Circle of confusion

[witz]

The way I see it is.... but I reserve the right to be wrong!

the larger the format... the more angle of view you have at a given distance ( D ) and focal length ( L ). The DOF ( F )  is determined by the F and is the same on any format at the same D & L.

see attached for a basic reference.

The advantage of smaller formats is that lens' can be made "faster" ( bigger F.. ie f1.2 ) because the image circle can be smaller and the distance between lens and film/sensor is shorter... therefore allowing one to use wider lens' at faster F's. And obviously a 1ds3 is easier to run about and shoot with.

The advantage of larger formats is that the cost of resolution is much cheaper ( except for extreme wide angle )... it's amazing to me that the canon 14mm 2.8 has about the same "field of view" as the infamous Hypergon for 8X10... at around 115º!

I went through a stint back in the late 80's where I ran around shooting "real world" type images in san francisco with a speed graphic 4X5 and an SA 65mm lens ( recessed board of coarse )... It was fun.... but I had to shoot hyper-focal to guaranty a high success rate of good images.  but now with the 1ds3 I'm convinced that my options are much better for more situations. I just love the 35mm f1.4.... as much as I loved my 50 2.8 on my old hassy 2000fcw.

[Morgan_Moore]

Yoiur attached image

Could you choose a wider lens on the Canon, open it up and recreate the big chip look - that is the question

I think in this case (f2) you couldnt because the lens speeds are not there, but this is a practical limitation

Is there a theoretical limitation too ?

S

[01af]

The DOF ( F )  is determined by the F ...[a href=\"index.php?act=findpost&pid=180224\"][{POST_SNAPBACK}][/a]

F is determined by F?

... and is the same on any format at the same D & L.[a href=\"index.php?act=findpost&pid=180224\"][{POST_SNAPBACK}][/a]

No, it's not. At the same distance and focal length (and same aperture, of course), DOF generally is smaller with smaller frame formats.

-- Olaf

[witz]

well... digging around I found this...

"Whenever the depth of field is compared between lenses, formats, or whatever, a complete framework must be sketched within which the comparison takes place. If not, the comparison is meaningless."


http://www.vanwalree.com/optics/dof.html

it makes my eye's hurt reading it.... They are probably correct. hehe.

looks like every lens has it's own qualities of DOF.... and the math can't be generalized.

[01af]

Looks like every lens has its own qualities of DOF... and the math can't be generalized.[a href=\"index.php?act=findpost&pid=180246\"][{POST_SNAPBACK}][/a]

Well ... yes and no. The math generalizes and idealizes, otherwise the formulas would become *WAAAY* more complex than they are. So when analyzing formulas, we effectively are looking at "ideal" lenses, i. e. lenses with no aberrations. In reality, such lenses don't exist, so the idealzied math doesn't perfectly reflect all the the actual properties of the lenses in our bags. Still it's not pointless; it does provide a framework which outlines the basic properties, behaviours, effects, and limitations of the optical laws and thus describes the lenses and their images properly, within limits.

-- Olaf

[Ray]

Umm---first you say you disagree ... and then you just repeat what I said, only in greater detail. Where's the dissent?

Olaf,
The dissent is in relation to your claim that the 'counting horses teeth' analogy is not a meaningful analogy because it's a simpler process.

I think it's a perfect analogy. When we were all there at the farmer's field in the early hours of the morning about 5,500 years ago, having had some difficulty in catching the horse which seemed a bit shy, there was some doubt as to whether the filly was in possession of a full set of teeth. The counting was made difficult by the horse shaking it's head and struggling to get away, although I must say that Plato and some of his burly students did a pretty good job holding that head still and keeping the mouth open. Aristophanes did his best holding the lantern, but when the torch was brought too close to the horse it really struggled, so we were never really sure just how many teeth that horse really had.

Nevertheless, the group decided that the count was 38 and that the matter was resolved. I protested that we couldn't draw any firm conclusions from this single, rather ambiguous result and that we should count the teeth of a few more horses. I conjectured that perhaps not all breeds of horse have the same number of teeth. The horse whose teeth we counted was a filly. Perhaps a fully grown horse has a greater number of teeth, or perhaps a stallion has a different number of teeth than a mare.

Socrates, who was older and wiser than the rest of us, agreed with me. However, the rest of the group seemed to have lost its enthusiasm for this entire process of gathering data, perhaps understandably because at least one of us was almost seriously injured by a kick from that filly.

I got the impression that most of us would rather sit in a confortable chair and speculate on such matters, and gather mathematical formulas and other opinions to support our preconceptions, instead of engaging in the dangerous activity of gathering real data.

[witz]

There are some odd horses out there though.... for instance, when I used to shoot jewelry with 4X5 I'd reverse my 210mm and it would have an odd effect on DOF... there was more room in front than in back rather than the normal more infocus behind than in front usual. So I'd imagine that different lens designs will be different? I'd guess that maybe iris blades will effect the bukeh as well?

but... that 110mm of mine on both MFDB and FF35 gave the same result of DOF.... but different FOV.

Fun topic! as I'm an addict to SDOF and always have been. One of the first questions I ask an art director is " do you want it all in focus or do you want it artsy"?

[Ray]

well... digging around I found this...

"Whenever the depth of field is compared between lenses, formats, or whatever, a complete framework must be sketched within which the comparison takes place. If not, the comparison is meaningless."
http://www.vanwalree.com/optics/dof.html

it makes my eye's hurt reading it.... They are probably correct. hehe.

looks like every lens has it's own qualities of DOF.... and the math can't be generalized.
[a href=\"index.php?act=findpost&pid=180246\"][{POST_SNAPBACK}][/a]

The article you refer to might appear to be confusing in relation to the narrow topic we are discussing here. It covers situations of different perspective, same f stop with different focal lengths, same magnification of main subject with different focal lengths on the same camera from different distances ( ie. different perspective).

The bone of contention in this thread, is a very specific one. Namely, having done apparently the right thing in adjusting focal length and f stop number in proportion to sensor size, and having ensured that the lenses on both formats are roughly comparable in quality, and having ensured that the sensors of the different formats have roughly the same pixel count and therefore the same resolving power, and having ensured that the shooting position (perspective) is unchanged in all comparisons, are there any significant differences in perception of DoF amongst the different formats in same size images or prints?B]

[AJSJones]

Think about this and it may answer your question..

If you shoot a pic on a FF camera with a certain lens and aperture and crop that image in PS down to 'crop sensor' size the DOF doesnt magically change - you just get less of the image

[a href=\"index.php?act=findpost&pid=180163\"][{POST_SNAPBACK}][/a]

This may be off topic by now, but it wasn't commented on  
The following is a version of that that I would agree with:

If you shoot a pic on a FF camera with a certain lens and aperture and make two 8x12 prints and crop the second print down in proportion to 'crop sensor' size (e.g. down to 5x7.5 for a 1.6 sensor) and view them from the same distance the DOF doesnt magically change - you just get less of the original print to look at.


Correspondingly, if you make an 8x12 after cropping in PS and view them from the same distance, the DoF will be different, due to the different degree of enlargement, no magic required

[Ray]

This may be off topic by now, but it wasn't commented on   
The following is a version of that that I would agree with:

If you shoot a pic on a FF camera with a certain lens and aperture and make two 8x12 prints and crop the second print down in proportion to 'crop sensor' size (e.g. down to 5x7.5 for a 1.6 sensor) and view them from the same distance the DOF doesnt magically change - you just get less of the original print to look at.
[a href=\"index.php?act=findpost&pid=180329\"][{POST_SNAPBACK}][/a]

Put even more simply, cropping is cropping, whether it's done by the camera or in Photoshop. But, as you admit, that's not the issue we're discussing here.

[Morgan_Moore]

Put even more simply, cropping is cropping, whether it's done by the camera or in Photoshop. But, as you admit, that's not the issue we're discussing here.
[a href=\"index.php?act=findpost&pid=180333\"][{POST_SNAPBACK}][/a]

Im not sure I'm with this

A crop of a FF image is the same as a Crop camera used with the same lens (etc) from the same point

How is it not one of the factors in the issue being discussed?

S

[01af]

The dissent is in relation to your claim that the 'counting horses teeth' analogy is not a meaningful analogy because it's a simpler process.[a href=\"index.php?act=findpost&pid=180308\"][{POST_SNAPBACK}][/a]

Since I never made any claim like that, I still cannot see any dissent.

I think it's a perfect analogy.[a href=\"index.php?act=findpost&pid=180308\"][{POST_SNAPBACK}][/a]

Well, not perfect ... but pretty good indeed. It would be perfect if, in addition to all the  hassles of taking a deep look into a struggling horse's mouth, there were objects to be found in horses' mouths than would easily get confused with teeth. So even if, against all odds, you managed to come up with a count of tooth-like objects, you'd still be unsure how many of them actually were teeth.

And if counting several horses' teeth and finding different numbers, you'd still not know if some horses had simply lost some of their teeth, or if you miscounted in some cases (and if so, in which?), or if the difference was due to age, sex, or breed.

... so we were never really sure just how many teeth that horse really had.[a href=\"index.php?act=findpost&pid=180308\"][{POST_SNAPBACK}][/a]

Exactly.

I got the impression that most of us would rather sit in a confortable chair and speculate on such matters, and gather mathematical formulas and other opinions to support our preconceptions, instead of engaging in the dangerous activity of gathering real data.[a href=\"index.php?act=findpost&pid=180308\"][{POST_SNAPBACK}][/a]

But gathering mathematical formulas isn't the same as disregard for real data! To the contrary---both strategies complement one another. Actually, my endeavour to improve that COC formula was triggered by the evidence found in empirical data.

... when I used to shoot jewelry with 4 × 5" I'd reverse my 210 mm and it would have an odd effect on DOF. There was more room in front than in back ...[a href=\"index.php?act=findpost&pid=180311\"][{POST_SNAPBACK}][/a]

That's impossible. Obviously your plane of focus wasn't where you thought it should be. Umm ... did you focus at full aperture and then stopped down for the shot? Most likely, your 210 mm lens has some aperture-related focus shift (not uncommon).

But ... that 110 mm of mine on both medium-format digital back and 35-mm format gave the same result of DOF ... but different FOV.[a href=\"index.php?act=findpost&pid=180311\"][{POST_SNAPBACK}][/a]

And different (relative) magnification, too. So DOF isn't comparable.

-- Olaf

[01af]

The article you refer to might appear to be confusing in relation to the narrow topic we are discussing here.[a href=\"index.php?act=findpost&pid=180321\"][{POST_SNAPBACK}][/a]

Exactly. The article referred to by "witzke"---which is very good by the way---covers DOF in great width. The topic we are discussing here is only one tiny aspect of all that ... it's an aspect the article covers with just one casual remark. About three quarters into the article, a few paragraphs above Fig. 9, it says:

The smaller format employed at an F-number N yields the same DOF as the larger format at an F-number of R × N. Please note that this rule of thumb is just a rule of thumb, not a mathematically exact relationship. It should only be applied at intermediate and long object distances, because at close range the equivalence breaks down.

Here, we are discussing how exactly that equivalence is breaking down, and how that rule of thumb might be replaced by a mathmatically exact relationship.

-- Olaf

[Ray]

The smaller format employed at an F-number N yields the same DOF as the larger format at an F-number of R × N. Please note that this rule of thumb is just a rule of thumb, not a mathematically exact relationship. It should only be applied at intermediate and long object distances, because at close range the equivalence breaks down.

Here, we are discussing how exactly that equivalence is breaking down, and how that rule of thumb might be replaced by a mathmatically exact relationship.

-- Olaf
[a href=\"index.php?act=findpost&pid=180389\"][{POST_SNAPBACK}][/a]

Shouldn't we first gather some evidence demonstrating that the equivalence actually is breaking down and to what extent it is breaking down. I never noticed this breaking down of the rule of thumb at the close range of my own tests, so how close is close? Macro?

[AJSJones]

Put even more simply, cropping is cropping, whether it's done by the camera or in Photoshop. But, as you admit, that's not the issue we're discussing here.
[a href=\"index.php?act=findpost&pid=180333\"][{POST_SNAPBACK}][/a]

I guess my point was that it does matter whether you crop before you enlarge or after you enlarge, vis-a-vis the DoF, because the CoC gets enlarged in the former case but not the latter.  Whether that affects the transition from in focus to OOF, as I understood the scope of the original post, I haven't the faintest

[Ray]

I guess my point was that it does matter whether you crop before you enlarge or after you enlarge, vis-a-vis the DoF, because the CoC gets enlarged in the former case but not the latter.  Whether that affects the transition from in focus to OOF, as I understood the scope of the original post, I haven't the faintest
[a href=\"index.php?act=findpost&pid=180446\"][{POST_SNAPBACK}][/a]

It's understood that perceptions of DoF change with print size and viewing distance. If you enlarge a crop of an 8x10 enlargement back to 8x10 size, then the result is effectively a crop of an A3+ size, with respect to the original FoV and image.

[AJSJones]

It's understood that perceptions of DoF change with print size and viewing distance. If you enlarge a crop of an 8x10 enlargement back to 8x10 size, then the result is effectively a crop of an A3+ size, with respect to the original FoV and image.
[a href=\"index.php?act=findpost&pid=180494\"][{POST_SNAPBACK}][/a]

Agreed - which is why the comment  from Sam

"If you shoot a pic on a FF camera with a certain lens and aperture and crop that image in PS down to 'crop sensor' size the DOF doesnt magically change - you just get less of the image"

looked possibly incorrectly "understood" without the necessary qualifiers...