Hi,

I have a favor to ask.

I did a group shot yesterday with my D800 and a Nikon 24-70. The center of the shot seems OK but the edges seem much more blurry.

Will somebody be willing to look at the NEF at my Web site and let me know what you think is happening?

Is there a lens problem, or a me problem?

www.osthus.us/group/

Visit my Nikonians gallery.

Visit my Nikonians gallery.

I think it's a you problem. The EXIF data shows you were at f/5.6, ISO100, 1/320s. ISO is great, shutter speed is great, but at f/5.6 you may not have enough depth of field. As well, there's a lot of haze and some water mist in the air which means that the distant parts of the scene are filtered through all the haze and mist - which makes for soft focus.

Forget about all that if depth of field is not your concern - that is, if you're only concerned about the edges adjacent to the group. Most lenses are slightly softer at the edges (compared to the center). If you want to guarantee sharper edges, crop.

Groups like this that are spread out w-i-d-e across the frame are bound to appear softer at the edges (e.g., Mr. Hardhat on camera right). The sharpest, cleanest shots of this type begin with a better organized group position - three rows of people, tallest in back, shortest in front (with a few exceptions here and there). What that positioning does, simply, is to position the group inside the sweet spot of the lens.

If it's not possible to organize the group in a proper position/pose, then all you're usually ever going to get is a snapshot - like the one you've already got.

My Nikonians Gallery

Howard Carson, Managing Editor
Kickstartnews Inc. - http://www.kickstartnews.com

You see the drop off in clarity as you go from the center towards the edge. I would have thought that since you are shooting at f/5.6 that you would have been fine with DOF. It is clear in the center so it isn't camera shake. I don't recall my 24-70 looking this soft towards the edges. I'll have to check mine this weekend.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

I think you should have been fine with the DOF for 24mm at 5.6. The 24-70mm at 24mm and f/5.6 does have fall off from the center out to the edges. If I want tack sharp at 24mm I try to keep my subject in the middle 50% of the image or I'll change to 35mm for sharpness from edge to edge.

In your photo the center is sharp, but not as sharp as I'd expect for a D800 and 24-70mm lens. And because the image is ever so slightly softer as you move from the center to the left or right, I'd suggest you had a miniscule movement of the camera when you shot. Much like pressing the shutter and the camera very slightly rotate around the axis from pressing on the shutter release.

If you want tack sharp with images like this, the guidelines in the D800 technical manual would apply.

Visit my Nikonians gallery.

I've got a little different interpretation. I downloaded the NEF and looked at it closely.

It's not depth of field. While the center of the image is tack sharp, the outer areas are not. My first thought was that the distance to the outer edges was longer leading to the softness. While that is true, it is normally minimized in modern lens design. And there is nothing truly sharp in the outer areas of the image in any plane.

There are a lot of grooves in the concrete visible at 100% I see those grooves in the center area of the image, but they are very soft as you move to the outer 25-30%.

My guess is the problem is that you were using a zoom lens at its widest focal length. Most zoom lenses are a little soft at the extremes and that is the main thing going on here. You just hit the design limit and have a camera that can show that level of detail.

There may not have been much you could do to change the outcome without using a wider zoom. I would probably edit the image by applying a vignette of increased contrast and increased sharpening at a little higher radius than normal. You could either apply it to the outer area or as a negative selection applied to the center.

I don't think an image like this suffers too much from being just a little soft across the crowd. It's a good image and captures the scene and group nicely.

Eric Bowles
Nikonians Team
My Gallery
Workshops - Smokies Oct 2012

Nikonians membership — my most important photographic investment, after the camera

Thank you all.

I believe I was bracing the camera on a railing.

This is the first time I have noticed any softness in the 24-70 other than when I blew the focus. But it's also the first shot I have tried like this.

I am feeling like his camera is much more difficult to shoot sharp than others. Maybe is my eyes but I also feel the autofocus is not as good as it was in my D700. The auto exposure seems different too.

I have read the sharp shooting manual and use those methods when practical. But a tripod with timer and mirror up and exposure delay mode is not always practical.

I'm starting to feel this is not a practical camera. With my Honeywell sp 1000 I could get shots so sharp they were like a razor edge, without jumping hoops. A large part of what I shoot is sort of like pj work. Maybe this is the wrong tool for that.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

>Hi,
>
>I have a favor to ask.
>
>I did a group shot yesterday with my D800 and a Nikon 24-70.
>The center of the shot seems OK but the edges seem much more
>blurry.
>
>Will somebody be willing to look at the NEF at my Web site and
>let me know what you think is happening?
>
>Is there a lens problem, or a me problem?
>
>www.osthus.us/group/
>
>


I think it is a lens "problem". Notice the quotes.
You are in DOF from the rails to infinity.
For such a lens F5.6 should be the sweet spot so stopping down further would not help unless something is wrong with your lens.
What I see for sure is that your lens has huge CA 2/3 from the center.
Another explanation is that at 24mm you are in a field curvature.
If you focus closer or tighter you might have avoided field curvature.
Do a newspaper on the wall test to find out if it's your lens.

Thank you. What do you mean by problem in quotes?

What settings and distance you recommend for the newspaper test?

Much appreciated.

PS. how can a CA problem be fixed? Would this be from something out of alignment? The lens is still under warranty.

Visit my Nikonians gallery.

Visit my Nikonians gallery.

What several respondents have referred to are the various technical limits of lenses. If we use a lens to make a certain shot, and certain portions of the shot are situated in weak areas of the lens, there's nothing wrong with the lens. Rather, we're not using the lens in the best way. We're not respecting the technical limits of the lens. The 'problem' with such lenses are the photographers who are using them. Me too, from time to time.

It's yet another reason why we can shoot so many hundreds (or thousands) of photos on a particular trip, outing, event or period of time, and only end up with a few dozen (or even fewer) genuine keepers.

As expensive as our cameras and lenses happen to be, they still have technical limits, areas or functions that are imperfect, and so on. I recall that it took me at least six months, using my 24-70 f/2.8, to begin making a respectable percentage of publishable shots out of thing on successive photography trips. Learning the strengths and the limitations of a particular favorite lens takes time, lots of photos, lots of examination, and lots of reminders-to-self about what works and what doesn't work.

Your CA question wasn't directed at me, but the answer has little to do with lens alignment or service calibration. (From Wikipedia): "Chromatic aberration (CA, also called achromatism or chromatic distortion) is a type of distortion in which there is a failure of a lens to focus all colors to the same convergence point. It occurs because lenses have a different refractive index for different wavelengths of light (the dispersion of the lens). The refractive index decreases with increasing wavelength.

Chromatic aberration manifests itself as "fringes" of color along boundaries that separate dark and bright parts of the image, because each color in the optical spectrum cannot be focused at a single common point. Since the focal length of a lens is dependent on the refractive index, different wavelengths of light will be focused on different positions. In the earliest uses of lenses, chromatic aberration was reduced by increasing the focal length of the lens where possible (which makes convergence positioning less critical)."

To varying extents, CA exists in single lenses (e.g., eyeglasses) just as it exists in multi-element photography lenses. There's nothing wrong with your lens. The minor CA exhibited by the lens you used is barely noticeable and can be reduced in some photo editing/post-processing software. It's a snapshot you've got though and I wouldn't waste time on it worrying about minor CA.

My Nikonians Gallery

Howard Carson, Managing Editor
Kickstartnews Inc. - http://www.kickstartnews.com

I''m no expert but isn't this problem a problem of diffraction?
Deep down I think if it was shot with about f/8 or 11 it would have been different. Also I do not know where in the image you focused.

I am very interested in this thread; a lot of excellent comments are being presented.
Thank you
-------------
Please visit my galleries: Reza Gorji Photography

>I''m no expert but isn't this problem a problem of
>diffraction?
>Deep down I think if it was shot with about f/8 or 11 it would
>have been different. Also I do not know where in the image
>you focused.

The OP's photo was shot at shot at f/5.6, so there cannot be any visible diffraction. With an aperture of f/16, diffraction might have been a softening factor in this type of shot, but it was made at a much larger aperture.

My Nikonians Gallery

Howard Carson, Managing Editor
Kickstartnews Inc. - http://www.kickstartnews.com

I apologize for my ignorance. I think I meant refraction but maybe I should just shut up and read!!! LOL@@@@

-------------
Please visit my galleries: Reza Gorji Photography

Meh - it's nothing. Still not refraction though. Refraction takes places when light passes through a medium such as water, glass and so on. Certainly refraction can be a quality altering or quality reducing factor in some very cheap lenses, but I don't think it's the OP's problem.

Please keep posting.

My Nikonians Gallery

Howard Carson, Managing Editor
Kickstartnews Inc. - http://www.kickstartnews.com

Thank you Howard.
Reza
-------------
Please visit my galleries: Reza Gorji Photography

Testing my 24-70mm on my D800E shows similar behavior at 24mm.... to get better edge to edge, you could have tried stopping down to F8, which helps with my copy of the lens.... As noted by others, there is some field curvature at the wide end of the lens also that can effect things.. But over all not a bad image, just sharpen outer areas and should be ok for most normal sized prints I would think...

Chris

Visit my Nikonians gallery.

The issue is caused by field curvature. I've seen this so many times with this lens. The 24-70 has strong field curvature at 24mm. The lens is still sharp at the edges, especially when stopped down to f/5.6 like in your case. But towards the edges of the frame, the plane of focus curves inwards, towards your position. I've had quite a few problems with this lens when trying to use it for flat subjects or for landscapes, until I learned it's characteristics. Now I usually avoid using it for such situations. If that can't be avoided, there is a way to deal with field curvature: You need to set the focus point farther away than the subject, but with the subject still within the DOF in the image center. This will bring the plane of focus at the edges out far enough to have the subject in focus there as well. Stopping down helps of course as well. Of course when you have people in the image, then experimenting with the focus distance is much less practical than when doing landscapes. Your best bet may be a prime lens, even cheap ones like the 24/2.8 or 35/2 have way less field curvature than fast zooms. And even slower consumer zoom lenses like the 24-85 and 28-105 handle this situation better, I have compared them.

Tom
Bay Area Nikonian

http://www.tkphoto.me/

>The issue is caused by field curvature. I've seen this so
>many times with this lens. The 24-70 has strong field
>curvature at 24mm. The lens is still sharp at the edges,
>especially when stopped down to f/5.6 like in your case. But
>towards the edges of the frame, the plane of focus curves
>inwards, towards your position.

I have this lens too, John and am very interested in your (Different) problem. Thanks for making the photo available in NEF so that all of us can view the symptoms, your focus point (ViewNX), etc. The focus point is clearly shown on the center back (man in yellow green shirt (no sunglasses) and man in blue plaid shirt to his front left. If you estimate 40-50 ft to the subjects, DOF calculations yield near limit from 8-9 ft to inf. In other words, the focus point appears just right to me.

There is plenty of evidence for the field curvature issue Tom cites. My thought is that field curvature needs to be dealt with separately from AF. We may be able to compensate by changing our focus points or even considering a different focus method, LV. After you read this, see what you think.

From the Nikkor AF-S 24-70mm f/2.8G Photozone review by Markus Stamm

http://www.photozone.de/nikon_ff/456-nikkor_afs_2470_28_ff

“When we reviewed this lens on our DX test camera we discovered a rather huge amount of field curvature at 24 mm. On the D3x it turns out the image field is not simply curved, but has a dent where the border of the DX frame happens to be. Further towards the edges, the images field folds back again.

Verdict
…
It's not without flaws, though. … The wavy image field at the low end might be an issue if you shoot flat objects and last but not least the lens is not exactly cheap.”

Klaus Schroiff reviewed the Canon EF 24-70mm f/2.8 USM L for Photozone. He described a similar "focus field curvature" showing a diagram, for the Canon 24-70mm here

http://www.photozone.de/canon_eos_ff/528-canon2470f28ff?start=1

“Remember that we mentioned that the outer focus field is distorted towards background? Well, the solution is to "pull" the focus field towards you. The sample crops ABOVE the sample scenes show the effect of this action - the corners are pulled into focus and the center remains sharp (within the depth-of-field). This "workaround" is viable at small to medium aperture settings but it does naturally fail at large apertures where the depth-of-field is not sufficient for such an approach anymore.“

My interpretation of Klaus’ diagram and ‘pulling’ the focus field toward you is

- the focus field curves back away from the you, the photographer
- to pull it toward you, you move the focus point toward you, in front of the subject (s)

Is this right? I am not at all sure of my interpretation? At this point, I believe I would adjust the focus point from back to front to see if there was improvement.

I’m going to do some tests with my D800 and the 24-70. Unfortunately, I don’t have 50 willing subjects, so I’ll have to try to simulate the setting with some suitable detail – no substitute for the human face, I’m afraid. Has anyone tried Live View for a group photograph like this?

And remember, when Howard says

>Combine the foregoing (Autofocus discussion) with the additional vagaries of lens design and the concommitant need to understand the strengths and weaknesses of a particular lens.

I have to say, 'Yep - agree'!

David

Visit my Nikonians gallery.

i made some small corrections in sharpness and clarity, then let LR due its thing for lens profiles.
then sent it to cs6 for the "adaptive wide angle adjustment", straightened the lightpole on right and fence pole on left so they were straight to side of photo.

after this you can really see the curve created by a 24mm shot.

i am not smart enough to know if f8 or 11 would have fixed, but after discussions above and my short trip to cs6 and Lr with this photo, i am smarter than before.

tell them to all get out there again and try with f/11 :<)

My Gallery

I seem to always be the odd ball in questions like these. I see something a little different then everyone else. If I look at the hand rail and steel plate with all the screws in it in front of the group I see them as being mostly sharp side to side. Certainly more so then the group is side to side. This is especially so for the lower left corner at the hand rail. Missed focus and lens field curvature?

.
.


Scott Martin Sternberg

Scotts Fine Art

>I seem to always be the odd ball in questions like these. I
>see something a little different then everyone else. If I look
>at the hand rail and steel plate with all the screws in it in
>front of the group I see them as being mostly sharp side to
>side. Certainly more so then the group is side to side. This
>is especially so for the lower left corner at the hand rail.
>Missed focus and lens field curvature?

You know what? You are correct!!!!!

This is what I love about this forum, I would not have noticed that. But there it is. Still wondering how the center of the group is sharper than the edges, but it looks like if they were at that front railing it wouldn't be a problem.

This raises some other questions:


I am seeing many instances where autofocus with my d700 and now this camera is not cutting it. I used autofocus in this image. I believe i put the sensor right on the center of the group. If autofocus is not reliable, what good is it? I understand not shooting a lion through a cage and relying on autofocus but why not this shot?

In other instances I was shooting houses in Carmel or valleys in Yosemite using autofocus and found some of the shots blurry and the focus data saying the distance was 10 feet or whatever. Why?

In the old days I had a manual focus camera and it worked great. Now I don't trust my old eyes any more. Do they make some sort of manual focus tool for guys with bad eyes who have lost faith in autofocus?

Do "real" pro shooters even use autofocus much?

Visit my Nikonians gallery.

Visit my Nikonians gallery.

We've had a separate thread about focus precision and focus accuracy. LensRentals had a good series of articles on the issues. Like any other mechanical process, focus is not perfect. It's easy to have a subject that fools the camera. It's also possible for focus to have small variation since it is a mechanical process.

Keep in mind that AF replaced manual focus. AF has been getting better and better through each subsequent generation of camera, but it will continue to improve and it is certainly not perfect.


Eric Bowles
Nikonians Team
My Gallery
Workshops - Smokies Oct 2012

Nikonians membership — my most important photographic investment, after the camera

Eric speaks the most important truth(s) abut autofocus I think.

Darrell Young and Thom Hogan, in their respective books on the D700, devote very large amounts of dissection, analysis and rational explanation of the ways in which the AF system best works. They're both recognized as very successful pros, so their extensive attention to the subject tells us mainly that AF is far from the micrometer-accurat, fully automated system that the camera makers and marketers would have us rely on. At a minimum, these latest AF systems require use by photographers who understand how best to use them. It's not just a point and shoot situation.

Combine the foregoing with the additional vagaries of lens design and the concommitant need to understand the strengths and weaknesses of a particular lens.

My Nikonians Gallery

Howard Carson, Managing Editor
Kickstartnews Inc. - http://www.kickstartnews.com

I believe that these are very valid points. From my experience with the D700, the AF of my camera is quite accurate. However, under certain conditions like large aperture or far away objects, the precision of the AF system could be better. Apparently, the range within which the AF system considers the object "in focus" exceeds the DOF under certain conditions. This should be even more apparent with the high resolution D800 which favors the use of large apertures. We should be aware that no amount of "fine tuning" improves the precision. All it does it "centers" the error distribution curve.

For the most precise (and accurate) focus I have found the manual focus using high magnification in LiveView to be superior to AF, provided the object is stationary or slow moving. We should note that professional movie and tv lenses generally use MF.

It almost seems that AF works best with low resolution cameras, small apertures and mediocre lenses...

Tristan

Visit my Nikonians gallery.

Visit my Nikonians gallery.

After waiting several weeks for word from Nikon I called a third time. I called once to ask what's napping. They said they would get back to me. I called a week ago and they said they would get back to me. So I called again today to find out what's happening and why they were not keeping their commitments.

I must have got a new guy because he immediately said they were looking into a problem with the d800 after getting many calls starting in July.

I asked why the camera came back from repair still not focusing and that stumped him.

In any case Nikon is starting to acknowledge that they have a problem.

Geez...

Visit my Nikonians gallery.

Visit my Nikonians gallery.