Why are classic focal lengths important?

[Sejanus.Aelianus]

It is like politics, with as many political parties as there are photographers.

...and like politicians; the dafter the idea, the harder they'll push it.

 

[Replytoken]

I could be wrong but from what I think I see on forums is that most enthusiasts are in the older age bracket (me included) and like to relate sizes back to what they used in the old film days. Same as I often still think of some sizes in inches instead of mm.

Just guessing but do the younger people worry about the older size relativities of lenses, doubt it. However, I still relate to the older sizes a bit when buying lenses because I remember what I liked back in my younger days. Perhaps I do not have any more memory slots left.

I suspect that if smartphone makers change from the 30mm eq. that younger folks might have something to say as it is what they are most accustomed to. They may enjoy a new focal length, but I am sure that some will bemoan the loss of the "classic" smartphone lens. Just give it some time. What goes around comes around.

--Ken

 

[ADemuth]

Ok, this thread has kind of derailed and we have some number crunchers here too and we're already on the subject of FOV, so I'll ask my question regarding focal lengths.

Depth of field between different focal lengths.

As I understand it, DOF (at a given aperture and focal length) increases with distance from subject, but decreases inversely to focal length at a given distance at that same aperture. Right?

So: If I take a picture at with a 50mm lens with a set FOV and switch to a 100mm lens (for easy math) and I back up (to twice the distance?) to make the same scene fit on the image sensor, is my DOF the same between the two, assuming the aperture is the same??

 

[SojiOkita]

Put a third way, if I take a standard portrait with an 80mm lens and get the typical head-and-shoulders shot, if I switch to a 35mm lens AND DON'T MOVE the perspective is identical, I'm just able to see more of the body now.

That's the whole principle of micro 4/3
The sensor is exactly like a center crop of a 24x36 sensor.
And a 25 mm in micro 4/3 gives the same perspective as a 50 mm.

We often read about "long focal length compression", but that's just an effect of distance, not focal lengths.

 

[AussiePhil]

Ok, this thread has kind of derailed and we have some number crunchers here too and we're already on the subject of FOV, so I'll ask my question regarding focal lengths.

Depth of field between different focal lengths.

As I understand it, DOF (at a given aperture and focal length) increases with distance from subject, but decreases inversely to focal length at a given distance at that same aperture. Right?

So: If I take a picture at with a 50mm lens with a set FOV and switch to a 100mm lens (for easy math) and I back up (to twice the distance?) to make the same scene fit on the image sensor, is my DOF the same between the two, assuming the aperture is the same??

Short answer yes.
Here is the maths courtesy of Depth of Field, Angle & Field of View, & Equivalent Lens Calculator - Points in Focus Photography{"c":[{"f":8,"av":"4","fl":100,"d":6096}],"m":0}
The key bits to look at is the Field numbers to see they match and the total DOF

Also note what is different as well

Subscribe to see EXIF info for this image (if available)

Subscribe to see EXIF info for this image (if available)

 

[AussiePhil]

We often read about "long focal length compression", but that's just an effect of distance, not focal lengths.

ummmm..... no it's about Focal Lengths as clearly shown here ..... How to use Focal Length and Background Compression to Enhance Your Photos

 

[SojiOkita]

So: If I take a picture at with a 50mm lens with a set FOV and switch to a 100mm lens (for easy math) and I back up (to twice the distance?) to make the same scene fit on the image sensor, is my DOF the same between the two, assuming the aperture is the same??

You can look at this type of thing in a dof simulator like this:
DOF simulator - Camera depth of field calculator with visual background blur and bokeh simulation.
The depth of field will be approximately the same, but the rendering of the background will be totally different.
That's why this simulator is good, it allows you to visualize the effect on the background.
The difference is linked to the compression we're talking about.

 

[SojiOkita]

ummmm..... no it's about Focal Lengths as clearly shown here ..... How to use Focal Length and Background Compression to Enhance Your Photos

On all the examples he gives, his distance with the subject is different, as explained in the comments.

Good question, David and Michelle. Yes I did have to move much farther back from my dad in order to keep him the same size in the frame. At 270mm I was...oh, I'm just guessing here, but maybe 10 or 20 meters back whereas at 18mm I was only about a meter away. None of those photos have been cropped, so what you're seeing is exactly what I got in camera.

So he clearly shows the effect of the distance on background compression, not focal length.

A 150 mm in micro 4/3 has exactly the same background compression as a 300 mm in 24x36.
And exactly the same background compression than a center crop of a 150 mm in 24x36.

If perspective & background compression depended on focal lengths, we wouldn't have any focal length equivalence between formats.

 

[SojiOkita]

A good article on the subject:
Lens Compression Doesn't Exist

Most photographers believe that the lens is "compressing" the scene because they will take a test shot with a wide lens, and then back up to capture a similar composition with a telephoto lens. It's easy to see that the telephoto shot is far more "compressed" and it's also easy to completely overlook the fact that the camera has been moved and the field of view has changed.

I think it's an important point to understand when composing a picture, because if you want a different subject/background perspective, the first thing to do is to move, not change focal length.

 

[AussiePhil]

A good article on the subject:
Lens Compression Doesn't Exist

Ok always happy to be educated and strangely i had read that article before, not sure why it didn't stick.

My issue with the F-stoppers article is that even if we accept it as truth. (and i think i will) The real world usage of, stick a longer FL on the camera, frame the same image then you will get background compression still in pure lay terms comes back to a longer FL will compress the background for the same framing of the subject.

 

[SojiOkita]

My issue with the F-stoppers article is that even if we accept it as truth. (and i think i will) The real world usage of, stick a longer FL on the camera, frame the same image then you will get background compression still in pure lay terms comes back to a longer FL will compress the background for the same framing of the subject.

Yes and I totally agree with that.
I often pick a focal length because of the perspective I will get: the focal length chosen will influence the position in relation the subject, and therefore the perspective & the background compression.

However, it's interesting to know that the real driver is the position.
If you use a zoom and don't like the perspective, zooming won't change anything.
Moving will.
And moving + zooming will change the perspective and keep the framing.
(as in the famous "dolly zoom" in Jaws or Vertigo)

 

[bjurasz]

However, it's interesting to know that the real driver is the position.
If you use a zoom and don't like the perspective, zooming won't change anything.
Moving will.

BAM! There it is, short, simple and accurate.

 

[brettmaxwell]

However, it's interesting to know that the real driver is the position.
If you use a zoom and don't like the perspective, zooming won't change anything.
Moving will.
And moving + zooming will change the perspective and keep the framing.

YES.

Distance is what dictates perspective and compression, and it goes hand-in-hand with field-of-view to get the desired framing. Field-of-view itself is a combination of focal length and sensor size.

In a very practical application, you could do street/travel photography with a 12-100mm (m4/3) lens and want a full-body framing of a human subject. If you want more of the scene, get close and go 12mm. If you want to exclude much of the scene, keep your distance and rack to 100mm.

 

[ADemuth]

Short answer yes.

I had suspected this, but every once in a while I'd hear someone say something like "Yeah, you sacrifice DOF with [insert crop sensor format here], but you can make up for that with a longer lens" and it would muck up my whole head. Thank you so much. I never thought to run it through a DOF calculator (mainly because I think those sorts of things are for the professionals - amateurs, like myself, need not apply)

It's good to know my understanding of optical principals wasn't horribly misguided.

The depth of field will be approximately the same, but the rendering of the background will be totally different.

When you say rendering, you mean the amount of background included, right? I totally understand this "compression" idea - if I want to illustrate context, I usually go wide and get in close, if I want focus on the subject, I generally want a larger FL and then back off. The OOF area should theoretically be identical in terms of out of focus-ness, just cropped (I know "cropped" isn't entirely accurate) The out of focus-ness is my question here. I can't make up for the increased DOF of a smaller sensor with longer glass and shooting further away. That won't decrease the DOF. A larger aperture is the only way to achieve this.

 

[SojiOkita]

When you say rendering, you mean the amount of background included, right? I totally understand this "compression" idea - if I want to illustrate context, I usually go wide and get in close, if I want focus on the subject, I generally want a larger FL and then back off. The OOF area should theoretically be identical in terms of out of focus-ness, just cropped (I know "cropped" isn't entirely accurate)

Yes.

The out of focus-ness is my question here. I can't make up for the increased DOF of a smaller sensor with longer glass and shooting further away. That won't decrease the DOF. A larger aperture is the only way to achieve this.

Yes again. The only way to "simulate" what a larger sensor can achieve is larger aperture.

 

[AussiePhil]

the reality isn't all doom an gloom as it's more than possible to get sufficiently thin doF for most things most of the time unless you really want tip of nose only razor thin stuff that is not used/needed by most
Generally the increased DOF is a bonus rather than a hinderance
Doesn't stop me lusting after a F1.8 200mm prime to get as close as possible experience to the legendary canon 400 F2.8 for framing and dof

 

[ADemuth]

Oh, I'm not complaining. Most of the time I actually appreciate the increased DOF I can still have when shooting wide open - too thin and I find I start to miss shots, especially with AF. Also, lots of bokeh doesn't instantly make a photo good.

 

[wimg]

A short note on DoF calculators: they are all based on the standard CoC, which is based on its definition dating back to 1930, i.e., film era and its limitations.

In this definition, CoC is defined as the largest allowable value of a point size so that an 8 X 10 inch print viewed at 10 inch distance shows a smallest distinguishable feature of 0.01 inch.

In practice, this corresponds to a CoC of 0.00125 inches on 35 mm film or FF, as an 8X magnification is required to print it this size, or 0.030 mm, which becomes 0.015 mm for MFT.

In reality human acuity at this distance with 20/20 vision is probably about 3X better, so 0.010 mm for FF, and 0.005 mm for MFT.

This drastically alters perceived DoF, especially for larger prints, where one gets to the limit or beyond of what this definition, and real acuity, says, something I have been noticing over quite a few years already. When using the standard CoC prints never looked really sharp to me at the limits for the standard CoC, and with digital, because everything appears to be crisper, it appears even worse to me.

All in all, for me, DoF is therefore 3X less, roughly, than DoF calculators indicate, although this may vary depending on your own perception, but I reckon a minimum safeguard makes for about 2X less.

This also means that you need to forget about hyperfocal distance (and IMO it never really worked properly anyway). For best results, IMO, you need to determine what the object, subject or area of interest in your image is, and make sure that that is appropriately sharp. IOW, focus on those, rather than on an average distance for so-called optimum DoF, which it never is.

Kind regards, Wim

 

[lookmore]

My only gripe is: Why don't we have a 22mm lens and maybe a 35-37mm as well?

I have used the 35mm Zuiko Digital on E-520 and with an adapter on µ43 but I find it an odd focal length ;-)
Good for what it is though: Macro shots. Or portraits maybe, but it isn't very fast.

 

[SVQuant]

I have used the 35mm Zuiko Digital on E-520 and with an adapter on µ43 but I find it an odd focal length ;-)
Good for what it is though: Macro shots. Or portraits maybe, but it isn't very fast.

That comment was a little tongue in cheek , but I owned that lens as well (just sold it a week ago). It is was good lens for macro, though as you said slow. I also used a 50mm lens on my Nikon APS-C DSLRs which is 75mm equivalent and it made a good portrait lens.

In a sense, 70mm is a "classic" focal length, but in the zoom world. With all the 24-70/70-200 FF zooms floating around and the P12-35/35-100 lenses on m43, I wouldn't be surprised if it is actually a very common focal length in terms of actual images taken with zooms.