Sensor/film size and DOF

[ashim]

<p >I am sure this would sound stupid but I need clear this at the earliest. So pls share your guidance.</p>

<p > </p>

<p >I was under the impression that as the sensor size increases, the DOF decreases – everything else remaining the same. One of the reasons, I thought, it is said that compacts have more DOF than a DSLR.</p>

<p > </p>

<p >Today, while doing some calculation on DOF master, I found that for the same lens, subject distance and f-values, the DOF is more on a D700 than a D300. When I changed the camera to medium format, the DOF increased even more!</p>

<p > </p>

<p >Was I missing something? Pls clarify. </p>

[photo_galleries]

<p>Facors that impact DOF: 1) aperture, 2) focal length, 3) subject distance. That is why those are the 3 variables that you can change at the dofmaster site. Sensor size, in and of itself, does not impact DOF; HOWEVER, the resulting crop factor does.</p>

<p>"<strong><em>One of the reasons, I thought, it is said that compacts have more DOF than a DSLR."</em></strong></p>

<p>Yes, but not because of sensor size. The ACTUAL focal length on a digicam is A LOT shorter than a "normal" DSLR lens (yes, I know that digicams are marketed with a "35mm equivalent focal length," but that is not the ACTUAL focal length).</p>

<p> </p>

[ashim]

<p >Thanks, Keith. </p>

<p > </p>

<p >I have found my mistake now. DOF is a function of magnification (FL and subject distance) and f-stop. I was comparing different magnification on different cameras. For the same magnification and f-stop, the DOF is more on a smaller sized sensor.</p>

<p > </p>

<p >Sorry for the quick post and bothering all, thanks.</p>

[richard_driscoll]

<p>Ashim,<br /> A rule of thumb is that a DX camera will give about the same DOF as an FX camera stopped down one stop when using a lens with the same angle of view. It turns out that if the crop factor were exactly 1.41 then the difference would be exactly 1 stop; since it is 1.52 then the difference is more like 1.2 stops.<br /> So a 35 mm lens on a DX camera at, say, f/8 gives the same DOF as a 53 mm lens on an FX camera lens at about f/11. In fact it's more like f/12.<br /> It gets confusing if you use a lens designed for FX on a DX camera and try to use the DOF scale, since the depth on DX will be <strong><em>less </em> </strong> than the DOF scale indicates. That is because on DX the lens is no longer wide angle but normal.</p>

[trevans]

<p><a href="http://www.dofmaster.com/dofjs.html">Depth of Field Calculator</a></p>

[luisarguelles]

<p>For keeping it simple, it must be taken into account that what determines DOF is the focal length of a given lens. Focal length is what dictates DOF. Forget about FX, DX formats, digital vs film, etc. Focal length is the key word. Distance and f-stop are only modifiers, that is, things that you can use in order to express your vision in a photographic composition.</p>

<p>As an example, take a Nikkor 50mm lens and a Zeiss 50mm lens for Hasselblad. While they are oriented towards diferent formats (35mm format vs Medium format) you will see that the DOF scale engraved on each lens is<strong> exactly the same</strong> for both lenses for a given distance and a given f-stop on the focal plane.</p>

<p>A closer example: A 50mm f/1.8 AF-D Nikkor lens will produce just the same DOF in a D700 and on a D300. You must realize that the 50mm working on a D300 is only a crop of that lens mounted on the D700, but DOF is exactly the same.</p>

[andylynn]

<p>Where this comes from is the fact that you need different focal lengths to take the same shots.</p>

<p>For example, suppose I use a D700 and a 50mm f/1.8 lens to take a photo of you from 5 feet away. Then I take the same photo using the same lens and a D300. They won't be the same photo - the D300 shot will be cropped. But FWIW, the depth of field will be the same.</p>

<p>So instead I need two lenses - the D700 gets a 50/1.8 and the D300 get a 35/1.8, which is approximately the same field of view, and now I can take the same shot with both cameras. Because the D300 needs to be used with a shrter focal length, it has more depth of field.</p>

[Two23]

<p>It is a MYTH that lens focal length determines DOF. All lenses have the same DOF. A 35mm has the same as a 200mm. DOF is determined by (1) image size (2) aperture. Since an FX image is larger than a DX image, it will have slightly less DOF. I noticed this kind of thing when I shot film. A 35mm slide (small image) had more DOF than a 4x5 (large image.) It is a myth that lens length determines DOF though. Here's the tests:<br /><a href="http://www.film-and-video.com/dofmyth.htm">http://www.film-and-video.com/dofmyth.htm</a><br /><a href="http://www.popphoto.com/Features/How-to/What-s-Up-With-Depth-of-Field">http://www.popphoto.com/Features/How-to/What-s-Up-With-Depth-of-Field</a></p>

<p>The reason the D700 has slightly less DOF than the D300 has nothing to do with the lens--that's crop factor. The D700 produces larger image size, hence slightly less DOF at a given aperture. What lens you compare won't matter. If you shoot a 50mm (or ANY) lens on a D700, it WILL have more DOF if used on a D300 because of image size. This is the same principle as shooting 35mm and 4x5--sensor/film size actually does impact DOF. This has been known for decades. </p>

<p>Kent in SD</p>

<p> </p>

[joel_covey]

<p><a href="http://en.wikipedia.org/wiki/Depth_of_field">http://en.wikipedia.org/wiki/Depth_of_field</a><br>

t'ain't a MYTH t's OPTIC<br>

see all them there "f"s<br>

that's where you put the focal length to make them there DOF numbers come out right</p>

[andylynn]

<p>That's really confusing. Of course focal length is a factor in DOF. True, that's because it affects the magnification, but these variables are all correlated. If I stand in one place and zoom in on you, the background becomes more out of focus.</p>

[richard_driscoll]

<p>This has all been discussed in these pages before.<br>

I recommend Bob Atkins' article:<br>

http://www.photo.net/learn/optics/dofdigital/<br>

It is written from a Canon perspective (1.6 crop factor) rather than Nikon (1.52 crop factor) but most of it still applies.<br>

Personally I keep hold of the fact that a small camera will have more DOF than a large camera when they are fitted with lenses giving the same angle of view, prints are the same size and they are viewed from the same distance. This observation goes back decades.</p>

<p> </p>

[Two23]

<p>Focal length isn't a DOF factor, magnification is. There's a difference. We need to clear up these myths in these online forums because having factual info to work from makes us better photographers in the field. Consider this potential shot: three wild flowers in a rough triangle, about 20 inches apart. You want a close up shot of them where they take up 2/3 of the frame. You try a 90mm macro, but quickly discover you can't get enough DOF to get them all sharp, no matter where you put the focus point. You think, "AHA, I'll use a 20mm lens!" You line up the shot so the composition is the same (move in closer,) and check DOF preview. You will find you have the same problem. So, you back way off and check the DOF preview. DOF will look better, but now you have a parking lot, telephone pole, and other stuff in the frame. You figure you'll just blow up the shot and crop out the composition and print that. What you'll discover is that all that 20mm DOF was an optical illusion. The shot will look exactly the same as if you had used the 90mm. Exactly the same. Just as the exposure stays the same lens to lens at same aperture, so does the DOF. It's physics. Back to the flowers shot wide with a 20mm lens vs. 90mm. Let's say you printed the 90mm shot to 4x6 inch, and the 20mm shot to 4x6. The 20mm print will appear to have more DOF, yes. But, it won't be the same composition--it has parking lot & telephone poles in it. If you were to blow up the shot made with 20mm to 16x20 inch, suddenly all that DOF will disappear and will be the same as how the 90mm looked at 4x6. If you blow up the 20mm lens shot so the composition is the same as the 90mm, DOF is exactly the same. It's not the lens that does that, it's image size. DOF is created by 1) image size 2) aperture. We should get rid of these myths. Generally the wide angle DOF myth is harmless, but there are real instances in the field where believing it can mess you up. <br>

Back to more of the original thread, consider LF lenses. I have three and they can be used on both 4x5 and 8x10. What happens to DOF if use my 90mm Super Angulon on a 4x5, and then put it on an 8x10? While I haven't shot 8x10, those who do tell me there is less DOF in an 8x10 tranny when you use the same lens on a 4x5 tranny. This also comes back to image size.<br>

<a href="http://www.luminous-landscape.com/tutorials/dof2.shtml">http://www.luminous-landscape.com/tutorials/dof2.shtml</a></p>

<p>Kent in SD</p>

[markci]

<p>Kent, your assertion is an approximation. It is not technically true. Depth of field is absolutely a function of focal length. It can be approximately considered a function of magnification, though the approximation degrades at very short distances.</p>

<p>http://en.wikipedia.org/wiki/Depth_of_field</p>

<p>Please stop "correcting" people who are more technically correct than you are. You are adding noise to the discussion.</p>

[andylynn]

<p>Kent: You're not reading what I wrote. I'm not talking about the scenario where you zoom the lens, then walk backward to get the same crop on your subject.</p>

<p>Try the following experiment:</p>

<p>Place two objects on a surface some distance apart. Place your camera with a zoom lens on the same surface so that one object is far and the other close. Zoom in all the way, so you have both objects in the frame, focus on the closer one and take a shot at f/5.6. Now zoom out all the way and focus on the same object, keep f/5.6 and take a shot.</p>

<p>In playback mode, zoom in on the far object in both shots. In the zoomed in shot it will be much blurrier.</p>

<p>Focal length is one of the variables in DOF.</p>

<p>Now, if you happen to have an FX camera that has a DX crop mode, do the same experiment, only instead of focal length you can vary sensor size. Take the shot with the camera in DX mode then the same shot (from the same point, with the same focal length and focus point and f/stop) in FX mode. Zoom in on the back object in both shots. It will look the same.</p>

<p>Sensor size, by itself, does nothing but crop the shot.</p>

<p>Now, if you have a film camera with a 50mm f/1.8 lens and like it a lot, then you get a DX camera and you want the equivalent lens so you get a 35mm f/1.8, you will quickly find that you have more DOF when taking the "same" shots.</p>

[andylynn]

<p>Here's what I'm talking about. Two shots taken at f/5.6 from the same spot, and cropped and resized to the same framing. I used my D90 and 18-105 lens.</p>

[andylynn]

<p>Oops. Here's the image.</p><div></div>

[Two23]

<p>I'll try this later, when I have more light. I was thinking of photos that involve a lot more distance than table top though. One thing I notice about my Nikon 17-55mm lens is that as I zoom it to 20mm, the lens actually gets longer, not shorter. Not sure if that affects things, but I'll check. I still have Nikons 20mm and 50mm lying around to compare. As for calling my posts "noise," note I'm quoting and referencing well respected sources, such as Luminous Landscape and Popular Photo magazine. et al. At any rate, you have my curiousity engaged.</p>

<p>Kent in SD </p>

[andylynn]

<p>Kent,</p>

<p>I don't think my results disagree with the articles. They're talking about different things - for example, if you step back but zoom in to put the same frame around your subject, you've changed two variables - focal length and distance - that cancel each other out. In the case of the distant tower in the background, it wasn't blurred the same in the sense that the number of pixels of image it was blurred by was the same but in the sense that the blur stayed the same proportional to the size of the building in the frame.</p>

<p>But none of them was addressing the case with all other variables removed - just standing in one place and zooming - which is important because the test I did using a crop is the same thing as if I had two cameras, with sensor size differences that made the 18 on one the same FOV as the 105 on the other and I took the same shot. Essentially, that's an exaggerated case of the whole 50mm on FX vs. 35mm of DX thing and it shows why you can hardly ever compress DOF with any P&S camera regardless of how good the lens is, unless you're doing extreme close ups.</p>

<p>This is all assuming that my working definition of DOF (which I believe is the working definition shared by most people) is not in conflict with the technical definition. E.g., in my working definition you can't change the DOF of a shot by cropping it; therefore you can't change the DOF by putting an FX camera in DX mode; therefore it should follow that you can't change the DOF by moving your lens from an FX camera to a DX camera.</p>

[radfordneal]

<p><em>You want a close up shot of them where they take up 2/3 of the frame. You try a 90mm macro, but quickly discover you can't get enough DOF to get them all sharp, no matter where you put the focus point. You think, "AHA, I'll use a 20mm lens!" You line up the shot so the composition is the same (move in closer,) and check DOF preview. You will find you have the same problem. So, you back way off and check the DOF preview. DOF will look better, but now you have a parking lot, telephone pole, and other stuff in the frame. You figure you'll just blow up the shot and crop out the composition and print that. What you'll discover is that all that 20mm DOF was an optical illusion. The shot will look exactly the same as if you had used the 90mm. Exactly the same.</em></p>

<p>This is not correct. When you decide to crop the 20mm image, you can put the camera in the same place as for the 90mm shot. The degree to which out-of-focus areas are blurred will then depend on the actual aperture (in mm, not as an f-stop), which will be less for the 20mm at f/8 than for the 90mm at f/8.</p>

<p>If you don't believe me, I suggest that you <em>actually try this</em> , and see what happens.</p>

[Two23]

<p>I took about 20 shots using three Nikon lenses: 17-55mm f2.8 (@20mm & 50mm), 20mm f2.8 AFD, 50mm f1.8 AFD. First thing I noticed is how much better the 17-55mm is at 20mm than the 20mm is, LOL. First batch I took the shots the same way you did, and got same results. Zoom or single focal lens didn't matter. Second batch I added a third item in the middle and put focus on that. Doing that, didn't see much (if any) difference between 20mm & 50mm, but will check on my big monitor tomorrow. Third set of shots I moved objects in close to preserve the same image size on my viewfinder when shooting 20mm. That also dramatically changed perspective, but it seems that DOF did stay the same on both lenses when I did that. That's viewing on D300 LCD though, will check on big monitor tomorrow. Hmmm. I changed from bottles since they are round, to small same type boxes since they are flat surface. Squared everything to camera, used a checkerboard table cloth for a grid to position boxes precisely. I used D300 AF, since it seems to be spot-on. Used Alien Bees B1600/umbrella for consistent light, camera on Gitzo 1325 tripod/AcraTech. Triggered camera with CyberSyncs, although considering I used 1/250 sync that probably didn't matter any. I tried to eliminate as many variables as I could. One of the reasons I put a third object into two sets, in the middle, was that by focussing on only one of two objects you are getting only half the DOF area. Generally, half of DOF falls in front and half in back, but I think I recall that changes some depending on distance. I tried to eliminate it as a variable. Will post photos tomorrow--too late tonight and wife is using my graphics computer/monitor. Will also take larger scale shots outdoors rather than these table top shots, if it doesn't rain.</p>

<p>Kent in SD</p>

[sjmurray]

<p>I found a good explanation of depth of field with examples and formulas here: http://toothwalker.org/optics/dof.html<br>

Here's an excerpt:<br>

<em>Neither the absolute nor the relative background blur should be used as a criterion to judge the depth of field. Much of the confusion in DOF discussions arises because people base their judgment on out-of-focus parts of the image. DOF should not be judged from background blur. A shallow DOF is not synonymous with a generously blurred background. A shallow DOF implies that there is a shallow region in object space that is rendered acceptably sharp, regardless of whether the background is just not sharp or completely blurred. The <a href="http://www.luminous-landscape.com/tutorials/dof2.shtml"> luminous landscape</a> puts considerable effort into examination of background blur, but to no avail. The <a href="http://toothwalker.org/optics/dof.html#definition">definition</a> of DOF is clear. It concerns the sharp parts of the image, not the blurred parts. </em></p>