Why is this photo not sharper?

[eric_arnold]

<blockquote>

<p>What remains unexplained (to me) is that the camera reported multiple focus points on the subject's body as well as on the bark.</p>

</blockquote>

<p>i never use the dummy AF modes for precisely this reason. cameras cannot defeat the laws of physics, so magically extending the focal plane was unrealistic. what likely happened was the camera defaulted to the most-in-focus area, the tree bark. the low-contrast moving subject was deemed a lower priority. as Elliot noted, you can prevent this by controlling the focus point manually.</p>

[ShunCheung]

<blockquote>

<p>Like I said before, the Nikon D7000 auto focus modes, (AF-A, AF-S, AF-C), focus more quickly and accurately in bright contrasty conditions, rather than in dim light, or open shade.</p>

</blockquote>

<p>Every Nikon DSLR I have used behaves that way; the D7000 is no different. More light (to a reasonable degree) helps us humans see better and also helps AF systems.</p>

<p>I too always select my one initial AF point. If you let the camera select multiple focus points, the camera may not always select what you want to be in focus.</p>

[barryhaugen]

<p>Guys, your get way to complicated with this. This just boils down to basic photography. It has nothing to do with "Pro" grade, focus points or calibration. The reality is this shot was done at f5.6 @max zoom creating a very shallow depth of field. With this camera/lens combo at 85mm@ f5.6 the DOF is about 2.5ft at 16ft. At f11 that distance becomes 5ft. Most lens are at their sharpest at f8 or f11. </p>

[ShunCheung]

<blockquote>

<p>The reality is this shot was done at f5.6 @max zoom creating a very shallow depth of field. With this camera/lens combo at 85mm@ f5.6 the DOF is about 2.5ft at 16ft.</p>

</blockquote>

<p>Barry, however, a DOF os 2.5 feet should be more than enough to make the entire head in focus, if the photographer focuses correctly (selecting the one appropriate AF point).</p>

<p>And BTW, the OP suggests that the subject was about 30 feet away, not 16.</p>

[barryhaugen]

<p>My point being that doing blame the equipment for being faulty or slightly unsharp images. I misread the OP about the distance. At 30ft yes the DOF is around 12-13ft but if the focal plane is near the near limit or far limit of the DOF area the subject will not as sharp. Yes I agree Shun that the use a single focus point will tell the camera exactly were the photographer wants the focal plane to be. By taking 3 or 4 strides closer to the subject and using f8 or f11 would have resulted in a sharper image at a shorter focal length and still filled the frame with the subject. </p>

[eric_arnold]

<blockquote>

<p>This just boils down to basic photography. It has nothing to do with "Pro" grade, focus points or calibration</p>

</blockquote>

<p>actually it has a lot to do with focus points, since the camera focused on the wrong area, i.e. the tree. had the OP manually selected the focus point, this might not have happened. i say 'might' because there is still the possibility that the combination of slow lens + low-contrast + dim shade conditions might have still challenged the d7000's AF module, depending on what the shooter focused on. this is why i suggested grabbing an 'edge' like the microphone or the glasses in my earlier post.</p>

[ShunCheung]

<p>I would put the AF point on the glasses rather than the microphone; that may make little difference in this particular case, but when the depth of field is shallow, the microphone could be in focus while the eyes are not.</p>

<p>Over a year ago when my D7000 was quite new, I found out that it is very demanding on lenses. My 500mm/f4 AF-S has served me very well since 1998 and provides very sharp images on my F5, F100, to D2X and D300 wide open at f4. However, on the D7000, when I pixel peep, that lens simply cannot create totally sharp images at f4. I locked it down on a big tripod with Wimberley head and manual focus with live view + exposure delay. Still, I need to close down to f5.6 to get really sharp images on the D7000 when I pixel peep. Therefore, the chance is that you'll benefit from stopping down a bit. Of course, if you don't pixel peep, perhaps it is not an issue.</p>

[bob_sunley]

<p>The exif data states the lens was focused on something 9.44 meters/ 30.5 feet away. What that was is another story, along with the distance to the subject.</p>

 

[eric_arnold]

<blockquote>

<p>I would put the AF point on the glasses rather than the microphone; that may make little difference in this particular case, but when the depth of field is shallow, the microphone could be in focus while the eyes are not.</p>

</blockquote>

<p>5.6 is medium as far as DoF--what you're describing is pretty much a given at wider apertures like 1.6 or f/2 or 2.8 (which is a moot point with the 16-85). but the reason to put the focus point on the mic is that its less liable to be subject to a herky-jerky head motion, as people sometimes do, and thus easier to acquire and lock focus;you just have to make sure your DoF is deep enough. (old concert photographers' trick) in any event, sounds like the OP could definitely have benefited from stopping down a click or two-- as well as using a faster shutter and just one focus point.</p>

[ShunCheung]

<p>Yesterday, I happened to capture the attached image with a fixed 85mm lens (my own copy of the 85mm/f1.8 AF-S) with the D800 at ISO 200 and 1/400 sec, the lens stopped down to f2.8.</p>

<p>I suppose the only similarity with the OP's image is focal length. Otherwise, I was shooting FX and f2.8 with a shallower depth of field. The camera was also much closer to the subject.</p>

<p>As you can see, at f2.8, depth of field is still quite shallow. I used single AF point focusing on the left eye (on our right). The girl's finger would probably be the same distance from her eyes as the microphone would be. My point is that I would form the habit to AF onto the eyes, not something slightly in front. When a subject is wearing glasses, focus on the glasses to approximate the eyes. In my case, even though I had used f5.6, I don't think I could have gotten away with focusing on her finger in front.</p>

<p>The large image is pixel-level crop from the D800. The inset shows the entire frame.</p><div></div>

[miguel_martinez4]

<p>I also have doubts regarding VR. It seems to me as if the shutter release wasn't depressed for long enough for VR to stabilise, adding extra motion blur.</p>

[ShunCheung]

<p>Here is another example from the same sequence, but the girl has turned a bit. In this case her right eye (to our left) is very sharp but her left eye is actually a bit out of focus. It is still the D800 with the 85mm/f1.8 AF-S at f2.8. Depth of field is indeed very shallow.</p>

<p>For good practice:</p>

<ul>

<li>Select one AF point</li>

<li>Focus on the eye</li>

<li>Use a higher shutter speed if possible</li>

<li>And I'll try to use f4 or even f5.6 more often when I use the 85mm lens. However, I printed an image that is essentially identical to this one to 8.5x11 and both eyes look sharp in the print. You see the left eye to be out of focus only when you pixel peep.</li>

</ul>

<p>Clearly some of these are contradicting issues. E.g. when you need to stop down, it also limits the shutter speeds you can choose from, unless you go to very high ISO.</p><div></div>

[Sycamoe]

<p>I would be willing to bet my kit that if the shooter used single AF, put the center focusing point on the eye, recomposed and shot, the pic would be excellent. It could be done in a split second. I have the 16-85 and a D200 and based on what I can see of this situation, it would be a piece of cake. It's far from a perfect lens but with realistic expectations, it's tack sharp and the VR certainly works well enough at that shutter speed (How fast could the guy's head be moving?). That said, of course the lens is a little better stopped down. </p>

<p>This is why I hesitate to use continuous AF unless there's a compelling reason to use it. To me continuous AF and/or multi points are for sports and wildlife. Although I don't have the long glass for those specialties so I don't do much of it, I understand the concept that you can't use single point, focus and recompose when something is moving like a bird or a running back. This situation is neither.</p>

[ShunCheung]

<p>If you read the opening post, this was what the OP wrote:</p>

<blockquote>

<p>The photo metadata shows a scattering of focus points that the camera thought were in focus: the tree bark immediately to the left of the face; the tip of the microphone; the fingers on the microphone; the collar on the shirt; the logo on the shirt; and others. There was no focus point on the face itself.</p>

</blockquote>

<p>It sounds like the OP let the camera choose where to AF. That is a practice I would avoid.</p>

[greg_janee]

<p>I'm the OP, and yes, the camera was in "Auto-area AF" mode as I think Nikon calls it. I'm definitely going to use Single-point AF for this type of scene from now on, as this is not the first time that the D7000 has screwed up (relative to my intentions). And, coming from a D40 with a grand total of 3 focus points, I'm actually more accustomed to using a selected focus point.</p>

[Sycamoe]

<p>Bet you're going to love that camera as you get used to it. Congrats on a good investment.</p>

 

[blu_dek_lawoe]

<p>I sometimes have had the same problem as the original post. After reading this today, I went out side and did a quick focus test. I used my 85mm 1.8D, all were shot at f4, AF-S. Some in the shade. I've got rid of the purple fringes and better results. However, I don't think the D7000 with a portrait will be able to show sharp nose hairs like the D800. Correct me if I'm wrong. It's twice the sensor size so it's not a good comparison. What I have just tested are hand held. And I have learned a lot what I have done wrong in the past 3 weeks since I got the D7000. Thanks.<img src="../photodb/folder?folder_id=1030999" alt="" /></p>

[eric_arnold]

<blockquote>

<p>even though I had used f5.6, I don't think I could have gotten away with focusing on her finger in front.</p>

</blockquote>

<p>there's no way to know that for sure from comparing a picture at 2.8 which is 2 full stops shallower DoF. in my experience that's enough wiggle room to pull it off. i should probably add this trick (for capturing subjects with fast/unpredictable movements subjects) works even better with a side angle than straight away.</p>