A technical dof question...

[anuragagnihotri]

<p>Hi guys,<br>

Here's a question, hope someone who has knowledge of this will help:<br>

The question is about DOF comparison of the two lenses: the new nikon 70-200 F4 and the old 180 AF ED.<br>

One would think that the 180 will have a shallower DOF because its faster, but a stop slower 70-200 has a better minimum focussing distance...<br>

New Nikon 70-200 F4 has a MFD of 3.2 ft...<br>

The nikkor 180 F2.8 has a MFD of 5 ft...<br>

Which one will have smoother/less DOF at the long end (200mm), at their respective MFDs, lets say, while shooting a portrait (head/head and shoulder)?<br>

What if we are shooting not at MFD but a full length plus some background kind of portrait?<br>

All wide open. <br>

regards,<br>

anurag</p>

[lornesunley]

<p>This handy tool will help you out with these calculations</p>

<p>http://www.dofmaster.com/dofjs.html</p>

<p>with a full frame camera (D700)<br /> the DOF of the 70-200 F4 lens set to 200/F4 at 3.2 feet will be 0.01 feet<br /> the DOF of the 180 F2.8 lens set to 180/F2.8 at 5 feet will be 0.04 feet</p>

<p>Given that the depth of field is dependent on the subject distance, focal length and aperture, it should not be surprising that a shorter focal length has greater depth of field in those conditions.</p>

<p>If you shoot at the same distance with the same focal length and the lenses set to their respective maximum apertures, the F2.8 lens will have the shallower depth of field.</p>

[hans_janssen]

<p>DOF is one thing, and can be calculated, but bokeh is another thing and can't be calculated.</p>

<p>What do you prefer? A very shallow DoF with an (maybe) awful bokeh or less shallow and a creamy bokeh?</p>

[anuragagnihotri]

<p>Sorry i will re-phrase my question a bit. I realize that you don't shoot a head and shoulder portrait from MFD, that is 3 feet and 5 feet. So how far one typically stands with 200+- focal length for head/shoulder and full body portrait?</p>

[anuragagnihotri]

<p>Lorne, <br>

I looked at DOF master. It was surprising to learn that there is hardly any DOF difference between a fast and a slower lens...in fact, the opposite because the slower zoom has a lesser MFD. </p>

 

[kari_oinonen]

<p>And the related question is: what is the focal length of the 70-200 F4 at minimum focusing distance when set to f=200mm.<br>

The 70-200 F2.8 decreases to about 140mm at minimum focusing distance at 200mm setting. So you set 200 and get actually about 140 mm. Photo .net has some discussion on this wrt to 2.8 zoom type II.<br>

The same occurs with the fixed 180 mm lens too but to a lesser digree. It is IF design.</p>

<p>You can get some idea by also comparing the maximum reproduction ratios at mfd. The zoom has 0.274x at 3.2 feet and the 180 af has 0.15x at 5 feet.<br>

Hopefully somebody can do a quick comparison and provide tested results.</p>

[jose_angel]

<p>Well, it cannot be said this way. In fact, the difference is mainly due to the aperture.</p>

<p>It doesn`t matter the focal lenght, or the focus distance. The bigger the subject on the screen the shallower DoF. If you keep the same size, the faster the aperture, the shallower DoF.</p>

[Two23]

<p>Jose is right. Bokeh is dependent mainly on the physical shape of the iris blades. I'd look to see which lens had the most blades, which should make the hole rounder. Really though, you need to shoot each lens yourself and then do a blind "taste" test. I'm betting on the newer lens, but that's a guess. </p>

<p>Kent in SD</p>

[rodeo_joe1]

<p>Any depth-of-field calculation has to be made using the <em>real</em> focal length of the lens. Since both the 180mm AF ED and the 70-200 f4 zoom use internal focusing, their focal length will change from the nominal value at infinity to some shorter focal length at minimum focusing distance. This can be quite a significant change, and needs to be known before the D-o-F can be properly calculated. Unfortunately, Nikon and most other makers keep very quiet about those figures.</p>

<p>It's almost certain that the 180mm f/2.8 lens will show a shallower DoF wide open, but those real-life focal lengths need to be known before an accurate DoF comparison can be made. Or someone needs to shoot the same subject at the same magnification with both lenses, and simply compare the results.</p>

<p>The other thing that needs to be known is the format the lenses will be used on. A DX body will show a shallower DoF than a full-frame camera for the same focused distance, focal length and aperture. However the full frame camera will need more magnification (i.e. closer focusing) to fill the frame with a head and shoulders portrait. The nett result of this is that for the same framing the DX camera will show about 1.5 times the DoF.</p>

<p>BTW, number of aperture blades will be totally irrelevant when both lenses are used wide-open.</p>

[anuragagnihotri]

<p>DOF master tells me that the difference between these 2 lenses is minimal at the most when it comes to DOF (surprise)<br>

Example, at 10 feet, the DOF difference is just .02 feet when you shoot at 180/2.8 or 200/4...<br>

Assuming the DOF is almost the same, will the creamy-ness of the background will also be similar? Suppose there are trees 15 feet behind the subject...which one will render them creamier? </p>

 

[anuragagnihotri]

<p>I read somewhere that 180 retains its focal length at close quarters and there is no breathing issue here. Its MFD is not very handsome though, 5 feet is lots.</p>

[rodeo_joe1]

<blockquote>

<p>"the DOF difference is just .02 feet when you shoot at 180/2.8 or 200/4..."</p>

</blockquote>

<p>That's assuming no change of focal length with focusing. Given the information provided by Kari, it seems likely that when focused to 3m the true focal length of both lenses might be very similar at around 170mm. In which case the f/2.8 lens would have a noticably shallower DoF.<br>

Another issue is that a DoF calculator only shows the limits where definition becomes noticeably blurred, it doesn't show how blurred an object 15ft behind the subject will be. So although the DoF calculator shows a minimal difference the actual amount of blurring 15ft behind the subject may be much more obvious.</p>

[anuragagnihotri]

<blockquote>

<p>Another issue is that a DoF calculator only shows the limits where definition becomes noticeably blurred, it doesn't show how blurred an object 15ft behind the subject will be. So although the DoF calculator shows a minimal difference the actual amount of blurring 15ft behind the subject may be much more obvious.</p>

 

</blockquote>

<p>Rodeo, exactly what i was thinking...the DOF master doesn't tell if the trees 10 feet behind the subject will be equally blurred or not. It tells that yes, they will not be in focus. </p>

[joseph_wisniewski]

"I realize that you don't shoot a head and shoulder portrait from MFD, that is 3 feet and 5 feet. So how far one typically

stands with 200+- focal length for head/shoulder and full body portrait?"

 

There's something very handy called "similar triangles". Say you're shooting on an FF camera. Your sensor (or film) is

24x36mm. In a portrait, we're concerned with the long side, 36mm. You want to shoot this with a 200mm lens. So, 36mm

base, 200mm height, that's the first triangle. The height is 5.55x the base (200mm/36mm). So, whatever you shoot, the

height of the triangle (distance to the camera) will be 5.55x the base (height of the subject, plus whatever you frame with

the subject)

 

A full body is somewhere around 6 feet tall, and you don't want the head or feet touching the frame of the image, so give

it about 8 feet. That means you need to be 8x5.55 feet = 44.4 feet away from the subject. (which also means that the

changing focal lengths that some folks mentioned aren't much of a problem, because you're not focused close).

 

Waist up (I like that term better than head and torso) is trickier. You're probably printing as an 8x10, not a 3:2, so,

although you have a 24x36mm sensor, you can really only use the 8x10 "portion" of it, 24x30mm. So, the triangle has a

height 6.67x the base, not 5.55. Let's call the nicely framed waist-up 4 feet. So, you shoot from 26.7 feet.

 

Nice backgrounds, but total disconnection from your subjects.

 

Now you know why the 85mm has been the "go to" waist-up portrait lens for decades.

[lwg]

<p>You can't tell from the DOF charts or the lens specs which will have a creamier background. Two lenses that look identical in specs will render both the foreground and even the focal plane differently. Look for lots of examples taken with both lenses. Comparing images is the best way to see the differences.</p>

[joseph_wisniewski]

"Bokeh is dependent mainly on the physical shape of the iris blades."

 

Not really. Bokeh is mainly determined by the lens's spherical aberration correction. Undercorrection leads to circles of

confusion that are brighter in the centers than the rim, so the middle of the blur is emphasized, and edges are "faded out"

gently. The background acquires "smoothness". Overcorrection (common on telephoto and macro lenses because it

increases the perception of sharpness) causes the rim of the circle of confusion to be brighter than the center, which turns

any out-of focus edge into two edges, the characteristic "nervous bokeh" look, where there's lots of sharp transitions.

 

The shape and number of aperture blades has some small effect on bokeh, mostly visible in bright "point sources" of

lights. It's "low hanging fruit" for the marketing people, because changing spherical aberration involves recompiling the

whole lens, setting up new tooling for the elements, basically relaunching the whole product, while changing the number

and shape of the aperture blades is fairly inexpensive. You're just rehashing marketing material.

 

And, as RJ mentioned, wide open, the aperture blades aren't even in the optical path.

[keirst]

<p>Shoot some tests before you do the real portraits and see which lens you prefer for its rendering. You can’t really predict the differences you will see based on DOF calculations. <br /><br />The 180mm/2.8 AF can be used with long extension tubes like the for closer focusing than the built in minimum focus distance, and it works pretty well this way optically, but you will loose AF and electronic communication between the lens and AF-SLRs and DSLRs.</p>

[joseph_wisniewski]

The tubes don't have to be that long. A 180mm focuses down to 1.5m by dropping its focal length to 160mm. So, a 20mm

tube is the "continuation" of it's focusing range, a 36mm is close to the continuation of the continuation, and a 12mm is

the "portrait booster".

 

12mm tube - 2.89m to 0.96m (9.5 ft to 3.1 ft)

 

20mm tube - 1.80m to 0.80m (5.9 ft to 2.6 ft)

 

36mm tube - 1.08m to 0.62m (3.5 ft to 2.0 ft)

 

Kenko tubes have electrical contacts and a mechanical extension of the AF "screwdriver" and preserve AF, all metering

modes, EXIF logging, and camera body aperture control.

[bebu_lamar]

<p>Although the 180mm f/2.8 is of IF design, its magnification ratio of 0.15x and minimum focusing distance of 1.5m suggest that is focal length isn't shorten at close focusing distance. </p>

[jose_angel]

<p>You`re right; by applying "standard" formulas, to reach that magnification the focal lenght should be even longer!</p>

<p>Optical design is simply too much complex to me... looks like everything can be achieved in one way, but also in the inverse (I`m still frightened with the front concave element of my Leica wide angle lens... :)</p>

[joseph_wisniewski]

It does?

 

If you take the formula for lens to image distance D=A(1+1/M), you get A = 195mm, a focusing extension of 15mm for a

180mm lens.

 

But if the focal length doesn't shorten, a 180mm has to extend 27mm to reach a magnification of 0.15x.

 

So, a magnification of 0.15x at a focusing distance of 1.5m doesn't "suggest" that the focal length doesn't shorten, it

states, quite clearly, that it does shorten.

[jose_angel]

<p>From my calculator, the extension needed to focus at 1.5 meters is 25mm... total bellows is 205mm...</p>

[jose_angel]

<p>Ok, I have just got it. I have just checked that "my" 205mm of extension still need another few millimeters more to reach that 0.15X. Too dense tonight... :P<br /> Thanks Joseph.</p>

[rodeo_joe1]

<p>I think we're getting away from the OP's original question guys.</p>

<p>Anurag, I re-wrote my Excel depth-of-field spreadsheet to work out the blur circle (circle of confusion) from the focused distance, rather than working out the DoF. As I suspected there's a very noticeable difference in the blur circle between a 200mm f/4 lens to a 180mm f/2.8 lens.</p>

<p>Conditions chosen were: Focused distance 3m (10ft); Background 3m behind subject. It's assumed that there's a pin source at the background position to render a perfectly circular OOF blur.<br>

The 180mm f/2.8 lens gave a massive blur circle diameter of 2.05mm at the image plane, while the 200mm f/4 lens gave a blur circle of 1.79mm. This is a big enough difference to be easily noticeable, I feel. The straightforward DoF calculator shows almost identical DoF for both lenses under the above focusing conditions.</p>

<p>I'm just going to double-check the maths, but I'm pretty sure I've got it right. This could be quite a useful calculator, so it's not time wasted.</p>

 

[rodeo_joe1]

<p>FWIW, the formula for calculating focal length from magnification and focused distance is F=d/(2+m+1/m). Remembering that focused distance is measured from image-plane to subject, and not from the forward lens node to the subject.</p>

<p>Using m=0.15 and d=1500mm in the above formula; the focal length works out at 170.13mm.</p>