Blurr Background

[RobertsMx]

There is no ‘argument’ about shells and frogs, merely a statement that we are moving away from those examples to another example using the OP’s camera
However, it was previously pointed out that conclusion from the example using of the three shells is flawed.

The concept of DoF is universal, not limited to one camera or other, one sensor size or another. As for the example using shell, what exactly is the flaw? Spelling it out will help have a conversation.

No.
That assertion is incorrect. Definitively NOTHING like that was ever written nor even implied.
There is definitely a need for a re-read.
It is quite clear what was written and meant.
Below is the extracted section of what WAS written and attention is drawn the underlined and bolded section for reference:

An argument based on fictitious setting that can't be achieved with the specific camera, won't get you the desired results. There is a reason the lens is rated: 4.3-129mm f/3.4-5.8. It is a variable aperture lens.

The DoF will be about 10 inches.
That was the point of the example - and the whole point of my original post.
Thanks for asking the question.
WW

It won't be because at 129mm, the max aperture won't be f/3.4 but f/5.8.

[William W]

You really do need t re-read what was written.

The flaws with the example of the three shells is that the text stated that the image #3 has a shallower DoF than Image #1 and it was asserted that the image supported that. There are two flaws:

- firstly that it is a mathematical impossibility for Image #3 to have a shallower DoF if the two images were framed identically

- secondly Image three is taken at a different camera elevation and that skews the imaging of the tile pattern, which it was assumed gives credence to the illusion of a shallower DoF

BOTH these flaws were previously pointed out.


***

There are no 'fictitious settings'

AGAIN please re read what was written - the text is written in very simple English and it suggests two things

FIRSTY take a picture at the longest Focal length and at F/5.8.
SECONDLY ----- Z O O M the lens so that you can get to F/3.4 and then F R A M E the image the same and make a second shot.

There is NEVER any mention of needing F/3.4 at the longest Focal Length.


***


There is an appearance of great confusion here.

It seems confusion is being caused by NOT reading what was written and then making up stories to argue against.

It is impossible to continue any sensible conversation under those conditions.

Accordingly, I withdraw from the conversation with you, having posted this last simply for the archival record of the thread.

WW

[RobertsMx]

You really do need t re-read what was written.

The flaws with the example of the three shells is that the text stated that the image #3 has a shallower DoF than Image #1 and it was asserted that the image supported that. There are two flaws:

- firstly that it is a mathematical impossibility for Image #3 to have a shallower DoF if the two images were framed identically

- secondly Image three is taken at a different camera elevation and that skews the imaging of the tile pattern, which it was assumed gives credence to the illusion of a shallower DoF

BOTH these flaws were previously pointed out.

And allow me to correct the assumptions.
-The images were shot at 35mm f/2, 70mm f/4 and 70mm f/2.8. You should expect the third to be shallower for a closely matched composition (the idea was to ensure same subject size), mathematically or otherwise. Why won't you?
-Except that the perspective changes so you're bound to have some differences. But you can't use that against the issue under discussion: DoF. The idea was to keep the subject size identical which is critical, not the slight difference in angle (and I did tell you that it was a quick take).

There are no 'fictitious settings'

AGAIN please re read what was written - the text is written in very simple English and it suggests two things

FIRSTY take a picture at the longest Focal length and at F/5.8.
SECONDLY ----- Z O O M the lens so that you can get to F/3.4 and then F R A M E the image the same and make a second shot.

There is NEVER any mention of needing F/3.4 at the longest Focal Length.

You calculated 18 in as the DoF at 129mm, f/5.8. And recommended that using f/3.4 would give you 10 in. The recommendation would be true, if you could do that at 129mm and you cannot. Hence the fictitious specifications.

[Stagecoach]

Manu

If 18" is the DOF at f5.8 at a lens FL of 129mm and you then zoom the lens in to achieve an aperture of f3.4 you will have reduced this lens FL. By reducing your lens FL you will have to move in closer (walk) to the subject to achieve the same image frame.

Having moved in closer to your subject with the new aperture of f3.4 with the shorter subject distance the DOF will have decreased significantly to around what Bill has quoted.

[William W]

The images were shot at 35mm f/2, 70mm f/4 and 70mm f/2.8. You should expect the third to be shallower for a closely matched composition (the idea was to ensure same subject size), mathematically or otherwise. Why won't you?

I said I wouldn't but I cannot resist - this is the last time:

Please reference the underlined sections which I have quoted.

Please understand that I am not shouting when I use capitals - they are used to emphasize key words


OK so we agree that the first Shell Image was taken at F/2
AND the THIRD Shell image was taken at F/2.8
AND we want BOTH those images to be FRAMED EXACTLY the same.

Clearly IF those two images are FRAMED the EXACTLY the same - we would expect the FIRST image to have a shallower Depth of Field.

This is because the first image was taken at F/2 - - - which is a BIGGER aperture than F/2.8

This is the concept and the AXIOM of DEPTH of FIELD which is the topic of most of this thread.

The answer to why won't I accept that the third shot should have a shallower DoF than the first is because it won't. To say it will, is simply: incorrect.



***


Now for the second bit - - -

You calculated 18 in as the DoF at 129mm, f/5.8. And recommended that using f/3.4 would give you 10 in. The recommendation would be true, if you could do that at 129mm and you cannot. Hence the fictitious specifications.

How many times do I have to state that I did NOT write using F/3.4 at 129mm?

How many times do I have to draw your attention to what was written - - - which was;

"SECONDLY ----- Z O O M the lens so that you can get to F/3.4 and then F R A M E the image the same and make a second shot."

So that means after you take the first shot at 129mm you zoom the lens UNTIL F/3.4 is available to use - - - that might be at whatever focal length it is - it is the Focal Length where F/3.4 becomes available.

AND then you FRAME the IMAGE the same size.

SO you end up with TWO shots: ONE at 129mm shot at F/5.8 and a second at whatever Focal Length, shot at F/3.4.

And again that is what the majority of this thread is all about . . . the fact that the Focal Length - doesn't matter if the shots are FRAMED the same size.

WW

[William W]

Thanks very much, Grahame: very much appreciated. You were typing as I was making my last reply . . .

It seems that I cannot explain that concept.

And also it seems I cannot explain the concept as to why Shell Image #3 cannot possibly have a shallower DoF than Shell Image #1.

I hope Manu gets it, now, reading your different words.

WW

[nimitzbenedicto]

To put it in simple words:

Manu is wrong!

[RobertsMx]

Manu

If 18" is the DOF at f5.8 at a lens FL of 129mm and you then zoom the lens in to achieve an aperture of f3.4 you will have reduced this lens FL. By reducing your lens FL you will have to move in closer (walk) to the subject to achieve the same image frame.

Having moved in closer to your subject with the new aperture of f3.4 with the shorter subject distance the DOF will have decreased significantly to around what Bill has quoted.

Then you're completely ignoring the composition, the size of the subject. Perhaps you can demonstrate your point with a pair of samples?

[RobertsMx]

OK so we agree that the first Shell Image was taken at F/2
AND the THIRD Shell image was taken at F/2.8
AND we want BOTH those images to be FRAMED EXACTLY the same.

Clearly IF those two images are FRAMED the EXACTLY the same - we would expect the FIRST image to have a shallower Depth of Field.

This is because the first image was taken at F/2 - - - which is a BIGGER aperture than F/2.8

This is where your argument falls on its face. Aperture size:
35mm f/2: 17.5mm
70mm f/2.8: 25mm

17.5mm is not bigger than 25mm, and if you keep match the subject size, you will get shallower DOF with larger aperture (and 35mm f/2 isn't it).

Now for the second bit - - -

How many times do I have to state that I did NOT write using F/3.4 at 129mm?

How many times do I have to draw your attention to what was written - - - which was;

"SECONDLY ----- Z O O M the lens so that you can get to F/3.4 and then F R A M E the image the same and make a second shot."

So that means after you take the first shot at 129mm you zoom the lens UNTIL F/3.4 is available to use - - - that might be at whatever focal length it is - it is the Focal Length where F/3.4 becomes available.

AND then you FRAME the IMAGE the same size.


SO you end up with TWO shots: ONE at 129mm shot at F/5.8 and a second at whatever Focal Length, shot at F/3.4.

And again that is what the majority of this thread is all about . . . the fact that the Focal Length - doesn't matter if the shots are FRAMED the same size.

WW

And for the second piece (following from the first), your idea would only work if the corresponding FL with f/3.4 is at least 76mm.

[William W]

Depth of Field formulae are computed using the aperture ratio, i.e. the F-Number, not the absolute linear size.

WW

[Stagecoach]

Manu,

Unclear what you are referring to when you comment that I am 'completely ignoring the composition'?

My reference to 'achieve the same image frame' means that at both focal lengths (129mm and smaller say 40mm) I am filling the frame with the same subject composition height and to achieve this I had to move in closer to my subject.

Not much point demonstrating this with samples as feeding the figures through standard FoV and DoF calculators will confirm these basic facts.

[jcuknz]

A good try Manu But some things cannot be re-written .

[RobertsMx]

Manu,

Unclear what you are referring to when you comment that I am 'completely ignoring the composition'?

My reference to 'achieve the same image frame' means that at both focal lengths (129mm and smaller say 40mm) I am filling the frame with the same subject composition height and to achieve this I had to move in closer to my subject.

Not much point demonstrating this with samples as feeding the figures through standard FoV and DoF calculators will confirm these basic facts.

Well, I do stand corrected on DoF, after all, getting closer at shorter FL will address the DoF issue (reduce DoF). However, background blur will be a different issue which is demonstrated in my example, where you will get a greater separation with longer FL. Also note that at shorter FL (40mm, f/3.4 for 1/2.3" sensor), the hyper focal distance will be substantially shorter than at 129mm, f/5.8 which will also have an impact on the blur.

So if you want a more blurred background with a portrait , which would you choose?
1- Shorter FL but moving closer
2- Longer FL

[William W]

Depth of Field:
A Practical Exercises to do, to explain the Axiom of Depth of Field.
As to why for any same FRAMING of a SUBJECT and for any one CAMERA FORMAT the DoF will remain constant for the SAME APERTURE, irrespective of the FOCAL LENGTH of the LENS.

EXERCISE #1:
We will take two pictures of the same object, for example a picture of a Shell, using the same camera for each picture.
The first picture we use a 35mm and set the aperture to F/2.
The second picture we use a 70mm and set the aperture to F/2.8.
We must ensure that the Object is in EXACTLY the SAME FRAMING for BOTH pictures.
Assuming this is NOT macro Photography and the Object at a normal Subject Distance, that is to say not beyond the Hyperfocal Distance, the question is:
a) Which image will have the SHALLOWER Depth of Field?
b) Provide some proof of your answer.


ANSWER:

a) The Picture taken at with the 35mm Lens and at F/2 will have a SHALLOWER DoF.

b) A practical proof substituting into the Formula for DoF at common Shooting Distances.\

1. The formula:
DoF ≈ (2ACF^2S^2) / (F^4 – A^2C^2S^2)
where:
A = Aperture of Lens expressed as the f-number
C = Circle of Confusion
F = Focal Length of the Lens
S = Shooting (Subject) Distance – i.e. the distance to the Plane of Sharp Focus.

2. The camera Format and Circle of Confusion:
Let an APS-C Camera be used.
Let the Circle of Confusion = 0.016mm
Note that the Shooting Distance to EXACTLY frame with a 70mm will be exactly TWICE that when we frame with a 35mm lens.

3. The 35mm lens used at F/2:
Let us frame this shot at a shooting distance of 2000mm:
Therefore:
A = 2
C = 0.016
F = 35
S = 2000


4. Computation for the DoF for the 35mm lens when it is used at F/2:
DoF35/2

≈ 313600000 / (1500625 – 4096)
≈ 313600000 / 1496529
≈ 209.55mm

5. The 70mm lens used at F/2.8
This shot must be framed at a Shooting distance of 4000mm:
Therefore:
A = 2√2
C = 0.016
F = 70
S = 4000

6. Computation for the DoF for the 70mm lens when it is used at F/2.8:
DoF70/2.8

≈ 7095957200 / (24010000 – 32768)
≈ 7095957200 / 23977232
≈ 296.00mm


By the way - - -
[Answers using the ‘f=calc’ DoF calculator, which is available for download from the www:
Answer for 35mm lens at F/2 on APS-C camera shooting at 2000mm * DoF = 202.8mm
Answer for 70mm lens at F/2.8 on APS-C camera shooting at 4000mm * DoF = 286.4mm]

I can derive that formula, from first principles, but it will be sufficient just to post it as is - as the bona fide of it can surely can be checked in any Standard Text.

Also the Axiom can be proved (for all situations) with a moderate level of algebra - but it is too time consuming to manifest the Equations to display nicely in line, here.

***

Background Blur / Quality of / Bokeh

Post #6

Where the FL becomes relevant, (but still only in a minor way), is when one considers the quality of the blur (Bokeh). But the OP did not ask that but rather only asked about how to make the background blurred.

I see no evidence of greatly different "Bokeh" in any samples in this thread comparing Telephoto lenses with normal lenses.

Also and more important to THIS THREAD: the great incapacity of small format cameras to make much, if any at all Background blur, when making any PORTRAIT shots wider than a tight head shot, kind of makes even thinking about Bokeh - moot.

Some may choose to disagree.


***

Separation (of the Subject):

Is in no way the sole priory of the Lens's Focal Length.

Separation is a multifaceted animal and also scene dependent.

Subject to Background Distance is usually more important than Focal Length of the lens.

And also, if the background is close the Aperture to create the Shallowest DoF is of course, very important.

For example - the Subject 'Separation' here is solely the factor of the LARGE APERTURE (F/3.5) creating a VERY SHALLOW DoF:



. . . there is no way one would want to go for longer lens and have to use F/5.6 or F/5.8 and double or more that slim DoF.

Others might choose to disagree.

WW

[Clactonian]

So glad that I have a brain that is untroubled by thought.

[Stagecoach]

Well, I do stand corrected on DoF, after all, getting closer at shorter FL will address the DoF issue (reduce DoF). However, background blur will be a different issue which is demonstrated in my example, where you will get a greater separation with longer FL. Also note that at shorter FL (40mm, f/3.4 for 1/2.3" sensor), the hyper focal distance will be substantially shorter than at 129mm, f/5.8 which will also have an impact on the blur.

So if you want a more blurred background with a portrait , which would you choose?
1- Shorter FL but moving closer
2- Longer FL

Manu,

Pleased that you now understand the affect that after the initial shot by then physically moving in closer + maintaining the same subject framing, by reducing lens FL + widening the aperture from the original gives you a shallower DoF.

As for the portrait background question this is another subject on its own where other factors come into account of which are of equal if not greater importance than just achieving max background blur to a portrait photographer. I'm not one of these but if you are looking for some advice and views it may be worthwhile you starting a separate thread on the subject.

[RobertsMx]

I see no evidence of greatly different "Bokeh" in any samples in this thread comparing Telephoto lenses with normal lenses.

You saw some of it comparing blur between 35mm f/2 and 70mm f/2.8 (and 70mm f/4) and in fact questioned why they looked different. You did get it right that they look different but why... is another story. Think field of view (more at smaller FL) and compression (more at longer FL), even as DoF may or may not be different.

Pleased that you now understand the affect that after the initial shot by then physically moving in closer + maintaining the same subject framing, by reducing lens FL + widening the aperture from the original gives you a shallower DoF.

Well, it is a no brainer that you'd get shallower DoF as you move closer to reduce the focusing distance. The key argument is how much can you get out of a particular camera (previous assumption being f/3.4 available at 40mm for specific calculation which may or may not be possible).

As for the portrait background question this is another subject on its own where other factors come into account of which are of equal if not greater importance than just achieving max background blur to a portrait photographer. I'm not one of these but if you are looking for some advice and views it may be worthwhile you starting a separate thread on the subject.

I don't see why this thread wouldn't suffice, since it is about background blur which applies to any situation including portraits. BTW, I'm not a proponent of max background blur, at all times. Sometimes, a blur used with context is far more useful as opposed to creating a wall like feel as here...

(Sony NEX-3, 135mm, f/5.6)
Whereas...

(Sony NEX-3, 135mm, f/5)

[William W]

You saw some of it comparing blur between 35mm f/2 and 70mm f/2.8 (and 70mm f/4) and in fact questioned why they looked different. You did get it right that they look different but why... is another story. Think field of view (more at smaller FL) and compression (more at longer FL), even as DoF may or may not be different.

NO.

Again that is a complete nonsense and simply another made up story.

There was a difference in the BACKGROUND noted between image #2 and image #3 and they should have been the SAME FIELD of VIEW.

REF Post #35:

Also (just an aside) there appears to be something strange when comparing images #2 and #3: IF these two shots were IDENTICAL except for the APERTURE, why is there such a significant difference in the width of the OoF white section: that surely cannot be explained simply by one stop of aperture change.

*

And the difference noted between image #1 and image #3 was the Camera elevation and referenced the Tiles Pattern - not about the background.


REF Post #35:

When we contrast images #1 and #3 we can note a flaw which will skew the visual reckoning: image #3 is taken from a significantly lower camera elevation than Image #1 which renders the tiled pattern differently in each image.

***

Well, it is a no brainer that you'd get shallower DoF as you move closer to reduce the focusing distance.

If that is so:
then it is becoming more apparently obvious that it is purposely wasting a lot of people's time acting for so long that this "No Brainer" wasn't understood and continuing to argue against these "No Brainer" facts.


WW

[Stagecoach]

Well, it is a no brainer that you'd get shallower DoF as you move closer to reduce the focusing distance. The key argument is how much can you get out of a particular camera (previous assumption being f/3.4 available at 40mm for specific calculation which may or may not be possible).

Manu,

Again you have attempted to read something totally different into what had been originally stated;

My post No 51 states "" My reference to 'achieve the same image frame' means that at both focal lengths (129mm and smaller say 40mm) I am filling the frame with the same subject composition height and to achieve this I had to move in closer to my subject. """

There was simply no assumption or specific calculation made using the figure of 40mm FL. It clearly states 'say 40mm' The reason for giving a numerical figure was to assist you in understanding that the FL was being reduced (widened). I also suspect that you fully understand that this example figure could have been 40,50,60,70 or 80mm which would all have been applicable to supporting the facts.


To me it has become obvious that continuing any dialogue with you on this subject is pointless as whatever is said you clearly attempt to twist and alter so obviously outside of its context.

[William W]

+1
ww