
Exposure is the spin wheel thingy on your right on the controller. Never set to auto.
You should have ISO100 (lowest) and exposure 1:(your video FPS)*2
EDIT: doesn't seem like the xposure matters
 
Last edited by a moderator:
Exposure is the spin wheel thingy on your right on the controller. Never set to auto.
You should have ISO100 (lowest) and exposure 1:(your video FPS)*2

That's hugely misleading.. you are removing all variables so you would inevitable end up with wrong exposure. There's nothing wrong with Auto in most cases, it will certainly do a better job than what you're suggesting...

Honestly, the whole 180 degree rule dogma is rarely ever important for drone footage where you typically shoot static landscapes from way up in the air...
 
  • Like
Reactions: MavicCF and Kilrah
Auto would be fine, but you say you locked it to the sky - so not able to be auto anymore.
Having the ND on was not relevant to the underexposure.

There is nothing wrong with using Auto just keep an eye on the result you are getting and adjust EV with the wheel if required.
 
  • Like
Reactions: Chirp
That's hugely misleading.. you are removing all variables so you would inevitable end up with wrong exposure. There's nothing wrong with Auto in most cases, it will certainly do a better job than what you're suggesting...

Honestly, the whole 180 degree rule dogma is rarely ever important for drone footage where you typically shoot static landscapes from way up in the air...
Interesting, i've been told its a rule for good quality footage
And i have quite bad experience with auto exposure on MP
 
Set to manual. Use D-Log and manual white balance (so that is locked too). Set to lowest possible ISO. Set sharpness to +1. Use the histogram to find a good compromis for exposure for the whole flight.
Adjust colour, saturation and sharpness in post with a decent program like PremierePro, FCPro or Resolve.
With D-Log you have the most information in your data and can play a lot in post, even (not too much) underexposed footage usually can be corrected that way.

The 180 rule is only to get cinematic blur that omits the shadows of the props and exposes at least half of the frame length. Without ND filter usually the frame is much longer than the part that is exposed, leaving a black gap between all the frames, inducing strobe flicker.
 
  • Like
Reactions: Babaco
ND Filters only allow for the correct amount of motion blur (cinematic footage), unless your filming a moving subject, ND filters have no use.

The histogram is far more useful and helpful 99% of the time.

Have a look at this video;

 
unless your filming a moving subject, ND filters have no use
Only when you take a static shot while hovering, without any motion. If you fly, everything in the frame is moving. Also, the motion blur helps to get rid of prop shadows in some lighting situations. And even while hovering and shooting without moving, the frames will have black gaps between them if the shutter speed is higher than 2x the frame rate.

But I already mentioned that in different wording.
 
Only when you take a static shot while hovering, without any motion. If you fly, everything in the frame is moving. Also, the motion blur helps to get rid of prop shadows in some lighting situations. And even while hovering and shooting without moving, the frames will have black gaps between them if the shutter speed is higher than 2x the frame rate.

But I already mentioned that in different wording.
I've been shooting and editing professionally for over a decade. Your black frame analogy is simply untrue. Each frame is, in essence, a photograph. Once the image is captured - no matter how high or low the shutter speed - that image is applied to the entire frame. There are no black gaps. While the 180 rule is a nice way to get motion blur, it's an unnecessary measure for most footage. I think folks have made it too important - when it's not.
 
I've been shooting and editing professionally for over a decade. Your black frame analogy is simply untrue. Each frame is, in essence, a photograph. Once the image is captured - no matter how high or low the shutter speed - that image is applied to the entire frame. There are no black gaps. While the 180 rule is a nice way to get motion blur, it's an unnecessary measure for most footage. I think folks have made it too important - when it's not.
Go ahead, have your opinion. It's your right. I won't bother you, no worries.
 
Go ahead, have your opinion. It's your right. I won't bother you, no worries.
That's lovely ~ if indeed it were just my opinion. It is, rather, exactly how video works. Under your theory, shooting at 30fps, with a shutter speed of 1/300 would render 9/10ths of each frame black. Sorry, my friend, it just doesn't work that way. The exposed image is applied to the entire frame.
 
I've been shooting and editing professionally for over a decade. Your black frame analogy is simply untrue. Each frame is, in essence, a photograph. Once the image is captured - no matter how high or low the shutter speed - that image is applied to the entire frame. There are no black gaps. While the 180 rule is a nice way to get motion blur, it's an unnecessary measure for most footage. I think folks have made it too important - when it's not.
Video Camera Shutter vs Frame Rate

There you go.
 
Last edited:
That's lovely ~ if indeed it were just my opinion. It is, rather, exactly how video works. Under your theory, shooting at 30fps, with a shutter speed of 1/300 would render 9/10ths of each frame black. Sorry, my friend, it just doesn't work that way. The exposed image is applied to the entire frame.
copied from mediacollege.com:
Think of the "missing time" as a gap between when the shutter closes and when the next frame begins.
 
  • Like
Reactions: truckstop
Sorry, my friend, it just doesn't work that way.
Please show me proof of how it works in your opinion my friend.
I'm always happy to learn, really.
 
Please show me proof of how it works in your opinion my friend.
I'm always happy to learn, really.
Imagine film - each frame passes by the lens and the shutter triggers for the appropriate amount of time, based on lighting and required exposure. The entire frame of film is "exposed", and imprinted with the image taken, even if the shutter was only open for 1/1000th of a second. As a single frame is the smallest element in video footage - and **is [essentially] just a single photograph** - it cannot be both fully exposed, and partially black at the same time. It's just one single image, that cannot be split - as there is nowhere to store the additional information (blackness). The image captured for that frame is assigned to the entire frame. Again, under your theory - a video camera shooting at 1/1000th of a second, would leave you with 97% of your video as pure black. You wouldn't be able to see anything with only 3% of each frame exposed. Having edited actual film stock, I can testify that no frames contain any trailers of black. Each frame is fully exposed.
 
Imagine film - each frame passes by the lens and the shutter triggers for the appropriate amount of time, based on lighting and required exposure. The entire frame of film is "exposed", and imprinted with the image taken, even if the shutter was only open for 1/1000th of a second. As a single frame is the smallest element in video footage - and **is [essentially] just a single photograph** - it cannot be both fully exposed, and partially black at the same time. It's just one single image, that cannot be split - as there is nowhere to store the additional information (blackness). The image captured for that frame is assigned to the entire frame. Again, under your theory - a video camera shooting at 1/1000th of a second, would leave you with 97% of your video as pure black. You wouldn't be able to see anything with only 3% of each frame exposed. Having edited actual film stock, I can testify that no frames contain any trailers of black. Each frame is fully exposed.
Again, I don't want to debate about stuff that is open for any opinion. You keep talking about my opinion, but it just pure fact.
I edited celluloid film with hand wound viewers many years ago. Turning the wheel slow (slow frame rate), one could see a strobe effect, while turning it fast (fast frame rate) it would be fluid. Hence the gaps between the frames were to small to be noticed. Now if the frame time (slow frame rate, turning the wheel slowly) is longer than the shutter speed, there's a relatively big gap until the next frame begins. Hence even bigger strobe effect with lower speeds(frame rate). Simple. You say all you want, but that doesn't change it.

A higher frame rate allows for faster shutter speeds....again, simple once you get your head around the basics.

The way you say it defies the science of filming. You made it up yourself. Again, show me your proof. Show me one website with reputation that confirms your theory. Please. People could get confused reading this discussion.

Have you read the info on mediacollege.com I posted at all?
 
Not wanting to get into the debate, but the histogram is the tool I use to confirm my lighting. It is sometimes difficult to tell with the display only.
 
Not wanting to get into the debate, but the histogram is the tool I use to confirm my lighting. It is sometimes difficult to tell with the display only.
Agreed! The histogram is the best tool to use indeed. The FPV image is not representative at all in most cases as most of us find out sooner or later.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Members online

Forum statistics

Threads
130,599
Messages
1,554,238
Members
159,603
Latest member
refrigasketscanada