Choosing focal length for DOF & OOF blurring

[New Daddy]

Doing google search for information that will give me better understanding about focal length and DOF, I found two opposite, battling views about the effect of focal length on DOF at the same aperture, IN A SITUATION WHERE THE PHOTOGRAPHER MOVES CLOSER TO OR FURTHER AWAY FROM THE SUBJECT IN ORDER TO MAINTAIN THE SIZE OF THE SUBJECT SAME IN THE FRAME.

One view says the focal length in that scenario won't have any effect in the DOF (http://luminous-landscape.com/tutorials/dof2.shtml), while another view says the focal length absolutely has an effect (http://theonlinephotographer.typepad...he-sequel.html). Our own trusted Cambridge in Colour tutorial is siding with the former (https://www.cambridgeincolour.com/tu...h-of-field.htm).

I'd like to put things in perspective here: so if you are changing the distance from the subject in order to maintain the size of the subject similar, (1) the DOF is pretty much the same across different focal lengths, (2) the distribution of the DOF changes such that longer focal length has more DOF in front, and (3) the blurred area (not the focused area) may have different characteristics such that longer focal length will afford more "enlarged" blurs. (I hope (1) through (3) are correct. I've spent quite a while to get to the bottom of these.) The second article above was incorrect, because it only ran test on an object that stood behind the subject (which would go more quickly out of focus as the focal length goes up) and it conflated the DOF with OOF (the rear subject appeared relatively sharp"er" in the wider angle, but may not have been in focus after all in the technical context of measuring DOF.)

The implication? If you are concerned about the focus (which is what the DOF concept is for), then the absolute "depth" may not change but the distribution of the DOF may matter - shoot wider if you want more focus behind the subject, and shoot longer if you want more focus in front of the subject. For this reason, longer would work better if the subject is moving towards you like in sports. If you are concerned about the BLUR (or bokeh), then shooting longer would give you more pleasing effects, not because the DOF changes but because the OOF blurs have more pleasing appearance in the numerically identical, physically larger aperture of longer focal length.

Can someone confirm that my summary of understanding is correct? I'm raising this because the information available on the internet is still confusing and conflicting, there being no final arbiter to judge who's right or wrong.

[Andrew76]

Have you read the thread "Blur Background", in the Equipment Forum? I think this issue has been beaten to death there. Unless you're asking something completely different that I'm missing?

[New Daddy]

Have you read the thread "Blur Background", in the Equipment Forum? I think this issue has been beaten to death there. Unless you're asking something completely different that I'm missing?

I may not be asking anything new to you. What was the "issues" that has been beaten to death? As far as I'm concerned, my google search dug up very conflicting and equally vociferous views on the issue, so I assumed the issue is highly controversial at least in some circles.

[Andrew76]

New Daddy,

I'm not trying to cause any controversy here. From what I read in your post, you're questioning whether focal length has any effect on DoF. Throughout the body of your comments, there is no question posed, then you finish off asking: "Any thoughts?", which is where I came up with my reply.

I've thoroughly read the thread that I've referenced, and that's why I referenced it - I just thought that after 3 pages, and fifty some odd replies, that you would get your fill of 'thoughts' from reading it.

What I don't know, is what more you would expect to get than what has already been posed, which is why I asked: "Unless your asking something completely different, that I'm missing?". If I am missing something, I do sincerely apologize.

[New Daddy]

New Daddy,

I'm not trying to cause any controversy here. From what I read in your post, you're questioning whether focal length has any effect on DoF. Throughout the body of your comments, there is no question posed, then you finish off asking: "Any thoughts?", which is where I came up with my reply.

I've thoroughly read the thread that I've referenced, and that's why I referenced it - I just thought that after 3 pages, and fifty some odd replies, that you would get your fill of 'thoughts' from reading it.

What I don't know, is what more you would expect to get than what has already been posed, which is why I asked: "Unless your asking something completely different, that I'm missing?". If I am missing something, I do sincerely apologize.

Ok. Let me rephrase the question then.

[RobertsMx]

I'd like to put things in perspective here: so if you are changing the distance from the subject in order to maintain the size of the subject similar, (1) the DOF is pretty much the same across different focal lengths, (2) the distribution of the DOF changes such that longer focal length has more DOF in front, and (3) the blurred area (not the focused area) may have different characteristics such that longer focal length will afford more "enlarged" blurs. (I hope (1) through (3) are correct. I've spent quite a while to get to the bottom of these.)

I agree with this. Even if we matched the DoF and image size using a 35mm versus 135mm lens on same sensor, the telephoto lens will render a more compressed background (blur).

[William W]

. . . Can someone confirm that my summary of understanding is correct? . . .

so if you are changing the distance from the subject in order to maintain the size of the subject similar, (1) the DOF is pretty much the same across different focal lengths

Yes.

Clarifying that is for most normal Shooting Distances and for the same Format Camera.

***

(2) the distribution of the DOF changes such that longer focal length has more DOF in front

Yes.

Clarifying that you mean the longer FL lens compared to the SHORTER FL lens.

Framing (Absolute Shooting Distance change) also has an impact of the DISTRIBUTION of DoF. (see below*A)

***

the blurred area (not the focused area) may have different characteristics such that longer focal length will afford more "enlarged" blurs.

Generally a reasonable description yes.

The aperture used and the constituent of the iris will also have an effect on other characteristics which may be quite noticeable – for example an F/2.8 100 mm lens compared to an F/1.4 50mm lens when both lenses used at F/2.8 might not fit that as a generalization.

Notwithstanding this, the factor of Subject to Background Distance, has a major effect.(see below*B)

***

The second article above was incorrect, because it only ran test on an object that stood behind the subject (which would go more quickly out of focus as the focal length goes up).

Yes, I believe so.

I quickly scan read it: appears that is a major flaw of fact and process, in the authors’ commentary.

***

If you are concerned about the focus (which is what the DOF concept is for), then the absolute "depth" may not change but the distribution of the DOF may matter – shoot wider if you want more focus behind the subject, and shoot longer if you want more focus in front of the subject.

Clarifying that you mean taking a shot with a WIDE ANGLE LENS compared to taking a shot with a TELEPHOTO LENS.

Not the whole story.

As mentioned the Framing (absolute Shooting Distance) has an impact.

(See above*A)
So (for example in Portraiture) a TIGHT shot will exhibit closer to a ½ : ½ distribution of DoF as opposed to the more oft quoted ⅔ : ⅓ distribution which we shall see in a LONGER SHOT.

***

For this reason, longer would work better if the subject is moving towards you like in sports. If you are concerned about the BLUR (or bokeh), then shooting longer would give you more pleasing effects, not because the DOF changes but because the OOF blurs have more pleasing appearance in the numerically identical, physically larger aperture of longer focal length.

Bokeh is both Subjective and is also Multifaceted.

Most Sports Shooters, I expect, choose a lens for necessary reach; the fast lens speed; and the very quick AF, rather than Bokeh as the primary criterion.

I do.

***

I'm raising this because the information available on the internet is still confusing and conflicting, there being no final arbiter to judge who's right or wrong.

Yes, there is much misinformation on the internet.

The final arbiter of validity of the answers to your questions (but NOT Bokeh) is: Fact.

These facts are predicated on mathematics.

You have already cited an example by noting the flaw in the second link that you posted: and you have rendered the flaw as apparent, by the use of mathematics.

*

Bokeh is Subjective: The Amount of Out of Focus at a particular Distance and the Depth of Field can both be calculated.

Although linked in several ways: Bokeh // The amount of Out of Focus // Depth of Field - are three different topics, but are often confused and mixed up as the same.

We can calculate the amount of (un)sharpness before and beyond the limits of the Depth of Field. This is only a PORTION of what constitutes "Bokeh", as Bokeh is a measure of both the quantity and also the quality of background blur.

The calculation of the quality of the (un)sharpness cannot be made by a simple method, even if the optics of the lens is known.

Moreover each quality of (un) sharpness, will be affected by other elements which will change from scene to scene. As for two examples: the physical nature of the background and nature of the lighting on it.

(see Above*B)
Also, one fact that becomes apparent when measuring the amount of (un)sharpness area is that the amount is dependent upon the DISTANCE from the Subject to the BACKGROUND.

If the Background is quite far away from the Subject and the Subject if reasonably close to the Camera: the amount of blurring is more closely related to the Absolute Linear Diameter of the Aperture.

For example, using a 50mm lens at F/5.6 the Linear Diameter of the Aperture is 8.93mm and for a 100mm lens used at F/5.6 the ALD is 17.86mm and therefore we would assume that the DISTANT BACKGROUND BLUR of the 100mm lens used at F/5.6 to be about twice as much as the 50mm lens used at F/5.6 . . . and this is so.

On the other hand, if we use the 50mm lens (for example at F/2 on a 135 format camera) to make a FULL LENGTH Portrait, Portrait Orientation, the SHALLOW DEPTH of FIELD (about 300mm or 12 inches) will tend blur the CLOSER BACKGROUND more.

So, the relative aperture, (that is the F/stop number) and also the Absolute Linear Diameter of the Aperture will vie for precedence as to which will be the more important to determine the blurring: and this will be dependent upon the SUBJECT to BACKGROUND DISTANCE. –

But as already mentioned the degree of the Out of Focus does not fully account for BOKEH: it is only one element of it.

WW

[jcuknz]

It is also worth mentioning that only the material at the distance the lens is focused at is sharp and in front and behind the image is getting more or less sharp and this depends on the person viewing the image.... so DoF is misleading because it is not a case that one centimetre on the soft side of the distance is out of focus and 1cm the other side is in focus.

Also worth being aware of as Stagecoach brought to our notice recently that shooting from further back and cropping in editing results in more DoF than going in close for the same framing.

So aware of those points I see little point in using DoF tables except for meaningless figures in discussions.

I normally choose focal length for the framing it gives me rather than ZWYF and focus on the subject.

[RobertsMx]

Also worth being aware of as Stagecoach brought to our notice recently that shooting from further back and cropping in editing results in more DoF than going in close for the same framing.
:
I normally choose focal length for the framing it gives me rather than ZWYF and focus on the subject.

Cropping is simply an alternative to using longer FL, and not the recommended one, besides additional considerations like impact of subject magnification.

In the other thread, for example, an assumption was also being made that if you keep magnification the same but via different FL, that there would be no difference in the backdrop. That would be true if FoV also matched but then, we would have to keep the same FL (or achieve the effect via cropping).

[DanK]

I found two opposite, battling views about the effect of focal length on DOF at the same aperture

I think the answer is simple. The tutorial on this site is correct. You should also read http://toothwalker.org/optics/dof.html, which explains this with concrete examples and also explains the difference between DOF and background blur.

I'm not being snide. You can find almost any view on the internet. That doesn't mean they are correct. No, there is no arbiter, which is a real danger. You have to learn which sites to trust. The tutorials here are a very good source.

BTW, welcome to the forum. It's our custom to put our real names and locations in our profiles. Could you please do that?

[New Daddy]

I think the answer is simple. The tutorial on this site is correct. You should also read http://toothwalker.org/optics/dof.html, which explains this with concrete examples and also explains the difference between DOF and background blur.

That link to toothwalker.org was really helpful. Thanks!

The differentiation between DOF and OOF blur, which is what I'm after, became much clearer -- no pun intended -- after reading toothwalker.org. And based on my new knowledge, I also found this website. I think this website will be more helpful than ODF calculator to those looking for nicely blurred background:

http://howmuchblur.com/#compare-1x-5...m-wide-subject

[ajohnw]

I'm glad that some one posted the toothwalker link. The 28mm and 100mm shots show the true effect very well. Just a pity that they didn't maintain the same angle of view as that might just have shown another effect.

The shots clearly show the for shortening of the longer lens and a knock on effect shown by the pavement that the blur between objects builds up more quickly in actual photographs.

It's a pity that there isn't a 28mm shot with the same angle of view. This is where portrait lenses come in while they can't offer the same angle of view they can offer the same "perspective". Values vary on 35mm between 70 and 100mm. Move on to a rich expensive portrait photographer that maybe does commercial add work as well and they are likely to have 3 lenses including a longer one than the usual range. That for getting more for shortening - big noses etc. The thing people miss in the this area is on a portrait or similar things, packet of cornflakes etc, the angle of view needs to be the same irrespective of the focal length of the lens. Under these conditions assuming a full frame is used there will be a lot of difference as the focal length is changed. The other complication is that when we look at things we are aware of some of the thing we are looking at but the rest isn't entirely sharp when subjects are at what might be called closer distances. That's why I put perspective in quotes it's not a simple angles etc argument. It's a fact that the portrait type focal lengths can match the view of the eye under the sort of conditions they are intended for but can't match the eye's field of view - that doesn't matter for the sort of things they are used for.

People these days just go bokeh, bokey, wokey but there is a more intelligent use of it. It can be seen in many add shots. The object being advertised will be in sharp focus. It will be clear what the rest of the view is but they will be slightly out of focus - the same sort of effect the eye has. Used correctly this automatically directs the viewers eye as well. Colin has an example of this approach in his flikr shots not the best but the circumstances aren't ideal but still real photography and it works. The best example I have seem was a model standing in front of a well trimmed hedge. Straight shot all in focus and very very ordinary. Out comes tape measure and lot of sums. Lens,distance and setting worked out including the final reproduction size, every thing set up and another shot taken. All of a sudden the model stands out. It has always amazed me that the same effects can sometimes be seen being deliberately used this way on the telly. Done on the fly.

Back to original question - put it all together and something fast in the portrait length range covers the above including bokeh. In my view that's a useful lens. For wider views I suspect the standard focal length - sensor diagonal -is the shortest that can realistically be used and again as fast as possible. However it's amazing what perspective correct can do even to the extent of defishing.

My example comes from the days when Hasselblad pro's might carry a suitcase full of lenses around with them. My friend was a cheap skate and used a 2x converter to fill the gap in his lens range. He also had a huge studio. I wonder what they do these days as fewer lenses are available - play with the distances I suspect.

John
-

[William W]

The differentiation between DOF and OOF blur, which is what I'm after. . .

Then I suggest that you should take a look at and perhaps download, Bob Atkins' "Blur Calc" program.

WW

[ajohnw]

That link to toothwalker.org was really helpful. Thanks!

The differentiation between DOF and OOF blur, which is what I'm after, became much clearer -- no pun intended -- after reading toothwalker.org. And based on my new knowledge, I also found this website. I think this website will be more helpful than ODF calculator to those looking for nicely blurred background:

http://howmuchblur.com/#compare-1x-5...m-wide-subject

This should show that there is a little more to think about. Any smaller than this and the effect is spoilt. I could have got round that by snipping off 2 of the leaves but would have disturbed the rain drops. It's most effective full sized really.



I just bought a 45mm F1.8 lens, more dof than 90 F1.8 on 35mm or so they say. I'm glad it has. This shows the difficulties in using it for full frames but it starts getting usable as the shots are reduced in size. Full frame it's 27 pixels blur 10cm away from the object. It also shows something along the lines of what I did with the flower. APS 85mm at F8. This can be useful at times.



John
-

[RobertsMx]

I'm glad that some one posted the toothwalker link. The 28mm and 100mm shots show the true effect very well. Just a pity that they didn't maintain the same angle of view as that might just have shown another effect.

The shots clearly show the for shortening of the longer lens and a knock on effect shown by the pavement that the blur between objects builds up more quickly in actual photographs.

It's a pity that there isn't a 28mm shot with the same angle of view.
-

Not sure what you mean by same AOV John... same from 28mm as from a 100mm lens? If so, how are you going to achieve that from two different focal lengths?

[ajohnw]

Not sure what you mean by same AOV John... same from 28mm as from a 100mm lens? If so, how are you going to achieve that from two different focal lengths?

By altering the distance between camera and subject until the angle of views match. I should have said another shot showing that against the 100mm really.

Another way of looking at it is taking head and should portraits with the 2 lenses with the viewfinders showing exactly the same view apart from lens effects.

If I understand the web calculator correctly the plots are all for the same angle of view. ie a head and shoulder sized subject is is filling the sensor in all cases. Looking at the 45mm 1.8 it makes me think mug, what did I buy it for as I am well aware of the problems with lenses like that. I found the long lens curve interesting as I have always thought they work out like that and 85mm x 1.6 isn't so long that forshortening is likely to be very evident at shorter distances. 135mm lenses used to be very popular, bit limited in tele terms but an easy hand hold.

John
-

[revi]

By altering the distance between camera and subject until the angle of views match. I should have said another shot showing that against the 100mm really.

Another way of looking at it is taking head and should portraits with the 2 lenses with the viewfinders showing exactly the same view apart from lens effects.

If I understand the web calculator correctly the plots are all for the same angle of view. ie a head and shoulder sized subject is is filling the sensor in all cases. (...)

Sorry, but that's not what I've been taught as Angle of View. The angle of view is only dependent on the focal length of the objective used, and not on the distance between subject and camera (not strictly true:
in general, when closing in on the subject, the angle of view is getting slightly smaller) (see e.g. wikipedia or free dictionary)

But yes, the web calculator gives results for the same sized image of the main subject. That means that with a 150 mm objective you'll be at 10× the distance to your subject compared with a 15 mm.

[ajohnw]

Sorry, but that's not what I've been taught as Angle of View. The angle of view is only dependent on the focal length of the objective used, and not on the distance between subject and camera (not strictly true:
in general, when closing in on the subject, the angle of view is getting slightly smaller) (see e.g. wikipedia or free dictionary)

But yes, the web calculator gives results for the same sized image of the main subject. That means that with a 150 mm objective you'll be at 10× the distance to your subject compared with a 15 mm.

Yes silly me but given the rest I suspect you know what I mean. I should have said same view in the frame which as you correctly point out will alter the angle of view and the distance from the subject as the lens is changed.

John
-

[ajohnw]

Still not clear really. Going back to the figurine shots in the link. 2 shots in portrate mode where the figure fills the view as much as possible with the camera always at the same angle and ideally square and centred on the figure. I'm not really on about back ground blur in this respect only the effects of the lenses focal length on the figure.

The perspective effects relate to how well the shot relates to the eyes perception of the depth - not the field of view the shot can show.

Neither relate to the topic really but I always feel that there is some confusion on this aspect of the subject and can't help commenting on it. No doubt some will disagree. As I see it magnification is changing as focal length is changing. Magnifications bring things that are further away closer which is why telephoto's show a for shortening effect. It so happens that some debatable focal length between say 70 and 100mm matches the eye. On the other hand from my own experiences some FRPS people from film days could tell if a portrait had been taken with a 135mm lens especially mine as I never bought one and went for 200mm. No comments at all when I bought a 100mm later.

These people still worry about it too
http://www.hasselblad.co.uk/products...em-lenses.aspx

The choice of the 1.7x converter is interesting.

Where is does relate to the topic is use for portrait/object type and landscape work. From DOF available at 45mm F1.8 on micro 4/3 for instance it can be seen that the very fast end of the F ratio is unlikely to be useful unless the shot is reduce by a fairly large amount. In the real world perfect fuzz sometimes needs the object and the scenery to be moved.

And people only worry about pixel count and densities.

John
-

[RobertsMx]

Yes silly me but given the rest I suspect you know what I mean. I should have said same view in the frame which as you correctly point out will alter the angle of view and the distance from the subject as the lens is changed.
John
-

John, field of view is dependent on the focal length. You cannot change that fact. A 100mm lens will provide a narrower field of view than a 28mm lens. What you may be alluding to is to use a larger magnification of the subject with 28mm by moving very close to it but keeping the magnification with 100mm lens smaller. You'd still not be changing the field of view.