How to stop exposure bias (EV)?

[Chopstix Kid]

I was out today testing the @fstop.labs 3-pack ND filter kit and when I got home to review the results, I noticed that the exposure bias adjusted automatically with each change in either aperture or shutter speed even though I was in full manual mode. I didn't event think to look at it while shooting as I assumed it would stay at my initial setting of 0. Any thoughts? Cheers.

 

[WithTheBirds]

If you were in manual mode displayed EV will, and must, change with any adjustments you make to AV, TV or ISO. It is simply informing you what the difference is between your settings and what the light metering has determined to be optimum exposure, i.e. EV=O. The camera doesn’t know you have an ND on the lens, only the amount of light reaching the sensor.

 

[Meta4]

I was out today testing the @fstop.labs 3-pack ND filter kit and when I got home to review the results, I noticed that the exposure bias adjusted automatically with each change in either aperture or shutter speed even though I was in full manual mode. I didn't event think to look at it while shooting as I assumed it would stay at my initial setting of 0. Any thoughts? Cheers.

There's nothing to stop because it's not changing anything.
Think of it as a meter needle that's just showing you how far off EV0 the meter thinks your settings are when the scene/lighting changes.

 

[CanadaDrone]

I was out today testing the @fstop.labs 3-pack ND filter kit and when I got home to review the results, I noticed that the exposure bias adjusted automatically with each change in either aperture or shutter speed even though I was in full manual mode. I didn't event think to look at it while shooting as I assumed it would stay at my initial setting of 0. Any thoughts? Cheers.

It's because any camera can only expose for one thing at a time. The exposure isn't actually changing in this instance (if you're in Manual mode), it's just telling you that your image is either over or under exposed by the amount shown. When you set an exposure with Manual mode, the camera can only expose for one brightness level - as the scene changes (e.g. a pan from the ground to the sky) there is going to be a significant exposure difference. All you can do is try to balance it as well as you can, or try to shoot more when the light is very even. You can also use graduated ND filters but they require a lot more effort. Best to air on the side of underexposure as well, shadows are much easier to recover than highlights - especially with the low DR of these sensors.

You can set the exposure to whatever you want in Manual mode, the camera's meter works independently, and will report how far you are under/over what it thinks is an ideal average exposure given the scene at all times. This is measured in EV.

 

[Chopstix Kid]

If you were in manual mode displayed EV will, and must, change with any adjustments you make to AV, TV or ISO. It is simply informing you what the difference is between your settings and what the light metering has determined to be optimum exposure, i.e. EV=O. The camera doesn’t know you have an ND on the lens, only the amount of light reaching the sensor.

Soooooo... does that mean it is metering ONLY and NOT actually affecting the exposure?

 

[CanadaDrone]

Soooooo... does that mean it is metering ONLY and NOT actually affecting the exposure?

Yes - they are separate functions. In Manual mode it is there to help you maintain something close to 0.0EV if that is the desired exposure. Meters also aren't perfect - they do their best to average out the entire scene's exposure and that is why you should always look at the histogram instead.

If you're flying along pointed down into a dark forest at 0.0EV in Manual mode, and then tilt the camera up to the sky, it might say +5.0 EV, and your image might be badly overexposed now, but the exposure components (Shutter, Aperture, ISO) have not changed because you are in Manual mode.

If I were you I would ignore that entirely and always use the histogram which is a far more accurate representation of scene exposure.

 

[Chopstix Kid]

What I am trying to do is test the accuracy of the ND filters. I started with no filter and then shot the same still with each of the ND8, ND16 and ND32, manually moving the stops to counter act the filter to see if the exposure would remain constant. All 4 photos were shot within a 10 minute window at dusk.

With no filter, the EV was -1.3 ISO100
ND8 EV -.7 ISO 800
ND16 EV -1 ISO 1600
ND32 EV -1.3 ISO 3200

Does this mean that the ND8 and ND16 filters are not true or could there be other factors? Cheers.

 

[Chopstix Kid]

Yes - they are separate functions. In Manual mode it is there to help you maintain something close to 0.0EV if that is the desired exposure. Meters also aren't perfect - they do their best to average out the entire scene's exposure and that is why you should always look at the histogram instead.

If you're flying along pointed down into a dark forest at 0.0EV in Manual mode, and then tilt the camera up to the sky, it might say +5.0 EV, and your image might be badly overexposed now, but the exposure components (Shutter, Aperture, ISO) have not changed because you are in Manual mode.

If I were you I would ignore that entirely and always use the histogram which is a far more accurate representation of scene exposure.

Ok I just read this after my post above. That makes sense. My positioning was not 100% consistent so that could be the reason for the sway in EV.

 

[WithTheBirds]

Soooooo... does that mean it is metering ONLY and NOT actually affecting the exposure?

I think you have confused yourself with your own "exposure bias" definition and what you thought it might mean.

EV (exposure value) is simply the value determined by a particular combination of aperture (AV) and shutter speed (TV) setting. A simple reference so you might alter either or both to arrive at a the save EV.

The number you are seeing is simply the difference between what the camera has determined to be the optimum EV and that determined by your current AV, TV and ISO.

 

[CanadaDrone]

What I am trying to do is test the accuracy of the ND filters. I started with no filter and then shot the same still with each of the ND8, ND16 and ND32, manually moving the stops to counter act the filter to see if the exposure would remain constant. All 4 photos were shot within a 10 minute window at dusk.

With no filter, the EV was -1.3 ISO100
ND8 EV -.7 ISO 800
ND16 EV -1 ISO 1600
ND32 EV -1.3 ISO 3200

Does this mean that the ND8 and ND16 filters are not true or could there be other factors? Cheers.

ND filters are actually very simple once you learn the terminology:

ND2 = 1 stop reduction
ND4 = 2 stop reduction
ND8 = 3 stop reduction
ND16 = 4 stop reduction and so on

All else equal (ISO and Aperture), every step up in ND filter will half your shutter speed.

For example if the exposure of the scene is 1/1000, F2.8, ISO 100 with a naked lens, leaving aperture and ISO alone:

ND4 will drop you to 1/250 (4 times slower)
ND8 will drop you to 1/125 (8 times slower)
ND16 will drop you to 1/60 (16 times slower)

And so on and so on.

You need to set up a controlled scenario if you want to test the accuracy of the filters - pick only one variable in the exposure triangle to manipulate.

On your Mavic pro, you also have aperture to play with, which changes things - you can get away with far fewer ND filters because you have another variable at your disposal - just don't go beyond about F4 as diffraction will begin to degrade the image at really small (high F number) apertures.

The other way you can test it is leave the exposure alone and just change the filters in manual mode - you should see the EV drop by the exact same ratio as above (naked lens at 0.0EV should show -2.0EV with a ND4, and so on - you will probably hit the limits of the meter before you run out of ND filters though).

 

[Chopstix Kid]

I think you have confused yourself with your own "exposure bias" definition and what you thought it might mean.

EV (exposure value) is simply the value determined by a particular combination of aperture (AV) and shutter speed (TV) setting. A simple reference so you might alter either or both to arrive at a the save EV.

The number you are seeing is simply the difference between what the camera has determined to be the optimum EV and that determined by your current AV, TV and ISO.

Yep I think I have it now. You gents are awesome. Thank you!

 

[WithTheBirds]

What I am trying to do is test the accuracy of the ND filters. I started with no filter and then shot the same still with each of the ND8, ND16 and ND32, manually moving the stops to counter act the filter to see if the exposure would remain constant. All 4 photos were shot within a 10 minute window at dusk.

With no filter, the EV was -1.3 ISO100
ND8 EV -.7 ISO 800
ND16 EV -1 ISO 1600
ND32 EV -1.3 ISO 3200

Does this mean that the ND8 and ND16 filters are not true or could there be other factors? Cheers.

Manually moving the stops? what were you changing? ISO?

 

[Chopstix Kid]

ND filters are actually very simple once you learn the terminology:

ND2 = 1 stop reduction
ND4 = 2 stop reduction
ND8 = 3 stop reduction
ND16 = 4 stop reduction and so on

All else equal (ISO and Aperture), every step up in ND filter will half your shutter speed.

For example if the exposure of the scene is 1/1000, F2.8, ISO 100 with a naked lens, leaving aperture and ISO alone:

ND4 will drop you to 1/250 (4 times slower)
ND8 will drop you to 1/125 (8 times slower)
ND16 will drop you to 1/60 (16 times slower)

And so on and so on.

You need to set up a controlled scenario if you want to test the accuracy of the filters - pick only one variable in the exposure triangle to manipulate.

On your Mavic pro, you also have aperture to play with, which changes things - you can get away with far fewer ND filters because you have another variable at your disposal - just don't go beyond about F4 as diffraction will begin to degrade the image at really small (high F number) apertures.

The other way you can test it is leave the exposure alone and just change the filters in manual mode - you should see the EV drop by the exact same ratio as above (naked lens at 0.0EV should show -2.0EV with a ND4, and so on - you will probably hit the limits of the meter before you run out of ND filters though).

My only variables were ISO and my accuracy in getting the M2P in the same position each time. Because of the time of day, I ran each shot f/2.8 with a 1/200 shutter. As you can see, I doubled the ISO each time since I had nowhere to go with the other two.

 

[Chopstix Kid]

Manually moving the stops? what were you changing? ISO?

Yes. See post just above... because of time of day, I had no room to move aperture and/or shutter. I am thinking that I could probably shoot a white wall to see if the metering is consistent across the filters?

 

[WithTheBirds]

Yes. See post just above... because of time of day, I had no room to move aperture and/or shutter. I am thinking that I could probably shoot a white wall to see if the metering is consistent across the filters?

You didn’t make it clear what you were changing. That’s not important.

Easiest way to perform your desired test would be to use aperture priority mode with the ISO fixed. You might then observe the light attenuation offered by the various ND filters by noting the camera assigned shutter speed.

This test will probably teach you more about basic camera settings and their relationships than it might about the filters, I would be amazed if they are more than 1/4 stop away from their advertised value.

Let us know how you go.