Full Frame vs. APS-C?

[Axel114]

Here is an example of what I meant -- what the same focal length would look like on the two sensors. I just think it gives people a better sense of what the actual result is.

Yeah that's fine. That works too. Of course, it doesn't provide context on why.

Some people want to know why and some just want to know effects. Either way, the more examples, the better.

On the page you linked they had a similar example, so hopefully people are seeing that the images are not smaller/cropped. To me the pics I posted above just give a more accurate representation of the formats. Some people complain when it's explained as reach and zoom, but that's the end product that people see at the same focal length.

 

[Great Bustard]

Here is an example of what I meant -- what the same focal length would look like on the two sensors. I just think it gives people a better sense of what the actual result is.

Yeah that's fine. That works too. Of course, it doesn't provide context on why.

Some people want to know why and some just want to know effects. Either way, the more examples, the better.

On the page you linked they had a similar example, so hopefully people are seeing that the images are not smaller/cropped. To me the pics I posted above just give a more accurate representation of the formats.

The examples about give an accurate representation of the differences in framing one can expect when taking a photo of the same scene from the same position with the same focal length on each.

Some people complain when it's explained as reach and zoom, but that's the end product that people see at the same focal length.

There are important differences between zooming and cropping with regards to resolution, noise, and DOF. For example, a photo at 100mm cropped to the same framing as a photo at 200mm on the same camera will have 1/4 the resolution as the 200mm photo if the lenses are the same sharpness. The cropped photo will will also be be made with 1/4 as much light as the zoomed photo if taken with the same exposure, and thus twice as noisy. Lastly, the cropped photo will have twice the DOF if taken with the same relative aperture as the zoomed photo.

Of course, the differences in resolution, noise, and DOF may not matter, depending on many factors (to include the scene, the display size of the photo, and the viewing distance). But the differences are there, and the reasons to choose FF over APS-C are the same reasons to choose an f/2.8 lens over an f/4 lens. Likewise, the reasons to choose APS-C are the same reasons to choose an f/4 lens over an f/2.8 lens.

 

[mostlyboringphotog]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

We do not say equivalent apertures are the apertures that result in same "exposure".

Correct. Equivalent photos on different formats, by definition, do not have the same exposure (they do, however, have the same brightness).

Mon ami - you got it...

So I've been told. ;-)

- "Equivalent photos" have different sensor size, different FL, different aperture...

Different *relative* aperture, but the same *effective* aperture (aperture diameter -- entrance pupil).

It's true that 50/4 ~= 75/5.6 ~= 100/8 ... but these will have different DOF on a given sensor size. So the same "effective" aperture diameter is not what keeps the DOF same between FF and crop sensors.

It's the same "effective" aperture diameter and different magnification (enlargement) is what equalized the DOF.

(same SS, though - thus different exposure) and different ISO (thus the same brightness).

The photos are equivalent rather than "equal" because different sensor size, different FL, different aperture (same SS, though) and different ISO resulted in equal looking photo.

Still not equal. There are many visual properties not part of the definition of Equivalent photos, such as resolution, bokeh, flare, color, distortion, etc..

OK - will a nearly equal do?

One of the better test of equivalence is a substitution.

For exposure you can substitute either f/2 1/200 or f/2.8 1/100. Therefore, both are equivalent exposure. But we do not say therefore, f/2 is equivalent to f/2.8.

Likewise, FF 75mm f/2.8 1/100 ISO200 and DX 50mm f/2 1/100 ISO100 will result in "equivalent photo" as you define it and now one can use either camera (substitute one for the other) to get "equivalent photo"; however, we should not say, therefore, 75mm is equivalent to 50mm and f/2.8 is equivalent to f/2, even though both lens have theoretically same "effective" aperture. diameter.

 

[Great Bustard]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

We do not say equivalent apertures are the apertures that result in same "exposure".

Correct. Equivalent photos on different formats, by definition, do not have the same exposure (they do, however, have the same brightness).

Mon ami - you got it...

So I've been told. ;-)

- "Equivalent photos" have different sensor size, different FL, different aperture...

Different *relative* aperture, but the same *effective* aperture (aperture diameter -- entrance pupil).

It's true that 50/4 ~= 75/5.6 ~= 100/8 ... but these will have different DOF on a given sensor size.

Nor will they have the same framing, so they are not equivalent.

So the same "effective" aperture diameter is not what keeps the DOF same between FF and crop sensors.

It is for a given perspective, framing, and display size.

It's the same "effective" aperture diameter and different magnification (enlargement) is what equalized the DOF.

More precisely, the same perspective, framing, aperture diameter, display size, viewing distance, and visual acuity yield the same DOF.

(same SS, though - thus different exposure) and different ISO (thus the same brightness).

The photos are equivalent rather than "equal" because different sensor size, different FL, different aperture (same SS, though) and different ISO resulted in equal looking photo.

Still not equal. There are many visual properties not part of the definition of Equivalent photos, such as resolution, bokeh, flare, color, distortion, etc..

OK - will a nearly equal do?

Not necessarily. Sometimes it's these other visual properties that are primary.

One of the better test of equivalence is a substitution.

For exposure you can substitute either f/2 1/200 or f/2.8 1/100. Therefore, both are equivalent exposure. But we do not say therefore, f/2 is equivalent to f/2.8.

f/2 is equivalent to f/2.8 if it results in the same DOF for a given perspective and framing and same total amount of light on the sensor for a given exposure time.

Likewise, FF 75mm f/2.8 1/100 ISO200 and DX 50mm f/2 1/100 ISO100 will result in "equivalent photo" as you define it and now one can use either camera (substitute one for the other) to get "equivalent photo"; however, we should not say, therefore, 75mm is equivalent to 50mm and f/2.8 is equivalent to f/2, even though both lens have theoretically same "effective" aperture. diameter.

Again, Equivalent Lenses are defined as lenses that result in Equivalent Photos on their respective systems, which means that they will have the same [diagonal] angle of view and the same aperture diameter.

There is a slight niggle with some specialty lenses, like Sony's new 100 / 2.8 STF lens which is t/4.5 at f/2.8. That is, it has the DOF of an f/2.8 lens but puts the same light on the sensor as an f/4.5 lens. In terms of Equivalence, it's like using the lens with a CPL.

 

[mostlyboringphotog]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

We do not say equivalent apertures are the apertures that result in same "exposure".

Correct. Equivalent photos on different formats, by definition, do not have the same exposure (they do, however, have the same brightness).

Mon ami - you got it...

So I've been told. ;-)

- "Equivalent photos" have different sensor size, different FL, different aperture...

Different *relative* aperture, but the same *effective* aperture (aperture diameter -- entrance pupil).

It's true that 50/4 ~= 75/5.6 ~= 100/8 ... but these will have different DOF on a given sensor size.

Nor will they have the same framing, so they are not equivalent.

So the same "effective" aperture diameter is not what keeps the DOF same between FF and crop sensors.

It is for a given perspective, framing, and display size.

It's the same "effective" aperture diameter and different magnification (enlargement) is what equalized the DOF.

More precisely, the same perspective, framing, aperture diameter, display size, viewing distance, and visual acuity yield the same DOF.

(same SS, though - thus different exposure) and different ISO (thus the same brightness).

The photos are equivalent rather than "equal" because different sensor size, different FL, different aperture (same SS, though) and different ISO resulted in equal looking photo.

Still not equal. There are many visual properties not part of the definition of Equivalent photos, such as resolution, bokeh, flare, color, distortion, etc..

OK - will a nearly equal do?

Not necessarily. Sometimes it's these other visual properties that are primary.

One of the better test of equivalence is a substitution.

For exposure you can substitute either f/2 1/200 or f/2.8 1/100. Therefore, both are equivalent exposure. But we do not say therefore, f/2 is equivalent to f/2.8.

f/2 is equivalent to f/2.8 if it results in the same DOF for a given perspective and framing and same total amount of light on the sensor for a given exposure time.

Likewise, FF 75mm f/2.8 1/100 ISO200 and DX 50mm f/2 1/100 ISO100 will result in "equivalent photo" as you define it and now one can use either camera (substitute one for the other) to get "equivalent photo"; however, we should not say, therefore, 75mm is equivalent to 50mm and f/2.8 is equivalent to f/2, even though both lens have theoretically same "effective" aperture. diameter.

Again, Equivalent Lenses are defined as lenses that result in Equivalent Photos on their respective systems,

This what we disagree and this is how we disagree -

FL and sensor size are Equivalent FOV; FL is not equivalent; sensor size is not equivalent.

Aperture (F#) and sensor size are Equivalent DOF; F# is not equivalent; sensor size is not equivalent.

ISO and sensor size are Equivalent Noise; ISO is not equivalent; sensor size is not equivalent.

FL and F# and ISO are Equivalent Photo; like "equivalent exposure", FL, F# and ISO are not the equivalents.

Now, drum roll please, we have an "Equivalent Photo Triangle" :-D :-O ;-)

Hope to share some beer for real...

which means that they will have the same [diagonal] angle of view and the same aperture diameter.

There is a slight niggle with some specialty lenses, like Sony's new 100 / 2.8 STF lens which is t/4.5 at f/2.8. That is, it has the DOF of an f/2.8 lens but puts the same light on the sensor as an f/4.5 lens. In terms of Equivalence, it's like using the lens with a CPL.

 

[beatboxa]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

We do not say equivalent apertures are the apertures that result in same "exposure".

Correct. Equivalent photos on different formats, by definition, do not have the same exposure (they do, however, have the same brightness).

Mon ami - you got it...

So I've been told. ;-)

- "Equivalent photos" have different sensor size, different FL, different aperture...

Different *relative* aperture, but the same *effective* aperture (aperture diameter -- entrance pupil).

It's true that 50/4 ~= 75/5.6 ~= 100/8 ... but these will have different DOF on a given sensor size.

Nor will they have the same framing, so they are not equivalent.

So the same "effective" aperture diameter is not what keeps the DOF same between FF and crop sensors.

It is for a given perspective, framing, and display size.

It's the same "effective" aperture diameter and different magnification (enlargement) is what equalized the DOF.

More precisely, the same perspective, framing, aperture diameter, display size, viewing distance, and visual acuity yield the same DOF.

(same SS, though - thus different exposure) and different ISO (thus the same brightness).

The photos are equivalent rather than "equal" because different sensor size, different FL, different aperture (same SS, though) and different ISO resulted in equal looking photo.

Still not equal. There are many visual properties not part of the definition of Equivalent photos, such as resolution, bokeh, flare, color, distortion, etc..

OK - will a nearly equal do?

Not necessarily. Sometimes it's these other visual properties that are primary.

One of the better test of equivalence is a substitution.

For exposure you can substitute either f/2 1/200 or f/2.8 1/100. Therefore, both are equivalent exposure. But we do not say therefore, f/2 is equivalent to f/2.8.

f/2 is equivalent to f/2.8 if it results in the same DOF for a given perspective and framing and same total amount of light on the sensor for a given exposure time.

Likewise, FF 75mm f/2.8 1/100 ISO200 and DX 50mm f/2 1/100 ISO100 will result in "equivalent photo" as you define it and now one can use either camera (substitute one for the other) to get "equivalent photo"; however, we should not say, therefore, 75mm is equivalent to 50mm and f/2.8 is equivalent to f/2, even though both lens have theoretically same "effective" aperture. diameter.

Again, Equivalent Lenses are defined as lenses that result in Equivalent Photos on their respective systems,

This what we disagree and this is how we disagree -

FL and sensor size are Equivalent FOV; FL is not equivalent; sensor size is not equivalent.

Aperture (F#) and sensor size are Equivalent DOF; F# is not equivalent; sensor size is not equivalent.

ISO and sensor size are Equivalent Noise; ISO is not equivalent; sensor size is not equivalent.

FL and F# and ISO are Equivalent Photo; like "equivalent exposure", FL, F# and ISO are not the equivalents.

Now, drum roll please, we have an "Equivalent Photo Triangle" :-D :-O ;-)

Hope to share some beer for real...

I don't know why or what you're arguing. The word 'equivalent' implicitly deals with the effects & counterparts. 'Equivalent' is not necessarily the same thing as 'equal'--the context has to be framed. Here, they have been framed as the output image. So all of this explanation is redundant and pedantic. It's arguing semantics for the sake of arguing instead of the actual concepts. Just like yesterday where (in a thread about equivalence), you were side-skirting the issue and arguing about cropping instead of equivalence (which is what the thread was about).

 

[Great Bustard]

One of the better test of equivalence is a substitution.

For exposure you can substitute either f/2 1/200 or f/2.8 1/100. Therefore, both are equivalent exposure. But we do not say therefore, f/2 is equivalent to f/2.8.

f/2 is equivalent to f/2.8 if it results in the same DOF for a given perspective and framing and same total amount of light on the sensor for a given exposure time.

Likewise, FF 75mm f/2.8 1/100 ISO200 and DX 50mm f/2 1/100 ISO100 will result in "equivalent photo" as you define it and now one can use either camera (substitute one for the other) to get "equivalent photo"; however, we should not say, therefore, 75mm is equivalent to 50mm and f/2.8 is equivalent to f/2, even though both lens have theoretically same "effective" aperture. diameter.

Again, Equivalent Lenses are defined as lenses that result in Equivalent Photos on their respective systems,

This what we disagree and this is how we disagree -

FL and sensor size are Equivalent FOV; FL is not equivalent; sensor size is not equivalent.

Aperture (F#) and sensor size are Equivalent DOF; F# is not equivalent; sensor size is not equivalent.

ISO and sensor size are Equivalent Noise; ISO is not equivalent; sensor size is not equivalent.

FL and F# and ISO are Equivalent Photo; like "equivalent exposure", FL, F# and ISO are not the equivalents.

Now, drum roll please, we have an "Equivalent Photo Triangle" :-D :-O ;-)

Hope to share some beer for real...

50mm is not equivalent to 75mm full stop. 50mm on 1.5x is equivalent to 75mm on FF. f/2 is not equivalent to f/3 full stop. f/2 on 1.5x is equivalent to f/3 on FF.

When people say, for example, that "f/2 is equivalent to f/4", it is because the context of that *particular* conversation is mFT vs FF. And, then, yes, 25mm f/2 on mFT is equivalent to 50mm f/4 on FF because the lenses have the same [diagonal] angle of view on their respective systems and the same aperture diameters.

 

[Stejo]

Nothing is ever that easy and straightforward.

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

If you can find it and manage to carry it, try focusing it on anything, moving or not.

Furthermore, there's limits to human vision.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

How about printing them on A3 or A2? At which point does the difference become relevant? Does it ever become relevant when downsampled for web use? Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

 

[Great Bustard]

50mm is not equivalent to 75mm full stop. 50mm on 1.5x is equivalent to 75mm on FF. f/2 is not equivalent to f/3 full stop. f/2 on 1.5x is equivalent to f/3 on FF.

When people say, for example, that "f/2 is equivalent to f/4", it is because the context of that *particular* conversation is mFT vs FF. And, then, yes, 25mm f/2 on mFT is equivalent to 50mm f/4 on FF because the lenses have the same [diagonal] angle of view on their respective systems and the same aperture diameters.

Nothing is ever that easy and straightforward.

Actually, lots of things are. and only get messy if you try to become more and more accurate. For example, the DOF for 25mm f/2 on mFT will not be *exactly* the same as the DOF for 50mm f/4 on FF because the lenses will not have the *exact* same design. But they will be close, just as the DOF for the 25 / 1.2, 25 / 1.4, 25 / 1.7, and 25 / 1.8 mFT lenses will all be close at f/2, but not *exactly* the same.

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

An 800 / 13. There might be some mirror lenses like that.

If you can find it and manage to carry it...

An 800 / 13 isn't going to be all that big and heavy.

...try focusing it on anything, moving or not.

That would be a challenge with PDAF. Wouldn't be any more a challenge for CDAF than would a 400 / 6.3 on mFT, however.

Furthermore, there's limits to human vision.

Sure.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Well, you won't see any resolution differences because you specified that they had to have equal resolution from the get-go.

How about printing them on A3 or A2?

Same.

At which point does the difference become relevant? Does it ever become relevant when downsampled for web use?

Not for photos with equal resolution.

Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Equivalence does indeed address this point:

Loading…

www.josephjamesphotography.com

Photography is all about the photo. But before we talk about the IQ of the photo, we must first get the photo. In other words, IQ plays no role in a photo that is out of focus. IQ plays no role in a photo that is missed due to slow focus or shutter lag. IQ plays no role in a photo that was not captured because the equipment was too bulky to be carried to the mountain top or too conspicuous to be used.

However, just as the amazing television series, Planet Earth, is significantly more impressive on a 65 inch 4K TV than it is on a 36 inch conventional TV set, there are plenty of TV shows and movies where the type of TV set they were viewed on would make no significant difference to the viewer. Thus, while the IQ of a photo, as well as the optional ability to achieve a more shallow DOF, can greatly enhance the impact that a photo has, this impact depends greatly on the photo as well as its display dimensions. In other words, sometimes IQ is paramount, and sometimes it is not. Just as we do not all watch the same TV shows, or even have the same opinions about the value of the shows that we do watch, different photographers will not take the same types of photographs or give the various elements of IQ the same value as another. Each photographer must balance the operation of a system against its IQ potential not only in concert with the display size of the photo, but also with both their skills in photography and post-processing, to decide what system best gets the job done for the type of photography that they do.

The debate between different sensor formats is very much like the debate between primes and zooms. While top-quality primes may have higher IQ, allow for a more shallow DOF, and be better suited for low-light photography, they do not zoom. That singular advantage of a zoom trumps all the advantages of a prime for many photographers, and so it is when comparing formats. It is not only a matter of whether one system is "better" than the other in terms of IQ, but at what display dimensions this difference becomes significant. For many, and likely most, it is more often a matter of available lenses, differences in DOF capabilities, and operational convenience, than it is a matter of IQ alone.

The bottom line is that we use a camera to create photos. It is important to understand the advantages of any particular system as a whole, both in terms of IQ and operation. The purpose of equivalence is to help evaluate the IQ end of that consideration, and, in conjunction with our individual "quality threshold", make an informed choice as to which system, or systems, best meet our personal needs for the photography that we do.

 

[beatboxa]

Nothing is ever that easy and straightforward.

(because people become pedantic and argumentative)

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

In terms of image quality, about 800mm F/12.6. This would result in an image with similar:

• Field of view
• Depth of Field
• Noise

At the same output size / resolution.

If you can find it and manage to carry it, try focusing it on anything, moving or not.

Furthermore, there's limits to human vision.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Depends on whether or not the details are at near the extrema. In some cases, yes. In other cases, no.

How about printing them on A3 or A2? At which point does the difference become relevant? Does it ever become relevant when downsampled for web use? Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Again, there are limits, and yes it sometimes becomes relevant for even web use. Nobody is claiming that in all cases the differences are noticeable. Most images at a tiny resolution like 64 x 64 pixels will be indistinguishable whether they came from a smartphone or medium format. But as you approach extrema, the differences become more pronounced. If anyone truly believes that there are no differences, then it would be hypocritical to own anything beyond a Pentax Q.

So I'd pose the question back to you: do you personally own any camera with a sensor larger than 1"? Why or why not?

I own and shoot multiple formats, including 1", micro four thirds, APS-C, and Full Frame. Equivalency is a real (but annoying when explaining) concept, and it has everything to do with the input & the output in different ways. When someone asks questions about equivalence, I give them straight answers aligned to the context of the question rather than throwing irrelevant variables in the mix. For example, pretty much nobody asking the question is usually asking about circle of confusion--they're asking about normal, observable, almost casual results. It's pedantic to try to argue points that are obviously out of scope of the topic for normal people and normal uses. The whole "well, *technically*...." argument gets old, fast. Things like: "Well technically depth of field depends on how far you are away from the picture you're viewing...." is not the answer to the question most people are asking.

 

[Stejo]

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

An 800 / 13. There might be some mirror lenses like that.

An 800 / 13 isn't going to be all that big and heavy.

...try focusing it on anything, moving or not.

That would be a challenge with PDAF. Wouldn't be any more a challenge for CDAF than would a 400 / 6.3 on mFT, however.

There's no such thing as a 800/13 because it's impossible to work with PDAF. And as I'm sure you know, there is a very high quality 400/6.3 available for mft that focuses extremely fast without any issues whatsoever. You can't disconnect theory from practical reality.

An 800/6.3 FF would be better in every aspect ofc, but it wouldn't be equivalent. It would be a very expensive specialized piece of equipment that can't be hand held or even carried at all.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Well, you won't see any resolution differences because you specified that they had to have equal resolution from the get-go.

There's plenty of 24MP FF and APS-C cameras around. So a 35/2 on APS-C produces identical images to a 50/2 on FF?

At which point does the difference become relevant? Does it ever become relevant when downsampled for web use?

Not for photos with equal resolution.

Look above, but still. I'll give you even that. How about different resolution. Take any sensor your like from what's currently available out there. Is the noise advantage perceivable when the images are printed on A4?

Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Equivalence does indeed address this point:

Loading…

www.josephjamesphotography.com

This is just words. I'm talking about actual numbers. Like taking photos with different sensors, printing them to the same size paper and analyzing the prints. Even better would be scanning the prints at very high resolution for everyone to see.

 

[Stejo]

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

In terms of image quality, about 800mm F/12.6. This would result in an image with similar:

• Field of view
• Depth of Field
• Noise

At the same output size / resolution.

f/12.6 is not something PDAF can handle. Thus, there is no FF 800/12.6 in production. There is however a 400/6.3 MFT in production which has no equivalent because... it's not that simple.

Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Again, there are limits, and yes it sometimes becomes relevant for even web use. Nobody is claiming that in all cases the differences are noticeable. Most images at a tiny resolution like 64 x 64 pixels will be indistinguishable whether they came from a smartphone or medium format. But as you approach extrema, the differences become more pronounced. If anyone truly believes that there are no differences, then it would be hypocritical to own anything beyond a Pentax Q.

So I'd pose the question back to you: do you personally own any camera with a sensor larger than 1"? Why or why not?

There are limits, we don't disagree. My question was pretty simple: what exactly are those limits? Well, not even exactly, a ballpark will do, since it varies according to your visual acuity.

Everybody is quick to say that f/2.8 = f/5.6. Nobody says under which parameters that holds true. Just as DoF is a function of visual acuity and scaling, so is noise. There's hard limits to what is perceivable. And just as you consider a range of planes in focus because of projection and perception limitations, the same should apply to noise.

And no, we're not talking 64x64 extremes. We're talking about limits that fall within current practical applications.

As for your question, I own or have owned in the past various cameras of various sensor sizes. Currently my largest sensor is MFT because it covers my delivery needs. When a beginner asks about equivalence, without any real grasp to real world use and you tell him that f/2.8=f/5.6 flat out, you're doing him a disservice imo.

 

[mostlyboringphotog]

Nothing is ever that easy and straightforward.

(because people become pedantic and argumentative)

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

In terms of image quality, about 800mm F/12.6. This would result in an image with similar:

• Field of view
• Depth of Field
• Noise

At the same output size / resolution.

If you can find it and manage to carry it, try focusing it on anything, moving or not.

Furthermore, there's limits to human vision.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Depends on whether or not the details are at near the extrema. In some cases, yes. In other cases, no.

How about printing them on A3 or A2? At which point does the difference become relevant? Does it ever become relevant when downsampled for web use? Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Again, there are limits, and yes it sometimes becomes relevant for even web use. Nobody is claiming that in all cases the differences are noticeable. Most images at a tiny resolution like 64 x 64 pixels will be indistinguishable whether they came from a smartphone or medium format. But as you approach extrema, the differences become more pronounced. If anyone truly believes that there are no differences, then it would be hypocritical to own anything beyond a Pentax Q.

If the pixel pitch and the technology is the same, and exposed at the same exposure, the 1000x1000 from FF would look exactly the same as 1000x1000 from a phone cam.

It's not that you would not get better image (less noise) from a larger sensor. However, you can get a better image from a smaller sensor as well, if you down sample the raw file. Of course, you'd end up with a smaller image with less resolution.

Since the FF would have more pixels, it can be down sampled to the same resolution as the phone cam and now the FF image looks much better.

So I'd pose the question back to you: do you personally own any camera with a sensor larger than 1"? Why or why not?

I own and shoot multiple formats, including 1", micro four thirds, APS-C, and Full Frame. Equivalency is a real (but annoying when explaining) concept, and it has everything to do with the input & the output in different ways. When someone asks questions about equivalence, I give them straight answers aligned to the context of the question rather than throwing irrelevant variables in the mix. For example, pretty much nobody asking the question is usually asking about circle of confusion--they're asking about normal, observable, almost casual results. It's pedantic to try to argue points that are obviously out of scope of the topic for normal people and normal uses. The whole "well, *technically*...." argument gets old, fast. Things like: "Well technically depth of field depends on how far you are away from the picture you're viewing...." is not the answer to the question most people are asking.

My specific argument is against the use of equivalent aperture for the smaller sensor cameras at DPR as if the equivalence of aperture is what result in the deeper DOF of smaller sensor cameras. This is like saying increasing ISO increase exposure to my ears ;-)

Of course, it is the fact that smaller sensor cameras use shorter FL that result in deeper DOF.

 

[beatboxa]

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

In terms of image quality, about 800mm F/12.6. This would result in an image with similar:

• Field of view
• Depth of Field
• Noise

At the same output size / resolution.

f/12.6 is not something PDAF can handle. Thus, there is no FF 800/12.6 in production. There is however a 400/6.3 MFT in production which has no equivalent because... it's not that simple.

Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Again, there are limits, and yes it sometimes becomes relevant for even web use. Nobody is claiming that in all cases the differences are noticeable. Most images at a tiny resolution like 64 x 64 pixels will be indistinguishable whether they came from a smartphone or medium format. But as you approach extrema, the differences become more pronounced. If anyone truly believes that there are no differences, then it would be hypocritical to own anything beyond a Pentax Q.

So I'd pose the question back to you: do you personally own any camera with a sensor larger than 1"? Why or why not?

There are limits, we don't disagree. My question was pretty simple: what exactly are those limits? Well, not even exactly, a ballpark will do, since it varies according to your visual acuity.

Everybody is quick to say that f/2.8 = f/5.6. Nobody says under which parameters that holds true. Just as DoF is a function of visual acuity and scaling, so is noise. There's hard limits to what is perceivable. And just as you consider a range of planes in focus because of projection and perception limitations, the same should apply to noise.

And no, we're not talking 64x64 extremes. We're talking about limits that fall within current practical applications.

As for your question, I own or have owned in the past various cameras of various sensor sizes. Currently my largest sensor is MFT because it covers my delivery needs. When a beginner asks about equivalence, without any real grasp to real world use and you tell him that f/2.8=f/5.6 flat out, you're doing him a disservice imo.

I think it's a disservice to downplay the effects of a larger sensor.

So let's be clear on a few things: A larger sensor has almost the exact same effects of a faster lens. A larger sensor allows more light less noise and a shallower DoF. So if you'd recommend a faster lens but not a larger sensor, you'd be hypocritical.

These limits are affected by many different things, not just resolution. There is no 'one size fits all' rule, and it's naive or misleading & argumentative to suggest there is one. Scene lighting, depth of field, resolution, size, etc. all contribute.

A while ago, someone tried to argue this point, and so I posted 2 images at 640x480--and the differences were clear. So again, let's say someone asked about the F/2.8 & F/5.6 lenses...on the same sensor. Would you suggest that there are no differences in the images? Because that's effectively what you're arguing right now, even though there clearly would be differences.

 

[beatboxa]

Nothing is ever that easy and straightforward.

(because people become pedantic and argumentative)

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

In terms of image quality, about 800mm F/12.6. This would result in an image with similar:

• Field of view
• Depth of Field
• Noise

At the same output size / resolution.

If you can find it and manage to carry it, try focusing it on anything, moving or not.

Furthermore, there's limits to human vision.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Depends on whether or not the details are at near the extrema. In some cases, yes. In other cases, no.

How about printing them on A3 or A2? At which point does the difference become relevant? Does it ever become relevant when downsampled for web use? Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Again, there are limits, and yes it sometimes becomes relevant for even web use. Nobody is claiming that in all cases the differences are noticeable. Most images at a tiny resolution like 64 x 64 pixels will be indistinguishable whether they came from a smartphone or medium format. But as you approach extrema, the differences become more pronounced. If anyone truly believes that there are no differences, then it would be hypocritical to own anything beyond a Pentax Q.

If the pixel pitch and the technology is the same, and exposed at the same exposure, the 1000x1000 from FF would look exactly the same as 1000x1000 from a phone cam.

It's not that you would not get better image (less noise) from a larger sensor. However, you can get a better image from a smaller sensor as well, if you down sample the raw file. Of course, you'd end up with a smaller image with less resolution.

Since the FF would have more pixels, it can be down sampled to the same resolution as the phone cam and now the FF image looks much better.

I've posted numerous samples but I have yet to see you do this. So instead of just making misleading claims, why don't you just provide the samples?

A while ago, someone made a similar claim--so I posted two images at 640x480, both at F/1.9. The differences were stark.

And yes, you do get a better image (less noise) from a larger sensor. You cannot get an image that's "better" (less noise) from the phone than the full frame--at best you may be able to equal it in ideal lighting and less than ideal viewing conditions.

And you're again now assuming that the FF would have more pixels, which isn't necessarily true. Note that within the past day or two, you tried to make the opposite of your claim here about downscaling in the other thread, but I had to correct you there.

So I'd pose the question back to you: do you personally own any camera with a sensor larger than 1"? Why or why not?

I own and shoot multiple formats, including 1", micro four thirds, APS-C, and Full Frame. Equivalency is a real (but annoying when explaining) concept, and it has everything to do with the input & the output in different ways. When someone asks questions about equivalence, I give them straight answers aligned to the context of the question rather than throwing irrelevant variables in the mix. For example, pretty much nobody asking the question is usually asking about circle of confusion--they're asking about normal, observable, almost casual results. It's pedantic to try to argue points that are obviously out of scope of the topic for normal people and normal uses. The whole "well, *technically*...." argument gets old, fast. Things like: "Well technically depth of field depends on how far you are away from the picture you're viewing...." is not the answer to the question most people are asking.

My specific argument is against the use of equivalent aperture for the smaller sensor cameras at DPR as if the equivalence of aperture is what result in the deeper DOF of smaller sensor cameras. This is like saying increasing ISO increase exposure to my ears ;-)

Of course, it is the fact that smaller sensor cameras use shorter FL that result in deeper DOF.

No, you don't understand the concept of a larger sensor or of ISO. This wouldn't be like increasing ISO to your ears. That's a very poor analogy. This would be like bigger eyes vs smaller eyes.

 

[Porky89]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

Ah! But what about equivalent noise? Apparently, according to that renowned photography expert Joseph James, the narrower the angle of view, the greater the image noise. And of course if you cut a photographic print in half it becomes four times as noisy. Full details here: Scissors, photos, and lenses.

--
The most startling incident in my life was the time I discovered myself to be a poet, which was in the year 1877.
William McGonagall

 

[beatboxa]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

Ah! But what about equivalent noise? Apparently, according to that renowned photography expert Joseph James, the narrower the angle of view, the greater the image noise. And of course if you cut a photographic print in half it becomes four times as noisy. Full details here: Scissors, photos, and lenses.

--
The most startling incident in my life was the time I discovered myself to be a poet, which was in the year 1877.
William McGonagall

This is the issue of removing relevant context or misunderstanding the basics. So let's clarify:

Each point of a subject has light that reflects or projects in relatively random directions. The light scatters.



There is a finite amount of light, especially when isolated with respect to time. To capture this light from a point, a 'wide open', lens focused on a certain distance captures all of the light on its surface and projects it onto a point on the sensor:



As below, a lens with a larger aperture captures more light, resulting in less shot noise. It is a larger area.

With a longer focal length combined with a larger sensor (ie. 'equivalent focal length, same F/number'), it can also be framed the same as the above; but the results will be more light / less shot noise as well as a shallower DoF. The shot noise will decrease because there is a larger sample to average. The DoF is because the light rays from the edge of the lens have to refract more than a smaller aperture, so they travel a different distance to the sensor. Out of focus areas are more prone to blur (like the red below).







If one used a longer focal length lens to get the same framing, one would capture less light or need a much larger aperture diameter (which would again result in a shallower DoF). This is because the photographer / subject would be further away and this would be more prone to shot noise. The projected rays are random over a larger physical area, resulting in more shot noise.



This is all basic physics. You can remove the context of any of the above to try to make it sound off--for example "Longer focal length results in more noise! How preposterous!" That just means you either didn't understand the quote or are intentionally misrepresenting it to try to prove a point. It doesn't have any affects on the actual facts, concepts, and physical phenomenon occurring--just in your communication of it. And frankly, it's detrimental to helping people because it gives them the wrong impression of what's happening.

 

[Great Bustard]

Question 1: What's the FF equivalent of a 400mm f/6.3 MFT lens?

An 800 / 13. There might be some mirror lenses like that.

An 800 / 13 isn't going to be all that big and heavy.

...try focusing it on anything, moving or not.

That would be a challenge with PDAF. Wouldn't be any more a challenge for CDAF than would a 400 / 6.3 on mFT, however.

There's no such thing as a 800/13 because it's impossible to work with PDAF.

There is an 800 / 8 mirror lens for about $200. No AF, though.

And as I'm sure you know, there is a very high quality 400/6.3 available for mft that focuses extremely fast without any issues whatsoever.

Actually, I didn't know. You mean the long end of the 100-400 Panasonic zoom?

You can't disconnect theory from practical reality.

Sure. Why were you under the impression that because a 400 / 6.3 on mFT is equivalent to an 800 / 13 on FF that an 800 / 13 must necessarily exist and have the same specifications?

An 800/6.3 FF would be better in every aspect ofc,

Not in every respect -- it would be larger, heavier, and more expensive.

,,,but it wouldn't be equivalent. It would be a very expensive specialized piece of equipment that can't be hand held or even carried at all.

Sure. But neither here nor there with regards to what is equivalent to what, really.

Question 2: Take an equal resolution image from FF and APS-C shot at same settings. Print them both on A4. Will you be able to spot any difference at this scaling?

Well, you won't see any resolution differences because you specified that they had to have equal resolution from the get-go.

There's plenty of 24MP FF and APS-C cameras around.

Yep.

So a 35/2 on APS-C produces identical images to a 50/2 on FF?

Nope. In fact, a 35 / 2 on APS-C doesn't even produce photos identical to its FF equivalent, which would be a 50 / 2.8.

At which point does the difference become relevant? Does it ever become relevant when downsampled for web use?

Not for photos with equal resolution.

Look above, but still. I'll give you even that. How about different resolution. Take any sensor your like from what's currently available out there. Is the noise advantage perceivable when the images are printed on A4?

I would imagine that it depends on the scene, the display size, the display medium, the viewing distance, and the visual acuity of the viewer.

Should there be an addendum to your equivalency study dealing with scaling factors? Is there any perceived noise advantage to shooting FF vs a smartphone when the final images are scaled down to the same 2MP size for example?

Equivalence does indeed address this point:

http://www.josephjamesphotography.com/equivalence/#conclusion

This is just words.

As is everything posted in this forum and written in books. Well, there are photos, too -- I can post some of them if you'd like that demonstrate the points made in the link above.

I'm talking about actual numbers.

Oh, numbers! How could I have forgotten about them! :-D

Like taking photos with different sensors, printing them to the same size paper and analyzing the prints. Even better would be scanning the prints at very high resolution for everyone to see.

In fact, I'm quite the fan of that and have even discussed that. For example, consider DPR's test scene that has all those photos of the same scene from different cameras in various levels of light. What if they were to print those photos at various sizes, scan them, and then present the scans? That would a monumental pain in the butt, of course, so I can understand why they would not do that with all cameras and all combinations. But if they picked just a few cameras in a few settings and did it, that would be awesome!

 

[Great Bustard]

I think "equivalent FL" is technically a wrong term; it should be equivalent FOV/AOV.

The Equivalent Focal Lengths are the focal lengths that result in the *same* [diagonal] FOV/AOV. Likewise, the Equivalent Relative Apertures are the relative apertures that result in the *same* aperture diameter (for a given AOV).

However, to correct the above wrong, DPR added another wrong concept of "equivalent aperture" and imho, as in most cases, two wrongs does not make it right,

Perhaps the language is now a bit more clear.

Ah! But what about equivalent noise?

What's "Equivalent noise"?

Apparently, according to that renowned photography expert Joseph James, the narrower the angle of view, the greater the image noise.

Joseph James never said, or implied, any such thing. For example, two photos of the same scene with the same camera, one at 100mm f/2.8 1/400 ISO 1600 and another from twice as far back at 200mm f/2.8 1/400 ISO 1600 will be equally noisy, despite the fact that the second photo has [roughly] half the angle of view as the first.

And of course if you cut a photographic print in half it becomes four times as noisy. Full details here: Scissors, photos, and lenses.

Indeed. You should have posted the text from that link -- it goes like this:

It makes it more noisy than if it hadn't been cut, yes. For example, if I take a photo of a scene from the same position with the same camera at 100mm f/2.8 1/400 ISO 1600 and 200mm f/2.8 1/400 ISO 1600, and crop the 100mm photo to the same framing as the 200mm photo, then the cropped photo will be twice as noisy because it is made with 1/4 as much light. Feel free to give it a try.

 

[Stejo]

When a beginner asks about equivalence, without any real grasp to real world use and you tell him that f/2.8=f/5.6 flat out, you're doing him a disservice imo.

I think it's a disservice to downplay the effects of a larger sensor.

So let's be clear on a few things: A larger sensor has almost the exact same effects of a faster lens. A larger sensor allows more light less noise and a shallower DoF. So if you'd recommend a faster lens but not a larger sensor, you'd be hypocritical.

It's not my intention to downplay nor overplay anything.

A larger sensor does not have the "almost the exact same effects" (what kind of syntax is that anyway?) of a faster lens. No almost, no exact, no anything. There's noise benefits to capturing more total light, which may or may not be apparent to the naked eye depending on how the final image is viewed.

You can't make a blanket statement that using a slower lens on a larger sensor and compensating with ISO is exactly the same. It absolutely is not. ISO is not exposure. There's multiple benefits to light intensity on the sensor as well.

I suggest you re-read my posts more carefully and if you wish to address my points, please do so with proof. It shouldn't be too hard to come up with some if equivalence has universal effect, no? Start with finding an f/13 lens that can be auto focused and move on to comparing noise for various print sizes.

I'm not arguing the benefits of a larger sensor in theory. I'm just putting some much needed asterisks when it comes to practice.