Are exposure meters in DX bodies calibrated differently from FX

michaeladawson · Mar 31, 2025

Hi all. I have a general question. I consider myself very experienced and knowledgeable, photography-wise. But every now and again, a question comes into my head that I'm suddenly puzzled by and can't answer. Today was one of those times.

Let's say I take an f/5.6 telephoto on an FX body. I add a 1.4X TC. The maximum f-stop is now reduced to f/8. To properly expose the same scene I now have to use a one stop slower shutter speed. I get why this is... you have basically enlarged the image circle such that half the light is no longer captured by the sensor.

Now I put that same f/5.6 telephoto on my DX camera body. The f-stop for exposure metering purposes is still f/5.6. But I'm "roughly" capturing the same scene as when the lens + TC is mounted on the FX body (Let's ignore the 2.0 vs 2.25 difference between a TC and a DX crop).

The only way this makes sense to me is if the DX and FX metering systems are calibrated differently so that f/5.6 delivers the same exposure (f-stop and shutter speed) on both formats.

Is this what is going on or am I missing something obvious?

SCoombs · Apr 1, 2025

Bill Ferris said:
michaeladawson said:

Hi all. I have a general question. I consider myself very experienced and knowledgeable, photography-wise. But every now and again, a question comes into my head that I'm suddenly puzzled by and can't answer. Today was one of those times.

Let's say I take an f/5.6 telephoto on an FX body. I add a 1.4X TC. The maximum f-stop is now reduced to f/8. To properly expose the same scene I now have to use a one stop slower shutter speed. I get why this is... you have basically enlarged the image circle such that half the light is no longer captured by the sensor.

Click to expand...

The non-TC'd lens captures the same number of photons from the subject as the TC'd lens. The image formed by the TC'd lens spreads that light over a larger area so, the exposure is reduced. But the SNR and noise level in the subject is the same in both photos.

Bill Ferris said:

Now I put that same f/5.6 telephoto on my DX camera body. The f-stop for exposure metering purposes is still f/5.6. But I'm "roughly" capturing the same scene as when the lens + TC is mounted on the FX body (Let's ignore the 2.0 vs 2.25 difference between a TC and a DX crop).

Click to expand...

The exposure is f/5.6 but the lens's entrance pupil is the same size as in the photos made with the full frame camera. So, the APS-C system captures the same number of photons from the subject and makes an image having the same SNR & noise level as the non-TC'd photo made with the full frame body. It's just been cropped in-camera by the APS-C sensor. Cropping doesn't reduce image lightness; photos don't get darker when cropped.

Thanks for asking a question In sure others have been pondering.

If this is correct, it would mean that the TC Fx, the non-TC Fx, and the non-TC Dx have the same SNR and noise, but I am not sure you intended to say this because you specifically said that the DX image would have the same SNR as the non-TC'ed Fx, which would seem to suggest it is different from the TC Fx. Did you intend to say that all three would be the same?

If so, one would wonder why anyone would ever use a TC, since it shouldn't make a difference, and yet we all know of good and knowledgable photographers who make use of TCs and have seen examples demonstrating that at least with more modern TCs the results with them can be superior to those obtained by cropping.

BasilG · Apr 1, 2025

SCoombs said:
If this is correct, it would mean that the TC Fx, the non-TC Fx, and the non-TC Dx have the same SNR and noise, but I am not sure you intended to say this because you specifically said that the DX image would have the same SNR as the non-TC'ed Fx, which would seem to suggest it is different from the TC Fx. Did you intend to say that all three would be the same?

If so, one would wonder why anyone would ever use a TC, since it shouldn't make a difference, and yet we all know of good and knowledgable photographers who make use of TCs and have seen examples demonstrating that at least with more modern TCs the results with them can be superior to those obtained by cropping.

My understanding is that the resolution benefits of using a TC are primarily due to the finite sampling of the sensor. If the lens used has resolution "headroom" (some people would say "out-resolves the sensor", but this is not quite right), you can access it by enlarging the image before it hits the sensor (which results in finer sampling of a given element of subject detail by the sensor). If the lens' resolving power is already mostly exploited by a given sensor even in the absence of a TC, all you get is a 1.4x (or 2x in the case of a 2x TC) drop in resolved LWPH, indicating that you might as well crop.

starbase218 · Apr 1, 2025

When applying a TC, the amount of light per square mm of your sensor is cut in half. When using a smaller sensor, the amount of light per square mm stays the same. But the number of square mm is less. This is why DX sensors have worse ISO noise performance than FX sensors.

Bob A L · Apr 1, 2025

Not necessarily. When operating a FX camera in DX mode, the noise situation should be the same as the same size pixels are being used, you are just only using part of them when in DX crop mode. All pixels on the sensor are all exposed to the same amount of light if focusing on a solid color target, and how many of them you use to make the photo does not matter. You just don't use the ones you don't want, this does not affect the light falling on the ones you do use in the photo. If a DX camera has a tighter pixel density, there is a chance of more noise than a FF sensor of the same mp rating but that is because of the difference in pixel size.

Bill Ferris · Apr 1, 2025

SCoombs said:
Bill Ferris said:

michaeladawson said:

Hi all. I have a general question. I consider myself very experienced and knowledgeable, photography-wise. But every now and again, a question comes into my head that I'm suddenly puzzled by and can't answer. Today was one of those times.

Let's say I take an f/5.6 telephoto on an FX body. I add a 1.4X TC. The maximum f-stop is now reduced to f/8. To properly expose the same scene I now have to use a one stop slower shutter speed. I get why this is... you have basically enlarged the image circle such that half the light is no longer captured by the sensor.

Click to expand...

The non-TC'd lens captures the same number of photons from the subject as the TC'd lens. The image formed by the TC'd lens spreads that light over a larger area so, the exposure is reduced. But the SNR and noise level in the subject is the same in both photos.

SCoombs said:

Now I put that same f/5.6 telephoto on my DX camera body. The f-stop for exposure metering purposes is still f/5.6. But I'm "roughly" capturing the same scene as when the lens + TC is mounted on the FX body (Let's ignore the 2.0 vs 2.25 difference between a TC and a DX crop).

Click to expand...

The exposure is f/5.6 but the lens's entrance pupil is the same size as in the photos made with the full frame camera. So, the APS-C system captures the same number of photons from the subject and makes an image having the same SNR & noise level as the non-TC'd photo made with the full frame body. It's just been cropped in-camera by the APS-C sensor. Cropping doesn't reduce image lightness; photos don't get darker when cropped.

Thanks for asking a question In sure others have been pondering.

Click to expand...

If this is correct, it would mean that the TC Fx, the non-TC Fx, and the non-TC Dx have the same SNR and noise, but I am not sure you intended to say this because you specifically said that the DX image would have the same SNR as the non-TC'ed Fx, which would seem to suggest it is different from the TC Fx. Did you intend to say that all three would be the same?

All three photos capture the same number of photons from the subject. As a result, the SNR and noise level in the subject portion of all three images is the same. The exposures aren't uniform but the light collected from the subject is.

While adding a TC magnifies & enlarges a central portion of the image, thus reducing exposure, it doesn't add more noise. Cropping doesn't alter the exposure needed to achieve a desired image lightness nor does it remove light from the remaining portion. So, again, no noise is added.

SCoombs said:
If so, one would wonder why anyone would ever use a TC, since it shouldn't make a difference, and yet we all know of good and knowledgable photographers who make use of TCs and have seen examples demonstrating that at least with more modern TCs the results with them can be superior to those obtained by cropping.

The advantage of a TC versus cropping is that the subject covers more pixels. You potentially get better resolution versus cropping. The result depends on the quality of the lens and of the TC. Among photographers who choose not to use a TC, it's often because - in their experience - the additional resolution is countered by the reduced image quality versus a bare lens. It's not everybody's experience but photographers who've seen that outcome often make that choice.

Interceptor121 · Apr 1, 2025

If you put the same lens on a cropped sensor and use the same exposure time and f stop you are changing the field of view so no longer capturing the same scene

The exposure density will stay the same but you are capturing a smaller area so the amount of light will be less

if you instead you capture the same area the focal length will change and with the same fstop and exposure time you will be projecting less light simply because the smaller format has a shorter lens and therefore the physical lens aperture is smaller even with the same field of view

The exposure meter works on density not total amount of light

starbase218 · Apr 1, 2025

Bob A L said:
Not necessarily. When operating a FX camera in DX mode, the noise situation should be the same as the same size pixels are being used, you are just only using part of them when in DX crop mode. All pixels on the sensor are all exposed to the same amount of light if focusing on a solid color target, and how many of them you use to make the photo does not matter. You just don't use the ones you don't want, this does not affect the light falling on the ones you do use in the photo.

That entirely depends on the magnification level you are looking at the photo. If it's on a monitor at 100%, for example, then both will have the same amount of noise. However, the DX photo will also be significantly smaller. If you're viewing both images at the same size on your monitor, and neither are magnified above 100%, then there will be a noise advantage for the FX photo, since there are more pixels in the photo, per pixel on your monitor. And so there is more information for the values of those pixels to be averaged out.

Bob A L said:
If a DX camera has a tighter pixel density, there is a chance of more noise than a FF sensor of the same mp rating but that is because of the difference in pixel size.

Pixel density (as in the size of the pixels) doesn't actually matter much. I mean, yes, smaller pixels will get you more detail, but also more noise at that detail level. But if you view the image at, say, 50%, the noise level drops, because again, there are more pixels in your photo for every pixel on your monitor.

Bob A L · Apr 1, 2025

If you have a FF camera it's easy to prove. Go out and take a photo of anything that is relatively evenly light and dark areas across the entire field or the proverbial brick wall for ease, take a photo with camera set at FF, then leave settings the same without moving the camera and take a photo with camera set to crop frame. The two photos should be identical concerning lightness/darkness (exposure to me). Only the field of view will change.

Leonard Shepherd · Apr 2, 2025

Rene Gr said:
indeed, a sensor with smaller pixels will get less light per pixel.

Will it ?:-O

The same amount of light falls on the top side of the micro lens over a pixel.

Smaller pixels affect noise but not exposure.

For those with a modestly long memory fine film grain increased resolution without changing the exposure.

A light reading equivalent to 1/125 at f8 gave the same exposure to both 35mm and half frame film cameras.

--
Leonard Shepherd
In lots of ways good photography is similar to learning to play a piano - it takes practice to develop skill in either activity.

BasilG · Apr 2, 2025

Leonard Shepherd said:
Rene Gr said:

indeed, a sensor with smaller pixels will get less light per pixel.

Click to expand...

Will it ? :-O

Per pixel? Yes.

In total? No (or not much).

Search

Are exposure meters in DX bodies calibrated differently from FX

michaeladawson

Forum Pro

SCoombs

Senior Member

BasilG

Forum Pro

starbase218

Senior Member

Bob A L

Forum Pro

Bill Ferris

Community Leader

Interceptor121

Forum Pro

starbase218

Senior Member

Bob A L

Forum Pro

Leonard Shepherd

Forum Pro

BasilG

Forum Pro

Similar threads

Keyboard shortcuts