[Wheatfield]

Yes, please. Is there anyone who could answer the OP's hood question (What hood would be appropriate to use for the FA 50/1.4 when used on APS-C?)?

I have a plastic hood for the SMC Pentax 100/2.8~4 that worked quite well on the FA 50/1.4.

[wallyb]

Not sure what you mean by this. You can get an image with the same field of view, the same DOF, the same shutter speed, and the same amount of noise. Whatever other difference you think might be attributable to sensor size, it's isn't going to be reflected in an of the above, so I'd question what relevance it has.

Seriously, this idea of equivalence is real. Like I said, takes a while to wrap your head around it, but it is completely 100% scientifically verifiable.

Perhaps I've been phrasing my rebuttals improperly. Let me clarify. The idea of equivalence is not a difficult idea for me personally to wrap my head around. I understand all of the technical details about a lens' circle of confusion, and I fully understand *how* you would 'equivalise' a cropped image to needing a smaller aperture in order to make the apparent DOF the same (and perhaps I'm thinking of something else when I say "apparent" DOF, since you're right that DOF is only what is apparent in the image).

But I completely fail to see the point. I see no application where you could possibly use this idea of equivalence, in regards to DOF at least. Why would it ever matter before-hand, instead of just doing it at the time of shooting the photograph, until you're pleased with the results? I've never seen where using f/7.1 versus f/8 would make or break an image I took. Do people really look at images taken with a 50mm lens on full-frame, at f/1.8, and say "Boy I wish I could go to that same exact place and take that same exact picture, but I only have an APS-C camera with me! Okay, so I need a 33.3mm lens, and I need to set the aperture to f/2.7, and adjust the shutter speed to..."

And other than having no visible practical purpose, I disagree with even talking about equivalising FF to APS-C images because when you discuss things like how the equivalent aperture needs to be smaller to give the same DOF, like was done previously in this thread, you make it sound like a lens becomes less-light-gathering and APS-C is somehow inferior in this regard. A 50mm f/1.8 lens doesn't "become" a 75mm f/2.7 lens, it stays the exact same lens no matter where it is, you just get a smaller crop of it. This crop (and subsequent upscaling) increases everything, as any kind of crop would do, be it a sensor crop or a crop on the computer, including chromatic aberrations and 'blurriness' (making the DOF more shallow).

I think my real beef only lied with how the subject was previously described.

[Rondec]

Beautiful photos, Gus (numbers 2 and 4 that is). You've given me one more excuse not to get involved with facebook. Clearly many can't appreciate great photography.

[Class A]

I have a plastic hood for the SMC Pentax 100/2.8~4 that worked quite well on the FA 50/1.4.

Interesting. Could you, at your convenience, check how deep and wide it is? I'd like to get a deeper hood but obviously would like to avoid vignetting. I would have thought that a hood that is effective for a 100 would cause vignetting on a 50mm (77mm FF-equivalent on a K-7).

A 50mm f/1.8 lens doesn't "become" a 75mm f/2.7 lens,...

I don't think anyone claimed that lenses transform. However, a 50/1.8 will take images on an APS-C camera that a 75/2.7 would take on an FF camera.
While the cases where someone wants to redo a shot using a different format are probably rare, the equivalence conversions are nevertheless useful for those with an excellent feel for numbers in FF-land who are now shooting in APS-C-land and thus can transfer their knowledge. There is also the benefit from being in a better position to make a decision between APS-C and FF, but that's another story.

---

Last edited by Class A; 12-20-2009 at 08:44 PM.

[wallyb]

I don't think anyone claimed that lenses transform. However, a 50/1.8 will take images on an APS-C that a 75/2.7 would take on an FF.
While the cases where someone wants to redo a shot using a different format are probably rare, the equivalence conversions are nevertheless useful for those with an excellent feel for numbers in FF-land who are now shooting in APS-C-land and thus can transfer their knowledge. There is also the benefit from being in a better position to make a decision between APS-C and FF, but that's another story.

I can agree with all of that

[Wheatfield]

Interesting. Could you, at your convenience, check how deep and wide it is? I'd like to get a deeper hood but obviously would like to avoid vignetting. I would have thought that a hood that is effective for a 100 would cause vignetting on a 50mm (77mm FF-equivalent on a K-7).

About 35mm deep, about 67mm across.

I don't think anyone claimed that lenses transform. However, a 50/1.8 will take images on an APS-C camera that a 75/2.7 would take on an FF camera.
While the cases where someone wants to redo a shot using a different format are probably rare, the equivalence conversions are nevertheless useful for those with an excellent feel for numbers in FF-land who are now shooting in APS-C-land and thus can transfer their knowledge. There is also the benefit from being in a better position to make a decision between APS-C and FF, but that's another story.

This whole debate is an exercise in wanking.
Just so you know.

[wallyb]

This whole debate is an exercise in wanking.
Just so you know.

How can you get good at something if you don't exercise?

[Class A]

About 35mm deep, about 67mm across.

Thanks!

This whole debate is an exercise in wanking.

Good to know that you are exercising (something) while we are discussing.

Beautiful photos, Gus (numbers 2 and 4 that is). You've given me one more excuse not to get involved with facebook. Clearly many can't appreciate great photography.

Sadly it is only my friends that comment on the pics... And they are all about the cell phone shots now!

I have another friend who is doing some really beautiful work with an Iphone... it still cant make a phone call but his photos are great ~

[Ben_Edict]

Yes, but you've got to think through all the implications of a crop. If you are stating that a 40/2.8 yields the same DOF independently of sensor format than your are not thinking through the implications of a crop that is enlarged to yield the same size as the bigger format. Marc said it all.

Well, I hope you agree that what DOF range a lens gives you is significant.

A 40/2.8 on APS-C will not give you the same DOF options as a 60/2.8 on FF. Most people are used to apply the crop factor to the focal length (40 -> 60 FF-equivalent) but fewer don't do this for the maximum aperture f-ratio (f/2.8 -> 4.2-FF equivalent).

Also, note that the FF-equivalent maximum aperture f-ratio is also the one to be used if you want to find out how "fast" (in the sense of what amount of light is allowed through by the lens, i.e., w.r.t. exposure as opposed to DOF) an FF-equivalent lens of a particular lens used on APS-C would be.

One must be careful not to confuse this with swapping the same lens from an APS-C camera to an FF camera. The equivalent focal length and widest aperture f-ratio (both obtained by multiplication with the crop factor) are meant to indicate what lens on an FF camera would produce an equivalent image.

Sorry, but I wholeheartedly disagree with your conclusions. I see the way your argument goes. But it is basically flawed: The aperture is fixed, completely independent of the camera attached to the lens. No way to circumvent that simple, basic fact. The DOF in the final image may vary. But the light gathering properties are part of the lens properties, not of the sensor size.

Especially, there is NO "equivalent widest aperture"...

I think the problem is, that you attribute the change of DOF to the lens, whereas it is only caused by a change final image format and an possibly higher magnification during the print stage if a small sensor is used for the capture. Viewed from that point, it is clear, that - following your arguments – any print or projected slide or projected digital image would change DOF, according to the size. This will make DOF a somewhat meaningless word.

Your arguments could be applied to only prints. So you could argue easily, that a 20x25 inches print shows much less DOF than a 6 x 4 inches print! Would that also mean, that a 40/2.8 lens is somehow transformed into a 180/12 lens?

I recognize, that all the talk about "equivalent focal lengthes" in all photographic commuinities leads to much more confusion, than it does to explain anything. I did use this notion myself too often, I realize now, we should really rethink that concept.

Ben

[Wheatfield]

Sorry, but I wholeheartedly disagree with your conclusions. I see the way your argument goes. But it is basically flawed: The aperture is fixed, completely independent of the camera attached to the lens. No way to circumvent that simple, basic fact. The DOF in the final image may vary. But the light gathering properties are part of the lens properties, not of the sensor size.

To me, it's one of those how many angels can we get to dance on the head of a pin arguments.

[Ben_Edict]

To me, it's one of those how many angels can we get to dance on the head of a pin arguments.

Wheatfield. it isn't. The thing is: lens properties are physical, unchangeable properties, like focal length and aperture.

Depth of field is mere convention, because it depends which size of the circle of least confusion appears acceptable. This is wholly dependent on the size of the final image and all the given sizes you'll find in the literature (for example the 0.025 mm acceptable circle of least confusion diameter for 35mm film) are based on assumptions or conventions (size of print, viewing distance), not on simple optics.

My opinion is, that it is not helpful in the least to mix hard facts (aka physical lens properties) with assumptions.

ALSO, I can still see no valid explanation, why the aperture would be different between different film formats (as it is a hard physical property of the lens). At least all the numerical aperture values remain unchanged between film formats, there is simply no other way.

Ben

P.s.: a personal note: I just thought, I am getting as grumpy in my answers as you sometimes have been lately! Haha…

[Marc Sabatella]

But I completely fail to see the point. I see no application where you could possibly use this idea of equivalence, in regards to DOF at least.

OK, *now* we're talking the same language. That's why I too questioned the *relevance* (not the *accuracy* of the comparison). As presented, the only value I can see is for the benefit of someone with no APS-C experience but tons of FF digital experience, who hears about a 40/2.8 lens and wonders what it be like to use. Then, the comparison to a 60/4.3 on FF digital would be enlightening. However, I don't know that many people here on this forum are in that boat.

For me, the idea has the one real-world relevance I alluded to: it helps me sort out my own upgrade possibilities. I do a lot of concert photography, mostly using short-medium telephoto primes are around f/2.8. I keep hearing how much better FF is, and I understand the reasons why, and then start wondering, OK, if I were to go FF, what lenses would I need to actually realize the potential improvement. And the answer is, I need a 105/3.5 to give me "equivalent" results to my 70/2.4, a 150/4 to match my 100/2.8, and a 200/4.5 to match my 135/3.5. To actually see a full stop of improvement in terms of noise for a given shutter speed or vice versa, I'd need a 105/2.4, a 150/2.8, and a 200/3.5. That's assuming, of course, that all of this were available in a stabilized system.

So being able to use equivalence in this answers some questions for me I'd find it very difficult to answer any other way short of renting a FF body and selection of lenses to try for a while. Instead, I can see right away tht getting any significant advantage from FF would involve using lenses whose the combined size and weight of those lenses would probably more double what I use now, and the cost would be even more out of line.

I would imagine others could apply the same ideas (and as likely as not come to the exact opposite conclusion given their own personal shooting preferences). As a way of considering future purchases - to evaluate the worth of a format & lens you are not familiar with - it's a handy tool. But you're right that it doesn't have a heck of a lot of applicability in actual shooting. Or, indeed, to the main gist of this thread.

A 50mm f/1.8 lens doesn't "become" a 75mm f/2.7 lens, it stays the exact same lens no matter where it is, you just get a smaller crop of it. This crop (and subsequent upscaling) increases everything, as any kind of crop would do, be it a sensor crop or a crop on the computer, including chromatic aberrations and 'blurriness' (making the DOF more shallow).

Oh sure, and of course, we'd be talking about two different lenses anyhow, so they are going to have different resolution characteristics, different amounts of vignetting and distortion, different bokeh, and so on. When one gets down to that level of nitty-gritty, obviously it makes no sense to compare a real flash-and-blod (well, glass-and-aluminum) 40/2.8 lens against a completely hypothetical 60/4.3 lens. but just to get a ballpark idea of its capabilities in terms of DOF and shutter speed / noise tradeoffs, it can be a remarkable tool. But again, in this particular context, it's really only relevant to someone who 8does* understand what a 60/4.3 would be like on FF but *doesn't* understand what a 40/2.8 would be like on APS-C. As it is, I think the comparison was about as useful to people to this forum as if someone were to offer to translate this whole thread into Swahili :-)

[Marc Sabatella]

But it is basically flawed: The aperture is fixed, completely independent of the camera attached to the lens. No way to circumvent that simple, basic fact.

Obviously. But the point of equivalence is, it is not the actual physical aperture (or even aperture ratio!) that we care about in photography. It is the *effect* this has on DOF and on what kind of shutter speed we can get for a given level of noise. And that's the point of "equivalence" - to tell us what focal length and aperture will give the same DOF/shutter/speed/noise characteristics betwene two different formats.

This is not something that "Class A" or I just pulled out of our hats for the purpose of this discussion. It's a widely reported and well-researched topic. there are any number of web sites out that will walk you through the calculation if you need to see the mathematical proof of this. But the proof *is* there, and I can tell you as a person with a degree in mathematics, while I've bothered to work through it all myself, it looks absolutely 100% rock solid.

I think the problem is, that you attribute the change of DOF to the lens, whereas it is only caused by a change final image format and an possibly higher magnification during the print stage if a small sensor is used for the capture. Viewed from that point, it is clear, that - following your arguments – any print or projected slide or projected digital image would change DOF, according to the size. This will make DOF a somewhat meaningless word.

Not at all. It just means that you have to establish the constraints under which it is to have a fixed meaning. And in this case, the constraints that you must make the *same sized* print (or monitor display) from both cameras. If you do that, you do indeed end up with more magnification on APS-C than on FF - and that's taken into account in the math that defines equivalence.

Your arguments could be applied to only prints. So you could argue easily, that a 20x25 inches print shows much less DOF than a 6 x 4 inches print!

Precisely. Are you suggesting this is not in fact true?

Would that also mean, that a 40/2.8 lens is somehow transformed into a 180/12 lens?

No idea what you're talking about here. Why would you even think about making a statement about the lens when all you've done is change print size? I suppose you *could* say, when printing 20x25, you'd need f/12 to yield the same DOF as you get printing 4x6 at f/2.8 (all else being equal, and assuming these numbers are actually correct). That would be a *different* form of equivalence. It might interest people interesting in comparing DOF at different print sizes for the same format size, if there were enough people sufficiently interested in that topic to be worth going through the math. But the type of equivalence being discussed here is not comparing DOF different *print* sizes; it's between different *format* sizes - and that's a topic that has interested enough people to have made it worthwhile for others to go through the math and yield an answer that can be used to say with absolute certainly what focal length and aperture you'd need on one format to match the DOF (and shutter speed / noise) of a given focal length and aperture on another format.

[wallyb]

For me, the idea has the one real-world relevance I alluded to: it helps me sort out my own upgrade possibilities. I do a lot of concert photography, mostly using short-medium telephoto primes are around f/2.8. I keep hearing how much better FF is, and I understand the reasons why, and then start wondering, OK, if I were to go FF, what lenses would I need to actually realize the potential improvement. And the answer is, I need a 105/3.5 to give me "equivalent" results to my 70/2.4, a 150/4 to match my 100/2.8, and a 200/4.5 to match my 135/3.5. To actually see a full stop of improvement in terms of noise for a given shutter speed or vice versa, I'd need a 105/2.4, a 150/2.8, and a 200/3.5. That's assuming, of course, that all of this were available in a stabilized system.

The only problem is: why? Why would you want to think that? It seems like an exercise in playing with numbers for a mathematical purpose, rather than a photographical purpose. You're never going to be able to re-shoot the same concert again in the exact same way, so who's to say you wouldn't want the keyboard in the background of the singer a little more or less in focus (at a narrower or wider aperture) than last time, for example? It's not something immediately applicable and important like "oh, my 50mm lens is going to act like a 75mm one, so I should step back a bit further to get a full-body shot of the guitarist", it is simply a numbers game that I still don't see the practical use for.