Is this element misalignment or something else?

[purpleroan]

I keep getting images like the attached one that just appear soft for some reason (this is straight out of camera). I thought it might have been camera shake, but I was shooting at 1/1600th, and I'm typically a very stable person. Any thoughts? It was taken on a 70-200mm f/2.9 is ii usm with a 5d mk iii body. I'm just about ready to send the lens in but figured I'd post here first. Thanks!!

 

[Overread]

Was the IS on? IS takes time to speed up to be ready to counter handshake and in the initial few moments when its speeding up it can introduce shake of its own. So a quick snap shot might well appear softer than it should. It's why if you're shooting as you are, ergo fast enough that you shouldn't have to worry about handshake, many will turn the IS off so that they can point and shoot as fast as the AF locks without having to wait a few moments to get the IS working.

 

[purpleroan]

Hmm..I believe it was. Actually, I remember turning it on that morning because it had been been off for some other reason on a previous shoot. I wasn't aware that could cause an issue though. Are you saying it must speed up every time focusing occurs? That'd make a lot of sense, as it seems to have happened for the entirety of the shoot. On the other hand, if you're implying its just when its first turned on, then I'm not too sure thats what it may be. Thanks so much for your input!!

 

[Overread]

Each time you press the shutter to half way the IS engages and starts to work. From that moment it has to take a few moments to detect and then counter the vibrations of the lens and camera. So its just like the AF in that it takes a moment to get ready for a shot. IT's not a huge length of time, but in sports it can make a serious difference.

One way to resolve the matter (asides from turning the IS off of course) is to follow the action in advance of the moment of the shot; this way you've already got the shutter half depressed and the IS running. If you use back-button AF control this means you can half press the shutter to get the IS working even if you don't yet need the AF working.

 

[fmw]

There is no camera shake at 1/1600th. It is better to turn the IS off because it can affect image quality negatively. My guess is that the lens is soft at whichever zoom setting you chose - probably fully extended. If an element were loose, the image would be far worse.

 

[photo1x1.com]

To me the grass in front looks sharper than the person.
Have you tried focussing on a still object from a tripod?
If it is a front focus issue, the 5D III has micro adjustment. But that only works if the issue appears over the complete zoom range. Datacolor has a nice tool called spyder lenscal that makes adjustments easier.

 

[TheLibrarian]

I've been considering my own soft focus issues and think I remember something about shooting on the lowest fstop will leave a soft focus. Your f2.9 is better than my kit f3.5 to f4 but only slightly and even still if you're all the way down at 2.9 thats your minimum. Not sure if thats the setting you were using but can try to not test the limits of the lens and keep it up around f8 or so especially for sports and distance shots where you're not going for bokeh or soft focus anyway.

He's moving and you haven't said what you do to get good focus or what focusing mode you were on though with that equipment I'd assume you know various methods and are using them properly.

 

[Overread]

A few thoughts
1) The 70-200mm f2.8 IS L MII is very sharp even at f2.8 at the 200mm end. It can more than stand up to being used wide open at any focal length and should deliver sharp results - all other properties being suitable for a sharp shot.

2) If it were a damaged element I'd expect it to appear every single time without error.

3) A subject moving toward the camera is one of the most challenging things for AF to work with because the subject is constantly moving the plane of focus. This can make it more risky and more likely to get miss-focuses than if the subject is moving across the view instead of toward the viewer.

 

[Braineack]

exif says 1/1000.

but looking at the image, i want to say the camera was moving (look at the BG) and you had IS on and the combo was causing your soft photo.

 

[KmH]

Most of the time IS/VR/OS etc should be turned off.
IS/VR/OS basically can't react fast enough for shutter speeds faster than about 1/500 of a second and then becomes an added battery drain that gives zero benefit.

If your camera has IS/VR/OS in the camera body rather than in the lens it can't correct camera movement with as much amplitude as in the lens IS/VR/OS can.

 

[purpleroan]

Ok, so a couple corrections..
I accidentally posted 2.9 rather than 2.8, and for that I apologize. Furthermore, I posted 1/1600th but it was apparently 1/1000th, however the issue was still occurring at the higher speed, so it's all more or less the same.

I have not tried focusing on a tripod since I don't currently own one, it rarely becomes necessary for my main shoots. It's very interesting though, what many of you are saying about the IS, as this is something I was not aware of. I think I will refrain from sending it off asap, as I have a handful of sports events coming up next weekend, and wish to see if the issue is still present with the IS off. I'm very grateful for all this information, and anybody has anything further to add I'd love to hear it!

 

[benhasajeep]

I think the lens was focused slightly in front of the player. I don't think it's a lens issue as around the edges of the picture it's about the same. If there were a loose element it would show worse in a corner. I also agree with the high shutter speeds IS/VR/OS should be off.

To check your lens. Since you don't have a tripod. Set your camera on a sturdy object. Set a lens test target on something appropriate distance away and focus on it. Then if you don't have a remote release, use the cameras self timer. If you really want to be accurate mirror lockup as well.

Look up lens focus test online. Don't buy anything. Self printed targets are available online. And or you can use something like a ruler!.

 

[TCampbell]

You are attempting to assess the optical quality of the lens using a moving image. Don't do that.

If the lens has an optical flaw, then that flaw would show up even in a shot where the camera and lens are on a tripod and you're taking a photo of a non-moving flat-field subject (such as a brick wall).

There can be lots of reasons for a blown shot... others have already mentioned image stabilization (the new Canon EF 70-200mm f/2.8L IS USM II is EXTREMELY good) but there are other questions such as ... was the camera in "one shot" AF vs. "AI Servo" (continuous AF)?

As I zoom in and read the "Mudcats Alumni" text on his shirt, I see blurring to the left - suggestion horizontal camera movement occurred.

Note that image stabilization isn't a guarantee that you'll get a keeper... it just improves the odds. Also image stabilization helps for camera movement but doesn't do anything for subject movement (where high shutter speeds help). If IS is enabled and you've been tracking, the IS system tries to stabilize (countering your camera motion) but at some point it reaches the limit of travel and has to reset the lens elements... for that moment in time, you'd have blur as the elements shift back to center.)

Before you send anything in for evaluation and service... do a simple test with a non-moving camera and non-moving subject to make sure the lens works in those situations. If you get sharp results the optics are fine. You can then experiment with the IS system to make sure it's working (hand held photography of a non-moving subject.)

Since this was a moving subject, you'd want to make sure you were using continuous focus (AI Servo focus mode) and that the focus point was actually on your subject (you could have a combination of missed focus and motion blur.)

 

[purpleroan]

I'm back everyone! So, after reading all of your replies I decided to try and use the micro adjustment feature on my camera body. I honestly didn't notice much of a difference, and I adjusted it specifically at the long end, which is where I have issues. Perhaps I didn't use a distance such as those I'm shooting at, but in order to test it I would have to have either a giant test sheet or something with equidistant marks, neither of which are practical at the moment for me. I have decided to post another sample to see if I can get any more ideas...again, it is sooc. I realize its a terrible photo in general but I was hoping to catch the issue, which I did. The lens is fine until 100/130mm-ish range, and it seems the effect is more dramatic the farther away the subject is, and the faster they're moving. This photo was shot at 1/800s, f/2.8, iso 125, IS off, at 200mm. It's still possible that it needs to be micro adjusted to a finer and more accurate extent, but I honestly don't see ANY area of sharper focus in the newest example, which leads me to believe that may not be the problem..

 

[photo1x1.com]

I'm back everyone! So, after reading all of your replies I decided to try and use the micro adjustment feature on my camera body. I honestly didn't notice much of a difference, and I adjusted it specifically at the long end, which is where I have issues. Perhaps I didn't use a distance such as those I'm shooting at, but in order to test it I would have to have either a giant test sheet or something with equidistant marks, neither of which are practical at the moment for me. I have decided to post another sample to see if I can get any more ideas...again, it is sooc. I realize its a terrible photo in general but I was hoping to catch the issue, which I did. The lens is fine until 100/130mm-ish range, and it seems the effect is more dramatic the farther away the subject is, and the faster they're moving. This photo was shot at 1/800s, f/2.8, iso 125, IS off, at 200mm. It's still possible that it needs to be micro adjusted to a finer and more accurate extent, but I honestly don't see ANY area of sharper focus in the newest example, which leads me to believe that may not be the problem..

I think that is a better example to judge. It looks like this is just a bad lens. I have never owned any of the 70-200 f2.8, but I had the same lens at f4.0 twice from Canon and once from Sony. I never had such blurry images. I did own an EF 50mm 1.4, which had a similar sharpness problem when shot wide open, but this particular lens was known for that issue. The 70-200 f2.8 on the other hand has a pretty good reputation. Any chance you can go to a store and compare your copy to another one?