[Marc Sabatella]

Unfortunately, an APS-C dSLR doesn't get equivalent exposure to a FF at the same aperture, since the smaller sensor loses a stop of light gathering ability. So, exposure on the APS-C at an aperture of f 2.8 is equivalent to f 4 on FF. Ergo, you must ALSO increase the ISO by a stop on APS-C to take your equivalent DOF/AOV/Shutter Time picture, given the same amount of light on subject.

You might want to think that one through again. Not sure what you meant by "exposure" or "equivalent" in the above, but the sense in which an exposure of f/2.8 on APS-C is equivalent to an exposure of f/4 on FF is that it collects the same actual amount of light for the same shutter speed, and the fact that you are shooting f/2.8 instead of f/4 means you shoot at an ISO setting *less* for APS-C than FF in order to get tha same shutter speed. That's why it works out to be about the same level of noise, because indeed, if you had shot at the *same* ISO, the FF shot would have had *less* noise.

This is easy enough to test; take some "Sunny 16" images with your APS-C dSLR of a front-lit subject on a bright sunny day at the appropriate aperture/shutter/ISO combination (e.g., ISO 200, 1/250 @ f 16 or 1/500 @ f 11 or 1/1000 @ f 8; ISO 400, 1/500 @ f 16, 1/1000 @ f 11, 1/2000 @ f 8). Your pictures will be dark. Increase exposure by one stop, and they'll be exposed spot-on as they would be on slide film (or, assuming accurate ISO values, on a FF dSLR).

This simply isn't true. I mean, any particular APS-C camera might happen to overstate it's sensitivity ratings, meaning it will shoot a bit darker than it should for a given ISO/aperture/shutter speed combo, but it's not an inherent feature of APS-C, but it's not an inherent feature of APS-C. And if it's overstating its sensitivity (so setting ISO 800 really gives you only ISO 700, say), then it's also giving you the slightly less *noise* of ISO 700. Meaning you really aren't paying any particular penalty here. Chances are, any given FF camera will be just as likely to somewhat overstate its sensitivity. You aren't really going to see a difference between APS-C and FF in terms of the exposure you get at a given ISO, aperture, and shutter speed.

So your argument here is based on a flawed assumption. A given ISO, aperture, and shutter speed really 8does* produce the same exposure on both APS-C and FF. With that in mind, I'd suggest reading through the entire thread. You'll find quite a lot of useful information was hashed out there. Not "proving" conclusively that one format is "better" or "worse" than the other, but providing new and more accurate perspectives on just what the differences really entail. Fascinating stuff, really.

[jogiba]

Great explaination, thanks!

Add to it the legacy of the registration (or mount hole) diameter. As studied in another thread, the PK mount is limited to f/1.2, some other SLR mounts to f/1.1 and some range finder mounts to f/0.95.

So, a PK mount 50/1.4 is possible while a 30mm/1.0 isn't. The excellent PK mount 30/1.4 from Sigma may be as good as it gets.

Canon 50mm F1.0

[falconeye]

Canon 50mm F1.0

jogiba, this topic was feeding another long thread already
So, for "some other SLR mounts" f/1.0 is possible, i.e., for Canon EF. Thanks for the correction.

[24X36NOW]

You might want to think that one through again. Not sure what you meant by "exposure" or "equivalent" in the above, but the sense in which an exposure of f/2.8 on APS-C is equivalent to an exposure of f/4 on FF is that it collects the same actual amount of light for the same shutter speed, and the fact that you are shooting f/2.8 instead of f/4 means you shoot at an ISO setting *less* for APS-C than FF in order to get tha same shutter speed. That's why it works out to be about the same level of noise, because indeed, if you had shot at the *same* ISO, the FF shot would have had *less* noise.



This simply isn't true. I mean, any particular APS-C camera might happen to overstate it's sensitivity ratings, meaning it will shoot a bit darker than it should for a given ISO/aperture/shutter speed combo, but it's not an inherent feature of APS-C, but it's not an inherent feature of APS-C. And if it's overstating its sensitivity (so setting ISO 800 really gives you only ISO 700, say), then it's also giving you the slightly less *noise* of ISO 700. Meaning you really aren't paying any particular penalty here. Chances are, any given FF camera will be just as likely to somewhat overstate its sensitivity. You aren't really going to see a difference between APS-C and FF in terms of the exposure you get at a given ISO, aperture, and shutter speed.

So your argument here is based on a flawed assumption. A given ISO, aperture, and shutter speed really 8does* produce the same exposure on both APS-C and FF. With that in mind, I'd suggest reading through the entire thread. You'll find quite a lot of useful information was hashed out there. Not "proving" conclusively that one format is "better" or "worse" than the other, but providing new and more accurate perspectives on just what the differences really entail. Fascinating stuff, really.

An exposure of f 4 on APS-C would be the equivalent of the exposure of f5.6 on FF, at the SAME ISO and shutter speed; it is you that needs to rethink.

The K10D sensitivity is quite accurate:

Pentax K10D Review: 14. Photographic tests: Digital Photography Review

APS-C loses a stop of light gathering ability for the same reason that you lose a stop of light gathering ability when you use a 1.4X teleconverter (though by a slightly different path). Since the image circle is much bigger (yes, before you say it, even APS-C lenses often cover FF, and if they don't, still have an image circle much bigger than the APS-C sensor diagonal covers) than the sensor, the light collected is reduced. At the end of the day, there's no free lunch - the 1.5x "multiplier" gets multiplied not only by the focal length, but also the aperture.

This is yet another effect of the legacy of using a less-than-half-frame sensor in a camera designed around a FF sensor (i.e., lens mount, register distance, mirror box).

---

Last edited by 24X36NOW; 09-23-2009 at 04:24 PM.

[falconeye]

An exposure of f 4 on APS-C would be the equivalent of the exposure of f5.6 on FF, at the SAME ISO and shutter speed; it is you that needs to rethink.

@24X36NOW, this thread turned out to be the most educated discussion of the topic so far. And it settled in a conclusive insight which is rare enough on the interweb.

Your post probably now falls below the required quality assurance level to retrigger a serious discussion. For a reason.

This topic is full of traps and one must be very careful not to jump to conclusion.

I will give you this much of feedback, by commenting your quoted statement:

1. Yes, an exposure of f/4 on APS-C is equivalent to the exposure of f/5.6 on FF, at the same ISO and shutter speed.

2. But, you get an underexposed FF image you would push by +1EV to get the required result. Which is 99.5% (PM Class A for the remaining 0.5%) the same result as using ISO200 and not pushing.

3. Therefore and disregarding the 0.5% effect which can be in favour of FF or APS-C, this statement is true too:

An exposure of f/4 on APS-C would be the equivalent of the exposure of f/5.6 on FF, at TWICE the ISO and same shutter speed.


Where you get an advantage for FF is in the dynamic range (more headroom for the highlights) at lowest ISO. But this doesn't affect the noise discussion as of this thread. The low ISO dynamic range advantage for FF was previously discussed in this thread. Read it.


Footnote:
________
Your image circle argument is flawed because you didn't take the change of field of view into account.

[Marc Sabatella]

An exposure of f 4 on APS-C would be the equivalent of the exposure of f5.6 on FF, at the SAME ISO and shutter speed; it is you that needs to rethink.

This is 100% in opposition to what any book on photography will tell you, what any experienced photographer will tell you, as well as what what physics and logic predict. There might be slight variations in how close cameras come to their specified ISO, but it is just ordinary variation between cameras, not a 1-stop advantage to FF. If you shoot f/4 on one camera and f/5.6 on another at the same shutter speed and ISO, the f/4 picture will be one stop brighter than the f/5.6 picture, and it doesn't matter whether you use f/4 on APS-C and f/5.6 on FF or vice versa. The same is true if you're comparing FF against 4/3, or medium format against a P&S. A given f-stop, shutter speed, and ISO generates exactly the same exposure regardless of sensor size.

APS-C loses a stop of light gathering ability for the same reason that you lose a stop of light gathering ability when you use a 1.4X teleconverter (though by a slightly different path). Since the image circle is much bigger (yes, before you say it, even APS-C lenses often cover FF, and if they don't, still have an image circle much bigger than the APS-C sensor diagonal covers) than the sensor, the light collected is reduced. At the end of the day, there's no free lunch - the 1.5x "multiplier" gets multiplied not only by the focal length, but also the aperture.

That's an interesting and almost plausible analogy, but it just isn't so. A TC does indeed multiply focal length as well as aperture. Cropping the sensor doesn't actually alter either focal length or aperture - all it alters is field of view. It is no different than simple taking the picture on a FF sensor and then cropping the result in Photoshop or with a pair of scissors on the print. Exposure doesn't change if you crop with Photoshop or scissors, and it doesn't change is you crop by using a smaller sensor.

[Rondec]

As Marc said, f4 is f4 on APS-C and FF. The difference between them is field of view and depth of field, not light gathering ability.

[falconeye]

As Marc said, f4 is f4 on APS-C and FF. The difference between them is field of view and depth of field, not light gathering ability.

This in turn, my friend, isn't true either. Read the thread. Marc said to use f/5.6 and increased ISO (I agree). However, "your" f4 lens on FF is bigger and therefore, gathers more light.

This is 100% in opposition to what any book on photography will tell you

Marc, you're right. But it is possible to follow a different path of reasoning where you end up using the same ISO (and underexposing by 1 stop).

I believe that 24X36NOW followed that path which led to the confusion. I replied above to explain how both paths of reasoning are equivalent.

[24X36NOW]

I invite you all to:

1. Take an incident light reading with a hand held meter of the same light as that falling on your subject.

2. Set your APS-C dSLR manually to the appropriate combination of ISO, F stop and shutter speed (e.g., 1/500 @ f 11 at ISO 200 for a front lit subject on a bright sunny day).

3. Shoot.

Your photo will be underexposed.

Then, add a stop of additional exposure by any method you wish, i.e., reduce shutter speed or open up one f stop or increase to ISO 400, and shoot the same image.

Your photo will now be properly exposed.

[No PP allowed, of course ]

Marc, a teleconverter does not change the focal length of the lens; this is an inaccurate oversimplification. Rather, it magnifies the image circle of the lens, and the light reduction is the result of the magnified image circle being larger than the film/sensor, with the light coming through the lens no longer all being captured. Think about this logically - it would not be possible for a teleconverter of a fixed size with fixed lens elements to alter the focal length of any lens you put it behind, as it would have to be a different size for each different lens to physically alter the focal length of each lens.

It is the light that doesn't hit the sensor in an APS-C dSLR that causes the reduction in light gathering ability, because of the small-sensor-in-camera-designed-for-bigger-sensor design and the resulting much-bigger-than-the-sensor image circle.

[24X36NOW]

This in turn, my friend, isn't true either. Read the thread. Marc said to use f/5.6 and increased ISO (I agree). However, "your" f4 lens on FF is bigger and therefore, gathers more light.


Marc, you're right. But it is possible to follow a different path of reasoning where you end up using the same ISO (and underexposing by 1 stop).

I believe that 24X36NOW followed that path which led to the confusion. I replied above to explain how both paths of reasoning are equivalent.

I'm not confused, just reporting an observation of actual shooting. The theory you are talking about is all accurate ASSUMING you have an APS-C dSLR that is designed with a register distance and lens mount chosen for APS-C sized sensors, so that the image circle of the lens covers (just) the necessary size for the APS-C sensor. The issue here is that the undersized sensor has been installed into a camera designed to cover a much larger image circle, and reduced light capture results since the image circle is much larger than the sensor.

In any event, you needn't belive me, check the EXIF data on your own images that should be "Sunny 16" exposures. Are the properly exposed images Sunny 16 exposures, or 1 stop over??

[falconeye]

I invite you all to:
[...]
Your photo will be underexposed.
[...]
Then, add a stop
[...]
Your photo will now be properly exposed.
[.../.]
It is the light that doesn't hit the sensor in an APS-C dSLR that causes the reduction in light gathering ability, because of the small-sensor-in-camera-designed-for-bigger-sensor design and the resulting much-bigger-than-the-sensor image circle.

@24X36NOW,

I get it now. Understand every jota. I see how this belongs to your set of core truths. Unfortunately, it is wrong. Just like earth turned out to be not flat. Not everybody accepted it.

As you are a thinker, please accept the challenge to accept that you may just be wrong.

1. The claim that APS-C underexposes where FF doesn't is probably rooted in experience. However, it must have been with diferent vendors rather than different sensor sizes. Yes, there can be up to 1 stop difference in ISO calibration in different cameras. But this doesn't relate to sensor size at all. The ISO standard is such that sensor size differences must cancel out.

And the mount has absolutely nothing to do here (provided it isn't too narrow for wide extreme apertures or inhibiting ultra wide angles). Please, think it through. Pls.

And just to be on the safe side... the mount does not create (or determine) the image circle. You may think so. Would be amazingly wrong.

2. DxO measures calibrated ISO sensitivities. You may want to have a look at the D300/D700 comparison yourself. at dxomark.com No difference. However, you'll see a difference between D3X and D700 which are both FF.

3. If I have to try to find the point where your line of reasoning breaks down, I believe it is here:

You believe that the FF image circle is cropped by the APS-C sensor which leads to less light being captured. This sounds correct. But isn't. You forget to take into account that the APS-C lens would also have a shorter focal length which bundles the available light onto a smaller area exactly cancelling out the loss due to cropping. By available light I mean the light rays (photons) hitting the front element from within the field of view.

At constant f-stop, this front element is larger for the FF lens hence the advantage in light collection. But not anymore for a lens of same physical size, like f/4 on APS-C and f/5.6 on FF. Then exactly the same amount of light is shed onto the overall sensor area.


Please, in replying, refrain from the reflex to shout "no". Try to understand every word I wrote here and further up. I did it with your posts. Listen, don't speak. I already heard you

Marc, a teleconverter does not change the focal length of the lens; this is an inaccurate oversimplification. Rather, it magnifies the image circle of the lens

Don't disagree with your magnification statement. Except that it absolutely makes no difference by which optical means you implement a given focal length. Adding a magnifying rear element is just part of standard lens construction. Go measure your 300mm lens. Is it 300mm long? Not? How can it be? Magnifying rear element?

The fact that you call teleconverter changing focal length "oversimplification" tells me that you don't have an accurate understanding of optical physics.

I really invite you to read the thread from the beginning. It is good lecture.
Anyway, I cannot spend more time to discuss it with you. Seize your chance. Or not.

[dnas]

At constant f-stop, this front element is larger for the FF lens hence the advantage in light collection. But not anymore for a lens of same physical size, like f/4 on APS-C and f/5.6 on FF. Then exactly the same amount of light is shed onto the overall sensor area.

Suppose you use the SAME FF lens (e.g. a 200mm F4 lens ) on both the FF & APS-C cameras. This means the front element is NOT larger for the FF lens, and that F4 means F4.

If you take a shot of a uniformly white wall, at F4, 1/500 second, then the resulting image will be the same brightness, whether you use FF or APS-C.

[Pentaxor]

@24X36NOW,

I get it now. Understand every jota. I see how this belongs to your set of core truths. Unfortunately, it is wrong. Just like earth turned out to be not flat. Not everybody accepted it.

As you are a thinker, please accept the challenge to accept that you may just be wrong.

1. The claim that APS-C underexposes where FF doesn't is probably rooted in experience. However, it must have been with diferent vendors rather than different sensor sizes. Yes, there can be up to 1 stop difference in ISO calibration in different cameras. But this doesn't relate to sensor size at all. The ISO standard is such that sensor size differences must cancel out.

And the mount has absolutely nothing to do here (provided it isn't too narrow for wide extreme apertures or inhibiting ultra wide angles). Please, think it through. Pls.

And just to be on the safe side... the mount does not create (or determine) the image circle. You may think so. Would be amazingly wrong.

2. DxO measures calibrated ISO sensitivities. You may want to have a look at the D300/D700 comparison yourself. at dxomark.com No difference. However, you'll see a difference between D3X and D700 which are both FF.

3. If I have to try to find the point where your line of reasoning breaks down, I believe it is here:

You believe that the FF image circle is cropped by the APS-C sensor which leads to less light being captured. This sounds correct. But isn't. You forget to take into account that the APS-C lens would also have a shorter focal length which bundles the available light onto a smaller area exactly cancelling out the loss due to cropping. By available light I mean the light rays (photons) hitting the front element from within the field of view.

At constant f-stop, this front element is larger for the FF lens hence the advantage in light collection. But not anymore for a lens of same physical size, like f/4 on APS-C and f/5.6 on FF. Then exactly the same amount of light is shed onto the overall sensor area.


Please, in replying, refrain from the reflex to shout "no". Try to understand every word I wrote here and further up. I did it with your posts. Listen, don't speak. I already heard you


Don't disagree with your magnification statement. Except that it absolutely makes no difference by which optical means you implement a given focal length. Adding a magnifying rear element is just part of standard lens construction. Go measure your 300mm lens. Is it 300mm long? Not? How can it be? Magnifying rear element?

The fact that you call teleconverter changing focal length "oversimplification" tells me that you don't have an accurate understanding of optical physics.

I really invite you to read the thread from the beginning. It is good lecture.
Anyway, I cannot spend more time to discuss it with you. Seize your chance. Or not.

forgive me for butting in falconeye, but just a quick question. is this the reason why DX lenses mounted on an FF body like the D700 or D3x have a different result than an FX lens mounted?

[Class A]

The issue here is that the undersized sensor has been installed into a camera designed to cover a much larger image circle, and reduced light capture results since the image circle is much larger than the sensor.

Assume you replace an FF lens (with the help of an adapter) with a MF format lens (of the same focal length) and mount it on your full frame camera. Will the camera suddenly start to underexpose? No. Just because the image circle is larger (the covering power of the lens is higher) this doesn't change anything about the exposure of the light captured by the sensor. The amount of light outside of the sensor is extra to what is needed, it is not lost.

When you compare equivalent images (in particular having the same FOV) you need to take into account that the focal length of the lenses have to change when you switch formats. The light lost to the APS-C sensor won't be collected by the FF sensor either because it must not be recorded in order to yield the same scene.

One could even argue that the FF sensor is disadvantaged as the same part of the scene is projected onto a larger area, hence, leading to a loss of light intensity (cd/mm^2). Yet, the larger light collecting area cancels this effect. It does not, however, overcompensate it yielding a net advantage in better light collecting ability.

[falconeye]

Suppose you use the SAME FF lens (e.g. a 200mm F4 lens ) on both the FF & APS-C cameras.

We seem to have a number of late arrivals in this thread
Comparing the same focal length on APS-C vs. FF is meaningless as it yields entirely different images (different FoV). Beware of the traps.

why DX lenses mounted on an FF body like the D700 or D3x have a different result than an FX lens mounted?

What do you mean by "different result"? Of course, a DX lens' optical design (whatever be the mount) is such that you have an unacceptable loss of resolution and/or illumination outside the APS-C image circle which would also affect exposure measurement. Spot metering or DX firmware cropping could minimalize the effect.

When you compare equivalent images (in particular having the same FOV) you need to take into account that the focal length of the lenses have to change when you switch formats.

Thanxs for coming to rescue