Is Moose Peterson wrong about teleconverters

<p>Moose has a ton of experience and access to the tech pros at Nikon so I would give a lot of credit to what he says but I don't seem to believe what he says about teleconverters reducing the depth of field. He says, "You should be aware that when using teleconverters, you lose either one or two stops of light when you use a 1.4x or 2x respectively. This is light that is just gone, sucked up by the optics of the teleconverters. At the same time, while the teleconverter increases the focal length of the lens, it does not change the size of the hole at the rear of the lens. This combo of longer lens with the same exit hole diameter means you have a slower lens but with less depth of field." He wants us to think that a 400mm f/2.8 with a 2x has less DOF than a 800mm f/5.6 (because the rear aperture hasn't changed size). I don't think the DOF would change but then I was thinking that it is possible to make things blurrier (such as use of then Nikon Defocus Control lenses.) Is Moose just wrong or am I thinking about this too much?</p>

<p>Frankly, I think you're thinking about it too much.<br>

The actual DoF depends on more than just the lens aperture. In case he meant you do not actually move your tripod back by a lot to compensate for the increased focal length, he is right. It would not be the same framing, though. But without a link to his point on TCs, it's hard to know the context in your quote fits. Maybe framing got mentioned, maybe not.<br>

Darn, now I thought too much about it too.</p>

<p>When you mount a 2x teleconverter onto a 400mm/f2.8 lens, you will have a 800mm/f5.6 lens, just like another other "native" 800mm/f5.6 lens. A "native" 800mm/f5.6 could well be sharper than the 400 with a TC and probably much longer physically, but there should be no difference between the two in terms of depth of field and many other characteristics.</p>

<p>Wouter, FYI if you google the first sentence of the quote, you arrive here:</p>

<p>http://www.vividlight.com/articles/704.htm</p>

<p>At any comparable aperture, a longer effective focal length is going to have shallower DOF.</p>

<p>I've been confused by Moose's statements about teleconverters before - or rather, I think he's confused, or not explaining very well. <a href="http://www.moosepeterson.com/blog/2010/02/25/the-teleconverters/">This video</a> may help, but unfortunately I can't view it at work. It may reveal all.<br />

<br />

As Shun says, a 400 f/2.8 with a 2x teleconverter should behave exactly like an 800 f/5.6, give or take image quality. You can think of the depth of field as happening entirely at the front of the camera: there's a cone of confusion that extends from the entrance aperture to the subject, and the angle of that cone is the same if you have a 400mm f/2.8 at the same distance from the subject as an 800 f/5.6, assuming both are shot wide open (because they both have the same entrance aperture of about 143mm - the actual apertures quoted probably aren't too exact). If you put a 2x teleconverter on the back - or take a crop out of the middle half of the image (a DX sensor is like a 1.5x teleconverter with ISO adjusted for the crop factor) - you'll get the same image from the 400mm as from the 800mm.<br />

<br />

If you take a photo from the same location of the same subject with a teleconverter, you get less depth of field. You're magnifying part of the image, so things that get out of focus, get out of focus faster. You get the same effect from cropping. However, I find it confusing to discuss this, because once you've got a completely different angle of view, you've taken a different image. For reference, if you move the camera so that your subject is the same size in the frame despite the different field of view, the depth of field at the subject is almost entirely determined by the f-stop - the focal length and distance cancel out.<br />

<br />

But do think about it. It's always good to know what's going on! (I recommend drawing some similar triangles, with a lens in the middle, a stick man on one side, and an image plane on the other; hours of fun...)<br />

<br />

By the way, the defocus control lenses don't make things blurrier (other than giving some uncorrected soft focus spherical aberration if you turn the DC ring past the actual aperture): they're for making the bokeh look smooth, but they don't change the <i>amount</i>. Now, tilt lenses can make bits of the image go very out of focus... and don't get me started on coded aperture EDOF. :-)</p>

<p>Thomas, I don't have anything to contribute that hasn't already been said, I just find it fascinating that you are referencing an article published in 2001 based on a camera/lens combination from the last century. The D1 had an APS-C sensor boasting less than 3 megs. <br>

I haven't followed Moose's writings for about 8 years. Does he still find it necessary to carry a 600mm lens/converter combination in the field given he now has the crop ability of larger sensors?</p>

<p>Gup: There was a comment on one of Moose's teleconverter pages about using a DSLR for cropping instead, and how this avoids the light loss. (I'd actually argue with that a bit, since noise increases as you magnify, but that's a different can of worms.) Besides, Nikon now make an AF-S 800mm f/5.6! My biggest issue with using a lens that long (in my case, trying a TC-16A on a 500mm f/4 AI-P - which works, but don't expect sharp corners) is that if you lose the subject in the frame, it's hard to find it again. But I should spend more time with my telescope getting used to star hopping!</p>

<p>He posts some pretty nice shots from those crappy old sensors.</p>

<blockquote>

<p>"At the same time, while the teleconverter increases the focal length of the lens, it does not change the size of the hole at the rear of the lens. This combo of longer lens with the same exit hole diameter means you have a slower lens but with less depth of field."</p>

</blockquote>

<p>That's either very badly worded, or the guy simply hasn't got a clue about how teleconverters work. Has he actually looked at the rear of a lens with teleconverter fitted? The 'hole' definitely doesn't stay the same size. The exit pupil, to give it its correct name, is divided in diameter by the magnification factor of the converter. That's the very reason for the loss of light and change in effective aperture number. (Well, actually the exit pupil is moved further from the image plane, but the nett effect is the same.)</p>

<p>What doesn't happen is that the converter 'sucks up' light. It genuinely increases the focal length and genuinely increases the relative aperture number of a lens. The shallower depth of field happens because DoF is proportional to aperture number, but inversely proportional to the <em>square</em> of the focal length. Simple maths.</p>

 

<p>Gup, Moose no longer carries the 600mm/converter combination in the field. He now has the new Nikon 800mm lens.</p>

<p>Barry: Of course he does. Oh to be a well-known photographer. (And, you know, have any photographic talent to earn the respect with! :-) )</p>

Good thread but did any body actually try the setup to get empirical data? I do not have the right equipment to do this but if a large check arrived to fund it, I would post the results {:o)

<blockquote>

<p>I do not have the right equipment to do this but if a large check arrived to fund it, I would post the results {:o)</p>

</blockquote>

<p>I'd be happy to carry out the experiment if someone with a deep pocket is willing to fund me for a 400mm/f2.8 and a 800mm/f5.6, both AF-S VR. I already have both optical versions of Nikon's TC-20e teleconverters to get a head start. :-)</p>

<p>Read any book authored by Moose and you'll see all kinds of typographical mistakes and often times his comments about gear is wrong. He comments about gear he's never used.</p>

<p>Thanks for the clarifications. I think Rodeo Joe might be on to the correction in communication when he says, " The shallower depth of field happens because DoF is proportional to aperture number, but inversely proportional to the <em>square</em> of the focal length. Simple maths." I wasn't aware of this so perhaps this is exactly what Moose was trying to say. I.E a 600mm 5.6 has less DOF than a 300mm f2.8. I admit I didn't know this, I assumed they would be the same. Thanks. I looked at some of the formulas but I need to look at them some more, the ones on wikipedia get much more involved than needed for this conversation and they don't clearly label the variables. </p>

<p>Not sure why you're looking on Pickyweedier for an answer Thomas. One of the many online DoF calculators - <a href="http://www.dofmaster.com/dofjs.html">like this one </a>- will give you the answer in a few mouse clicks.</p>

<p>Keep the same camera model and subject distance - say 40ft. Then just select a 300mm lens at f/2.8 and compare the DoF with that of a 600mm lens at f/5.6. The result shows you that the 300mm lens gives almost exactly twice the DoF as the 600mm lens.</p>

<p>The DoF isn't <em>exactly</em> double, because the focal length of the longer lens is a more significant fraction of the subject distance.</p>

<p>RJ: Because DoF calculators like that one give you a limited precision result based on a large number of hidden assumptions and tell you nothing about the actual relationships that you're looking at. I'd prefer to be told how the different factors that can affect depth of field are related than be given a load of contextless numbers and try to work out how they're related from them. While I believe the one you've linked is okay, many DoF calculators are also <i>wrong</i>, having messed up the basic formulae.<br />

<br />

With the subject at the same distance, the depth of field cone in the scene is the same shape for a 600 f/5.6 and 300 f/2.8 (both used wide open), because it's only the absolute aperture not the focal length that matters. The 600mm lens magnifies the 300mm lens image by a factor of two, this giving a larger apparent depth of field. You're also taking an entirely different image. It's possible that this is what Moose meant (though it's a heck of a misleading way to say it), but it's also possible he was just confused.<br />

<br />

Try to match images by changing sensor format, cropping, or moving the camera and the depth of field also changes, but differently. Anyone writing about this typically picks the configuration that they were thinking of at the time, and reports only that behaviour as if it applies in all situations. It's like the "airplane on a treadmill" argument. Understand what's actually going on and you won't get confused by these discussions. Put in the numbers for just one case into a DoF calculator and you'll probably just get confused. (But the Wikipedia page should be better labelled.)</p>

<p>I was going to say what Rodeo Joe said. A TC does not suck up light, it simply magnifies the image, so the sensor receives less light. Increase the magnification, and you reduce DOF. Mouse Peterson's explanations are poor.</p>