Two primes on dual format cameras (24/1.4 and 85/1.4 on D700 and D300)

<p>I came up with the following idea: mounting <strong>24/1.4 and 85/1.4</strong> on dual format cameras, let's say <strong>on d700 and d300</strong>.</p>

<p>In terms of field of view and depth of field, this gives effectively 24/1.4, 35/2.2, 85/1.4 and 135/2.2. I would consider this set for portraits on location, weddings, baby shoots and travel.</p>

<p>If anyone uses that lens set with dual format cameras, I'd love to hear their opinion because I have some concerns.</p>

<ol>

<li>First, while 135/2.2 is still impressive for portraits and sports, the exotic 24/1.4 becomes a boring 35/2.2 on crop camera.<br /><br /><br /></li>

<li>Second, I'm concerned if the added noise will degrade your expensive lens collection.<br /><br /><br /></li>

<li>Third, I wonder if you don't get tangled up with the lens changes and their different properties on dual format cameras. <br /><br />At any point in time, you will either have 35/2.2 and 85/1.4 or 24/1.4 and 135/2.2 (in terms of FoV and DoF). <br />Is this feasible in practice? </li>

</ol>

<p>Why are you indicating a different f-stop for the lenses on a crop camera?</p>

<blockquote>

<p>Why are you indicating a different f-stop for the lenses on a crop camera?</p>

</blockquote>

<p>Because a smaller sensor produces more depth of field. It's called <a href="http://www.josephjamesphotography.com/equivalence/">equivalence</a>. You can calculate the difference <a href="http://www.dofmaster.com">here</a>.<br>

In other terms, 24/1.4 on d300 will not produce the same images as 35/1.4 on d700 given the same distance and FoV. <br />24/1.4 on d300 will look more like 35/2.2 on d700, excluding noise and detail.</p>

<p>What?</p>

<p>I don't see why it should be confusing - my Nikkors are all full frame, but I've switched between a crop-sensor Canon and a film camera without any difficulty; you could roughly achieve this with one camera if you got a (third party) 1.4x teleconverter, after all - or force the camera into DX mode (or just crop the result yourself). There are more differences between the optical characteristics of the native lenses at these focal lengths than just the aperture, for what it's worth. It may be a sensible way of picking your lens selection if you already have both bodies, but I'm not sure I'd go that way from scratch. I'm not sure I understand your point about noise, by the way.</p>

<blockquote>

<p>I'm not sure I understand your point about noise, by the way.</p>

 

</blockquote>

<p>Full frame cameras generally produce less noise and more detail. I'm concerned about losing detail and adding noise when screwing prime lenses on a crop camera.</p>

<p>Oh, right ... maybe I should read more carefully.</p>

<p>I've never thought of doing something like that, I have a D700 and a D200 so I guess it could be done, but I don't think it would make much difference. I'd be more likely to use the D700 and just stop down and/or move closer to the subject if I was looking for changing the DOF or wanting different framing. You shouldn't get any degradation from noise if you don't try to use a higher ISO than the camera will produce good images with when properly exposed.</p>

<blockquote>

<p>I'd be more likely to use the D700 and just stop down and/or move closer to the subject if I was looking for changing the DOF or wanting different framing.</p>

</blockquote>

<p>The thing is: I do not want to have more depth of field. I pick f/1.4 lenses to achieve shallow DoF. However, on crop cameras, part of that shallow DoF is lost.</p>

<p>Moving in closer is not always an option but is a good idea nonetheless :)</p>

<blockquote>

<p> boring 35/2.2 on crop camera.</p>

</blockquote>

<p>Lenses aren't boring. Photographs can be boring, photographers can be boring, but a lens is just a tool. <a href="http://www.citysnaps.net/"> This guy </a>shoots almost exclusively with a 35 on a full frame camera. </p>

<blockquote>

<p>Lenses aren't boring. Photographs can be boring, photographers can be boring, but a lens is just a tool. <a rel="nofollow" href="http://www.citysnaps.net/" target="_blank"> This guy </a>shoots almost exclusively with a 35 on a full frame camera.</p>

</blockquote>

<p>Well good for him! That was such a bland, boring remark... sigh.</p>

<p>You're just confusing yourself by pretending that the aperture is "effectively" different between sensor formats. The most important effect of the aperture is that it balances against shutter speed to get you the right exposure. This doesn't change at all when you switch between FX and DX camera bodies.</p>

<p>Really, all the stuff you hear about a lens being "effectively" something different on a DX body is nonsense that causes lots of misunderstanding among people who don't actually understand how cameras and lenses really work. Ultimately, you cannot take "exactly the same picture" with cameras of different formats. You can try to adjust various parameters to get close, but you can never really get it exactly.</p>

<p>So every time you need to change a lens you actually have to change TWO lenses (unless you're leaving the body cap on a body)<br>

For me this would get frustrating but why not just try it. Everyone has different ways of working.</p>

<blockquote>Full frame cameras generally produce less noise and more detail. I'm concerned about losing detail and adding noise when screwing prime lenses on a crop camera.</blockquote>

 

<p>Ah, I wondered. Full frame cameras tend to produce less noise <i>at the same relative aperture</i> as a crop sensor camera (mostly) because, at the same pixel count, more light hits each sensor site. At f/2, the amount of light per unit area doesn't depend on focal length, but a D300's pixels are (1.5x1.5=)2.25x smaller than a D700's. The actual difference is slightly more in favour of the larger sensor, partly because a fixed amount of the sensor area has to be occupied with logic and data transfer, but within a sensor generation it's the pixel size that matters.<br />

<br />

If you consider an 85mm f/1.4 on a DX camera to equate to a 127.5mm f/2.1 lens on FX, bear in mind that a native 127mm f/2.1 lens lets in less light by a factor of 2.25, which balances out the noise issue (mostly, as I said). Hence you could say that an 85mm f/1.4 on a DX body at ISO 200 is like a 127.5mm f/2.1 lens on an FX body at ISO 500, in terms of field of view, exposure and depth of field. But that might just confuse people who prefer to think in terms of a focal length being a focal length and an ISO setting being an ISO setting. (You can get into the same arguments about teleconverters, cropping images, and film grain. And I have, and I don't want to again!)<br />

<br />

In summary: don't worry about noise, to a decent approximation the noise behaviour of a D300 with an 85mm f/1.4 lens and a D700 with a 135mm f/2 lens, both used wide open, will be similar. It is, however, easier to make a lens that behaves well optically at f/2 than at f/1.4, so you might be right to have concerns about resolution, especially away from f/8-ish.</p>

<p>Craig's response is exactly the way previous threads on this subject have gone - "<em>you're</em> confused by thinking in the way that exactly explains the relationship between depth of field, focal length and sensor size; you should think of it in my terms instead, and then you'll see that it makes absolutely no sense!" - and I recommend that you agree to differ. I suspect that some people whose expertise I greatly value might be refusing to read any of my posts after similar disagreements...<br /> <br /> Craig - I don't mean to be disrespectful or antagonistic, I'm just explaining how your post sounds from the perspective of those who believe that the geometry of the situation explains the "equivalence" perfectly. I'm happy to be told there's a flaw in the reasoning of this principle, but I've been in very long threads in which no specific problem with it has been stated, other than that it's another way of thinking about the situation. What you can't do is explain the depth of field change by stating a different focal length - which nobody seems to mind discussing - and relative aperture, and then ignore both sensitivity and shutter speed. Treat everything together, and it makes sense, with the understanding that the existing concepts of "focal length", "aperture" and "ISO" are equally valid and have their uses.<br /> <br /> And I wasn't going to get involved again... *sigh*.</p>

<p>Actually I shoot events with a pair of D700/D300 and my preffered lens setup is based on this:<br>

Nikon 24/1.4<br>

Sigma 85/1.4<br>

Nikon 180/2.8<br>

Sometimes I may bring another lens if I know I have particular needs but for most the three lenses are my workhorses.<br>

I have no problem switching lenses to get the appropriate FOV I need. The only observation I have: Sigma 85/1.4 and Nikon 18/2.8 are shinning on both cameras. Nikon 24/1.4 is really my best lens on D700... but on D300 it seems that does not show so much magic... I do not know if this is because of my lens, because of my D300 or because of my style of shooting... :-)</p>

<blockquote>

<p>You're just confusing yourself by pretending that the aperture is "effectively" different between sensor formats. The most important effect of the aperture is that it balances against shutter speed to get you the right exposure. This doesn't change at all when you switch between FX and DX camera bodies.</p>

 

</blockquote>

<p>I'm not talking about exposure but about depth of field. Major difference!</p>

 

<blockquote>

<p>So every time you need to change a lens you actually have to change TWO lenses (unless you're leaving the body cap on a body)</p>

 

</blockquote>

<p>Good point but even with one body and two lenses, you still need to grab two lenses. Where else do you put your unscrewed lens? In a lens case or on a body... doesn't change a thing. Except that if you screw it on a body, you don't need the cap!</p>

 

<blockquote>

<p>I came up with the following idea: mounting <strong>24/1.4 and 85/1.4</strong> on dual format cameras, let's say <strong>on d700 and d300</strong>.</p>

<p>In terms of field of view and depth of field, this gives effectively 24/1.4, 35/2.2, 85/1.4 and 135/2.2. I would consider this set for portraits on location, weddings, baby shoots and travel.</p>

</blockquote>

<p>Very interesting. Merely two days ago, the OP was telling people that he was using two Canon lenses, a 24-70mm/f2.8 and a 24-105mm/f4 on a 5D Mark II for weddings: <a href="../wedding-photography-forum/00YxpC">24-70/2.8 vs 24-105/4 IS at wedding</a>.</p>

<p>To me, obviously the 85mm/f1.4 is good for portraits. Otherwise, it makes little sense to use a 24mm/f1.4 and a 85mm/f1.4 for weddings, baby shoots and travel. The OP's existing Canon lenses make far more sense for wedding photography.</p>

<p>Just my personal opinion, but I have always found a 35mm and 85mm lens combination to be extremely versatile on a film or FX body. A little wide and a little long covers a lot of shooting situations.</p>

<p>It seems to me a pretty theoretical approach in photography; it sounds feasible.</p>

<p>But in the real life, (-or at least in my shooting life-), I don`t see the point of using -only- (superexpensive) primes that are forced to be used in two inseparable bodies/formats. They are actually nice, but maybe not as practical in the field.</p>

<p>I think I`d feel more confortable (portraits, wedding, baby shoots and travel) with my D700 + 24-120/4 than with four primes and two bodies.<br /> ... Or with two FX bodies, and two lenses (24-70 + 70-200) for weddings,<br /> ... or with one DX or FX body and a 85VR or 105VR for baby shoots,<br /> ... or with one DX or FX body with a prime for portraits,<br /> ... or with a DX or FX camera with a 16-85 or 24-120/4 for travel.</p>

<blockquote>

<p>I'm not talking about exposure but about depth of field. Major difference!</p>

</blockquote>

<p>The problem is that aperture relates to both. When people make overly-generalized, technically-wrong statements like "24mm f/1.4 on FX is effectively 35mm f/2.2 on DX", it confuses people who don't understand enough about the technology to understand that you mean something much more limited than you seem to be saying. If a newbie has learned that a smaller aperture lets in less light and requires a longer exposure, he may think, based on a statement like yours, that a lens set to f/1.4 on a DX camera requires an extra stop or so of exposure time to compensate for the fact that it is "effectively" f/2.2 -- which, of course, is not true.</p>

<p>A 24mm f/1.4 lens is always a 24mm f/1.4 lens. It doesn't matter what camera you put it on. A smaller sensor uses a smaller portion of the image circle, resulting in tighter framing; that's the only real difference. DOF is not, in fact, any different at all. What makes it <em>seem </em>different is one of three things:</p>

<p>1) You compensate for the narrower angle of view by using a shorter focal length, which increases DOF;<br>

2) You compensate for the narrower angle of view by moving the camera farther away from the subject, which increases DOF;<br>

3) You calculate DOF on DX using a smaller CoC value, which <em>reduces </em>DOF.</p>

<p>In this post, I think you're doing #3 (because that's what the dofmaster calculator does when you switch from FX to DX), and then you're narrowing the aperture to compensate for the smaller CoC value. This is how you end up claiming that f/1.4 is "effectively" f/2.2 on DX.</p>

<p>My next question for you is whether you understand why the dofmaster calculator is using a different CoC for a DX camera, and how the CoC value relates to assumptions about how the image will be used. It's really not as absolute as you seem to think.</p>

<p>@Craig -</p>

<p>Your last post is actually the most confusing post I've seen since I don't understand your #3. Your wording doesn't make sense to me and is very baffling.</p>

<p>Andrew and I have had very lengthy conversations, both in forums and through email, on how sensor size effects DOF and it boils down to this:</p>

<p><em>The single biggest benefit of an FX/Full Frame sensor is it's behavior when using large aperture lenses. More specifically (referring to the Nikon AF System), 50mm -135mm lenses shooting with an f-stop of f/1.4 - f/4. The larger sensor produces a depth of field roughly 1.5x smaller than an equivalent focal length (75mm on FX/50mm on DX) on a 1.5x crop sensor.</em></p>

<p><em>This is a tough concept to follow since DOF is not a measurable, tangible entity. It is up to one's individual perception of DOF that makes the case. It is less noticeable with smaller apertures, wider focal lengths, and at longer shooting distances from the subject. Portrait photography is where this difference is most noticeable due to the specific conditions that are related to the subject.</em></p>

<p>I hope this brings the DOF quibble to an end...</p>

<p>@ Shun --</p>

<p>I also find it interesting that Hocus Focus is using a 5DMkII, D700, and D300... The only reason I can fathom is that the OP was second shooting for a pro that lent out a 5DMkII and lenses. Any light shed on this would be helpful.</p>

<p>RS</p>

<p> </p>

<p>Richard: DOF is a mathematical entity and it can be calculated very precisely, but only within the context of certain assumptions. To really understand DOF conceptually, you have to understand what variables are involved.</p>

<p>Here is a thought experiment: In a controlled studio environment, with a stationary tripod, shoot a scene using the same 24mm f/1.4 lens at f/1.4 using both FX and DX cameras. Print the resulting images at, say, 20x the size of the sensor. Then crop the FX print to match the narrower framing of the DX print. How do the two resulting prints differ?</p>

<p>No discussion of depth of field is meaningful without a definition of the acceptable circle of confusion, but, on balance, I'd say...</p>

<p>I'd say it's time to stop feeding the troll.</p>

<p>I have to agree that I'd rather stick to one camera, but I can imagine that a D300 as a back-up for a D700 makes some sense, and in the scenario of already having the cameras, a couple of expensive primes might trump four of them. I'd be really sure you want the primes, though - a fast zoom does sound like a better solution for the stated scenarios unless you really want a shallow DoF.<br />

<br />

Craig: I agree, it's entirely possible to be confused about this if you think about it in terms of the ISO (sensitivity to light per unit area) and the aperture actually changing. However, it's also completely valid to ignore that part of the calculation, and - if someone's going to claim that "a 24mm lens used on a DX camera is equivalent to a 35mm lens on an FX camera" and "DX cameras have less depth of field" - also discuss exactly what happens to the depth of field if you think of the lens on the DX camera in terms of a lens used on FX. If you think of this scenario in terms of the "virtual FX lens" having a smaller aperture, as you say, you have to think of the sensitivity changing to explain the lack of difference in exposure time; stated in terms of ISO this makes no sense, but thought of in terms of light per sensor pixel if <i>does</i> make sense; you just have to accept that the amplification setting that the camera reports as its ISO value is not the quantity that you should be thinking about.<br />

<br />

However, the ways in which a newbie might be confused by the "equivalence" statement does not make it invalid - it just means that we need to know what we're referring to, and compensate for <i>everything</i>.<br />

<br />

Generally, when we talk about equivalent lenses for different sensor sizes, we assume that the lens is in the same place (I get to the limits of my optics here, but I'm going to suggest that the <i>entrance pupil</i> needs to be in the same place) and that we're trying to replicate the same field of view - that is, we're trying to take indistinguishable pictures with two different formats. If we move the camera, the perspective changes; if we use a different (equivalent) focal length, the cropping changes. If we talk about the effect of a change in focal length on depth of field, we do need to consider whether we've already moved things, but in the context of a changed sensor size and the premise that we already have an "equivalent focal length", I think it's clear what we mean.<br />

<br />

And we're talking in terms of a generalised lens, not any particular lens that may have focus breathing or other concerns. Strictly, the distance to the subject has an effect on the effective focal length of the lens, because the back focus distance changes, but this isn't so significant at normal distances. If it affects anything, it's the "effective focal length", not the aperture.<br />

<br />

I strongly disapprove of anyone plugging numbers into an on-line depth of field calculator and trying to extrapolate from the result - the magic in the box isn't that complicated. (Not to suggest that Craig does this, I just don't think it's the right way to learn about what's going on - although it's a useful tool for specific cases. I've seen a lot of DoF discussions which claim "DoFmaster says..." without further explanation.) It's much less confusing to think of the geometry of the real world. If we first set up two lenses so that two cameras with different sensor formats have exactly the same field of view (by scaling the focal length of the lens by the difference in sensor size), and we eventually intend to produce prints of the same total size from each, the depth of field is solely determined by the size and position of the entrance aperture: there's a cone of confusion passing from the entrance aperture through every point in the focal plane in the scene. What goes on behind the lens has no effect on this - to get the same depth of field (and size of background blur) it's the entrance aperture size that matters.<br />

<br />

F-stop is focal length divided by aperture. Increase the focal length by a factor of "x" and you need to reduce the f-stop by the same factor to keep the entrance aperture the same. Lens in the same place, same cones of confusion, same field of view, same magnification, same image.<br />

<br />

I would hope that DoFmaster scales the CoC down by the crop factor in the assumption that you're enlarging the result by a corresponding amount, because it talks in terms of the CoC at the sensor - this lines up with the reduced aperture that we're talking about for a larger sensor (since CoC at the same distance from the focal place scales with aperture). Generally speaking I hate talking about depth of field anyway, because it depends so much on how much you're magnifying the final image. However, that doesn't invalidate looking at the relative effect of aperture on depth of field as the format changes.<br />

<br />

But I agree that this is a subject that causes a lot of confusion, and that we should be careful to state our premises, and to warn novices to learn the "conventional" way of thinking about a camera first. (That said, I'm not convinced the "crop factor" is ever a useful concept for novices who don't come with a predisposed idea of the field of view of an 85mm lens...)</p>

 

<p>there are times when dual format makes sense and times when it doesn't. some of this depends on what lenses you are using. for instance, the 24-70 and 70-200 shine on both FX and DX, but i haven't done extensive switching with primes. i tend to use the 50/1.4 only on FX, though i suppose i could use it on DX as well. but most of the time in my DX kit i'm carrying 2 zooms and only one fast prime, and i usually go for the 30/1.4.</p>

<p>besides the confusion over DoF--yes, FX cameras have shallower DoF, but the aperture remains the same: 1.4 is 1.4 on both formats--it can be confusing with FF UWAs, which aren't quite so ultra on DX. it doesn't really make sense to carry DX-only lenses along with FX lenses, although something like the tokina 17/3.5 can work on both formats. really, the whole point of this is to carry less lenses, not more, and switch when you need more or less reach. if i was overly concerned about DoF and losing the "magic" of a certain lens, it would probably drive me crazy.</p>

<p>it's great when you need more reach from a tele lens, but if you're shooting, say, an all-day concert and the light drops in the evening, then the high-ISO limitations of DX come into play. in that situation, i'd probably rather have a 2-body FX combo.</p>

<p>i like the idea of mihai's kit, though. the only way to really know if this works for you is to try it for yourself.</p>

<p><em>The single biggest benefit of an FX/Full Frame sensor is it's behavior when using large aperture lenses.</em></p>

<p>well, the lower noise capabilities of FX at high ISOs are a pretty big benefit to that format, IMO.</p>

<p> </p>