FX lens on DX body - impact on aperture not just field of view

[William W]

You have not changed my mind, just tried to bamboozle me with a lot of totally irrelevant information. Par for the course IMO

John,

To whom is the "you" that you refer?

WW

[george013]

You have not changed my mind, just tried to bamboozle me with a lot of totally irrelevant information. Par for the course IMO

Is that another way to say you don't understood what I wrote?

It's the same as William wrote, only I started with the print and a minimal visible point, William and all calculators start with the CoC. But all CoC's in the calculator are based on the same minimal printed point that's visible in a standard print size.

Maube this will help.
smallest point at viewing distance 0.25mm
print size 30x20cm

CoC for FF 36x24 sensor 0.25/(300/36)= 0.03
CoC for DX 24x16 sensor 0.25/(300/24)= 0.02
CoC for fourthird 17x13 0.25/(300/17)= 0.014
CoC for Sigma Foveon 21x14 0.25/(300/21)= 0.018 for Ted
CoC for iPhone 4.5x3.4 0.25/(300/4.5}= 0.00375

George

[DanK]

This is a great thread. Thanks to all who contributed.

However, I suspect it is more detailed than the OP needed. Dennis, if you are still tracking this, I am going to offer some simpler suggestions because your post suggests you are relatively new to all of this arcane stuff.

First, the simplest summary of the effects of format on sensor size that I have ever seen is this post by Bob Atkins It clearly explains the point one in one of the posts that it matters whether you are referring to keeping the distance the same or the framing of the image the same. Bob is a Canon shooter, so for your purposes, you have to substitute 1.5 for his 1.6.

Second, to add to Bill's comment that bokeh is not the same as DOF: a number of things go into bokeh, including the characteristics of the lens and the nature of the background. Apart from that, it is important to understand the difference between background blur and DOF. Check out this post and search for the "background blur" heading. The bottom line: the narrower the field of view, the more background blur, even if DOF is identical.

Third, while Bill is of course completely correct in his description of the effects of shooting FF rather than crop, I would add a caveat: for most people under many circumstances, it doesn't make much difference. I shoot both. I'm quite sure that if you look at my site and don't look at exif information, you won't know which images I shot with the four different bodies (3 crop, one FF) that I used for those images. That's not to say that there is no difference. There are specific uses for which I always take either my FF or my crop, if I can. But much of the time, it just doesn't make much difference.

[William W]

Hey Dennis,

. . . the effects of shooting FF rather than crop, I would add a caveat: for most people under many circumstances, it doesn't make much difference.

Good call Dan.

This comment maybe got lost - ". . . both Nikon and Canon make a very good, very affordable 85/1.8 Lens. (I have a Canon EF 85/1.8) . . . I am sure that you will be overjoyed with the Nikon 85/1.8 - it will give you heaps of pleasure on your current DX body."

BTW Bob Atkins has a "Blur Calc" tool.

WW

[jcuknz]

Is that another way to say you don't understood what I wrote?
George

Frankly I could not be bothered reading either it or Bills because it is totally irrelevant to what I wrote.

It is quite amusing that you both ignore the alternative meaning which can be taken and I am not interested is any of your very clever calculations.

The CoC depends on the angle of light approaching the sensor obtained by a given aperture and the distance between lens and sensor along with where it is measured.
Neither of these factors change when you change the camera body.
It is simple physics.
"...because the Circle of Confusion differs dependent upon the film or sensor size."
Hence I questioned the statement by Bill and I believe Manfred sees my point which both of you have missed and entered into long winded irrelevances. LOL

On the other hand I can see that my extract of the longer paragraph on its own makes sense ... English is a very flexible language.

[xpatUSA]

CoC is not a physical fact, it's a variable factor for calculation.

Well said. Whenever I read, for example, that THE (my caps) CoC for some particular sensor is 0.019 mm (that's to the nearest micro-meter !!), that's when I stop reading.

I have researched this subject considerably and have determined that the so-called CoC is whatever you want it to be in other words, as George says, it's a variable factor . . extremely variable, in fact. Shooting macro 1:1 with a simple lens? The CoC doubles from that found at infinity.

I rather like the "cone of confusion" principle in such matters, as propounded by Richard Lyon of Foveon fame:

http://kronometric.org/phot/iq/DepthOfField-Lyon.pdf

Hard going but it blows a myth or two.

My humble apologies for referring to a link where angular measure is used

[PhotomanJohn]

I have tried to keep out of this but have decided to jump in regarding COC. We have one individual who is using the "optical" definition for COC (sometimes referred to "circle of least confusion") which is the inability of the lens to focus a point source as a single point. The definition given: "The CoC depends on the angle of light approaching the sensor obtained by a given aperture and the distance between lens and sensor along with where it is measured." is correct.

Then we have several other folks talking about the "photographic" definition of COC that associated with depth of field. This has to do with objects at different distances appearing to be in focus in a standardized size of print (about 8" X 10") viewed at a fixed distance (around 12"). This COC does vary with sensor size because of the different levels of enlargement needed to produce the standardized print and is somewhere between 1/1500 to 1/1730 of the diagonal of the sensor, and for that matter, of the standardized print.

Being that this is a photographic forum, we should be discussing the "photographic" definition of COC and not confusing folks with the "optical" definition which in practice has little to do with DOF which is our primary concern.

John

[xpatUSA]

I have tried to keep out of this but have decided to jump in regarding COC. We have one individual who is using the "optical" definition for COC (sometimes referred to "circle of least confusion") which is the inability of the lens to focus a point source as a single point. The definition given: "The CoC depends on the angle of light approaching the sensor obtained by a given aperture and the distance between lens and sensor along with where it is measured." is correct.

Then we have several other folks talking about the "photographic" definition of COC that associated with depth of field. This has to do with objects at different distances appearing to be in focus in a standardized size of print (about 8" X 10") viewed at a fixed distance (around 12"). This COC does vary with sensor size because of the different levels of enlargement needed to produce the standardized print and is somewhere between 1/1500 to 1/1730 of the diagonal of the sensor, and for that matter, of the standardized print.

Being that this is a photographic forum, we should be discussing the "photographic" definition of COC and not confusing folks with the "optical" definition which in practice has little to do with DOF which is our primary concern.

John

I fully agree with the sentiment, but what do we do if "the print" does not exist?

What if someone only views their masterpieces on a disgustingly high-definition tablet? What if they only view on a 54" 2K TV? Or what about more humble folks like myself who only view on an ordinary 0.3mm dot pitch monitor?

[William W]

You have not changed my mind, just tried to bamboozle me with a lot of totally irrelevant information. Par for the course IMO

- PLUS -

Frankly I could not be bothered reading either it or Bills because it is totally irrelevant to what I wrote.

That’s an Interesting logic and conversational technique to employ: that is to conclude that others’ contribution to a conversation is totally irrelevant, without first reading those contributions.

***

It is quite amusing that you both ignore the alternative meaning which can be taken and I am not interested is any of your very clever calculations.

I don’t believe that response Post #14 did ignore the ‘alternative meaning’, but obviously that would, not be noticed if post #14 was not ever read, but for clarity and the archival integrity of the thread the quote from Post #14 is here:

Don't confuse the application of Circle of Confusion for purposes of Depth of Field calculations - with the definitive measure of the smallest blur spot achieved by a lens at the lens’s point of best focus. Traditionally the term: Circle of Least Confusion is used for the latter, to avoid this confusion.

And in any case, the whole point is this ‘alternative meaning’ is not relevant to the Opening Post, which is discussing and considering the effects of the COMPARATIVE resultant larger DoF when using a lens at a particular Aperture on an APS-C Camera as COMPARED and CONTRASTED to using it on a 135 Format Camera.

WW

(This was written before John posted #27: I thought there was still worth in posting it, now, anyway.)

[jcuknz]

Sorry PhotomanJohn but I see no difference between the theoretical types of CoC that you suggest. I like the idea of a cone with its top chopped off but do not know what such a thing is called. In any discussion it is usually illustrated by plane diagram vertical through the centre of a simple meniscus lens.

However what I originally bawked at was I believe a suggestion that changing a body immediately changes the CoC which is of course ridiculous ... it is only when you discus its relevance to depth of field and I was not taking it to that further stage which others have inundated us at extreme length with examples.

It is probably of great interest to academia, though I would feel sorry for students subjected to that, but apart for the general principle of no interest to a practical photographer. In my original training theoretical occupied just two hours a week intended to pass C&G in the 1950's.

However this is a geek site in part so I shouldn't have got involved in this thread except I was prompted by my interpretation of the original part paragraph which I thought was strange and questionable. Which it is of course unless qualified by more info than was provided.

edit .. and to check for that qualification I returned to the post and found it preceded the quote .

[William W]

However what I originally bawked at was I believe a suggestion that changing a body immediately changes the CoC which is of course ridiculous ...

I was prompted by my interpretation of the original part paragraph which I thought was strange and questionable. Which it is of course unless qualified by more info than was provided.

Understood.

***

edit .. and to check for that qualification I returned to the post and found it preceded the quote .

Thank you for at least re-reading that small portion.

Indeed. Had the sentence been read and then quoted as it was originally written and also quoted in its entirety, the meaning is abundantly clear.

Similarly, the APERTURE of the Lens does not change: what does change is the computation for the Depth of Field and that is because the Circle of Confusion differs dependent upon the film or sensor size.

WW

[Manfred M]

I like the idea of a cone with its top chopped off but do not know what such a thing is called.

Mathematicians are do not use flowery language when it comes to names; try truncated cone.

[george013]

I think the Airy disk is meant. It's the only thing that depends on the ratio diafragma diameter and image distance. Independent from the sensor size. What's meant with the optical definition like in http://en.wikipedia.org/wiki/Circle_of_confusion looks to me more depending on the lens.

George

[DennisS]

Second, to add to Bill's comment that bokeh is not the same as DOF: a number of things go into bokeh

Thank you for your response Dan but what threw me initially was the Tony Northrup's reference to aperture in the video which was misleading. I was aware that the DOF was smaller in Full Frame (not fully understanding why but Bill cleared that up) and I did not confuse Bokeh with DOF. My attraction to the 85mm was only because of the smooth bokeh - not as smooth on my 60mm Full Frame macro lens. I initially purchased the 60mm to do double duty as a macro and portrait lens but now would like a dedicated lens for portrait.

This same 60mm FX prime macro gives better bokeh but same DOF when compared with my 55-200mm DX set at 60mm and the framing is the same at that zoom on my APS-C camera. Of course, the field of view would be very close to a 90mm on a Full Frame camera - I get that.

[William W]

I find Bokeh really interesting. I don’t place too much importance on it for my own work but I do find it interesting.

As mentioned, there are many elements which make Bokeh, not just the lens. Sure there is an intrinsic ‘Bokeh value’ if you will, to each lens, but that is only where is starts.

Just as one example showing the difference in Bokeh of EF 135/2, being dependent upon three of many elements which go to make the Bokeh in the final image:

> Background TEXTURE
> Lighting on the Background
> Ratio of distances – Camera to Subject :: Subject to Background

Sample A:




Sample B:


WW
All Images ©AJ Group Pty Ltd Aust 1996~2015, WMW 1965~1996

[tclune]

The thing that I find a bit disconcerting about the idea that FF lenses on DX cameras change both field of view and depth of focus is that it seems to be counting the same effect twice, but maybe I'm wrong. If we shoot from the same place at the same focal length, we would need to crop the FF camera shot to get the same view. That act will also lower the DOF of the image, compared to the DX shot. Is that what we really mean by lower DOF? Alternatively, we can use a shorter lens on the FF to get the same crop, which will increase the DOF of the FF image. Again, aren't we really looking at one phenomenon here?

[george013]

The thing that I find a bit disconcerting about the idea that FF lenses on DX cameras change both field of view and depth of focus is that it seems to be counting the same effect twice, but maybe I'm wrong. If we shoot from the same place at the same focal length, we would need to crop the FF camera shot to get the same view. That act will also lower the DOF of the image, compared to the DX shot. Is that what we really mean by lower DOF? Alternatively, we can use a shorter lens on the FF to get the same crop, which will increase the DOF of the FF image. Again, aren't we really looking at one phenomenon here?

It's hard to explain with words, so I borrowed some figures from Wiki.

Field of view or angle of view, http://en.wikipedia.org/wiki/Angle_of_view

The angel of view is specified by the ratio image distance and the size of the sensor. The bigger the sensor, the larger the angle of view. Under the same conditions off course.

Sharpness is defined as the maximal diamater the truncated light cone may have. What that maximal size should be I explained some posts before, post 22.

Also from Wiki http://en.wikipedia.org/wiki/Circle_of_confusion

As you can see that light cone is defined as the ration of the diafragma diameter and the image distance. This is independant of the sensorsize.

Having a FF camera and cropping out a part, is something as using a smaller sensor. Your angle of view is changing. But the light cone in the second picture doesn't change.

And then the just mentioned post 22 comes in picture.

George

[tclune]

Alternatively, we can use a shorter lens on the FF to get the same crop, which will increase the DOF of the FF image.

I seem to have fallen off the reality wagon here. What I intended to say was: Alternatively, we can use a longer lens on the FF to get the same crop as the DX image, which will decrease the DOF also.

Now, all of this is hazy enough in my mind that I can't even manage to express what I intend for an entire paragraph without confusion. It is certainly possible (indeed, probable), George, that your diagrams express precisely what I am failing to grasp. Alas, I am still failing to grasp it. It still gnaws in the back of my mind as a tautology -- kind of like saying that a macro lens has higher magnification and can focus closer than a standard lens of the same focal length.

[William W]

The thing that I find a bit disconcerting about the idea that FF lenses on DX cameras change both field of view and depth of focus is that it seems to be counting the same effect twice, but maybe I'm wrong. If we shoot from the same place at the same focal length, we would need to crop the FF camera shot to get the same view. That act will also lower the DOF of the image, compared to the DX shot. Is that what we really mean by lower DOF?

{Alternatively, we can use a longer lens on the FF to get the same crop as the DX image, which will decrease the DOF also.}

Again, aren't we really looking at one phenomenon here?

For convenience, I have placed your re-written sentence in the above quote.

Maybe this will assist your understanding:

If we shoot from the same place at the same focal length, we would need to crop the FF camera shot to get the same view. That act will also lower the DOF of the image, compared to the DX shot.

If you mean that we put a 50mm lens on APS-C Camera and a 50mm lens on ‘full frame’ Camera and stand in the same spot and take two identical pictures and then we crop the FF shot: Cropping does NOT change the DoF of the image made with the FF Camera.

The DoF remains the same for any image no matter how much we crop it.

Depth of Field (BTW – we are discussing Depth of ‘Field’ not Depth of ‘Focus’) is a ‘measurement’ which is used as a guide.

We must not confuse the indicated DoF with what we actually will determine as ‘acceptable’ or ‘unacceptable’ in the final image. For example an heavily cropped image might display an unacceptable amount of Out of Focus area. So, if your workflow is to usually crop a large amount of your images all the time, then it might be a good idea to use a different DoF calculation (i.e. use a different Circle of Confusion for your DoF calculation) to get a better idea of what will be acceptable or not acceptable, for you, in your final image.

I'll just repeat the DoF calculations do NOT make a definitive measurement - it is a guide: a guide which you can easily manipulate be a better guide to suit your own workflow or taste; and this is easily done by choosing the Circle of Confusion which best suits your: Workflow; Taste and Genre of Photography.

What is important, however, is when we are discussing COMPARATIVE Depth of Field is that we keep CONSISTENCY in our formulae. And that's all we are doing now. We are discussing in general, COMPARATIVE DoF between cameras of different formats. We therefore must say we will use one particular CoC for calculations of DoF for FF cameras and another different CoC, for calculations of DoF when using APS-C Cameras.

*

Alternatively, we can use a longer lens on the FF to get the same crop as the DX image, which will decrease the DOF also.

If you mean that we use a 50mm lens on a FF camera and then a 100mm lens on a FF camera and we and the Subject both stand in the same position and we make two pictures of the Subject using the same Aperture, then yes, the picture using the 100mm lens will have a shallower DoF.

WW

[D L]

I've been following this thread, out of interest. I thought I understood this concept. But I had never considered this scenario:

If you mean that we put a 50mm lens on APS-C Camera and a 50mm lens on ‘full frame’ Camera and stand in the same spot and take two identical pictures and then we crop the FF shot: Cropping does NOT change the DoF of the image made with the FF Camera.

WW

Mind blown. It completely makes sense to me that cropping can't change the DoF.

However, if you use the exact same lens on both bodies, same aperture and same focus point, will it not project the exact same light (i.e. image) onto both sensors? The only difference being, some of the light being "missed" and not recorded around the edges of the smaller APS-C sensor. How can the DoF be different? Or is that not how it works?

I am happy to just accept the concept and go on taking pictures, but now the cogs in my mind won't stop spinning.