fredmiranda.com

Moderated by: Fred Miranda

New fredmiranda.com Mobile Site

New Feature: SMS Notification alert

New Feature: Buy & Sell Watchlist

FM Forums | Canon Forum | Join Upload & Sell

Archive 2012 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #1 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

With its impressive specs and reviews, the 70-200 f/2.8 L II seems like it ought to be ideal for a task that comes up from time to time in my typical seascape work. Over the last year I have ordered it from B&H (great service) 3 times, returned 2 of them, and am trying out the third. All 3 lenses seemed to show the same problem. With the zoom at 200mm, focused at infinity (at least 1500 feet where I test), and with IS turned on (mode 1), at f/5.6, and shutter speed faster than 1/1000th, the rightmost 1/4 of the frame is soft, really soft. Center and left side are fine, excellent, really. So good that I very much want the lens to work as expected so I can keep it.

I have tried the test two ways. First way, I just pick a nicely textured and illuminated subject (a grassy bluff that rises on the far side of an inlet) that stretches across the frame, and shoot a single shot hand-held. Inspect. Repeat. All pictures come out the same, soft on the right.

Second way, I pick a prominent target in the middle of the frame (Boston Light, about a mile away across water, with still air and zero mirage), focus, shoot, and then turn off auto-focus and recompose and shoot twice more, to put the target first on one side of a frame, and then on the other. When the target is on the right side, it's, uselessly soft, when on the left, sharp�almost indistinguishable from when it's in the center (which is great, impressive).

A variation focuses separately for each frame, with the target in the middle, then recomposes to put the target first on one side to shoot, then again to put it on the other. Same result, sharp left, soft right.

A test with IS turned off, camera mounted on tripod, mirror lockup, cable release, live view focusing, shows the entire frame sharp, including the right side, although the right then appears to be minutely less sharp than the left. But basically, using the camera without IS seems to cure the problem.

Three new lenses, the first two ordered in succession about 10 months ago, and the last just delivered this week, have shown the same symptoms. I'm getting ready to return the third lens and give up, but thought maybe someone could help me out if I am missing something obvious, or making a mistake because I have never used IS except with this lens.

Any suggestions appreciated.

Oct 25, 2012 at 08:42 PM

lowa2
Offline
• • • •
Upload & Sell: On

p.1 #2 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

What about trying the lens on a different camera

Oct 25, 2012 at 09:07 PM

jcolwell
Offline
• • • • • • •
Upload & Sell: On

p.1 #3 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

Same camera? ...

The lens is wickedly sharp, wall to wall and corner to corner, at least mine is. Maybe, there's some asymmetry in your camera mount. Of course, this would be even more pronounced with a longer lens.

IME, the IS works fine on this lens. OTOH, maybe a "minutely less sharp" right-side image with no IS, can be associated with a "less sharp" right-side image with IS, i.e. maybe it's the camera mount.

Of course, you could send the camera and lens in for a check up. I normally wouldn't do this, as they're calibrated to independent criteria, but it could be worth a shot.

Oct 25, 2012 at 09:14 PM

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #4 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

Maybe should have mentioned that all other lenses work fine on the same camera, showing no sign of this problem. That includes the 135L using the 1.4 Extender II, for an effective focal length of 189mm�same ballpark. The camera is a 5D II with which I have years' of experience. The suggestion to try another body is a good one though, if only for the sake of being thorough.

Oct 25, 2012 at 09:57 PM

jerbear00
Offline
• • •
Upload & Sell: Off

p.1 #5 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

Bummer. No complaints with mine.

Oct 25, 2012 at 10:20 PM

skibum5
Offline
• • • • • •
Upload & Sell: Off

p.1 #6 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

splathrop wrote:
With its impressive specs and reviews, the 70-200 f/2.8 L II seems like it ought to be ideal for a task that comes up from time to time in my typical seascape work. Over the last year I have ordered it from B&H (great service) 3 times, returned 2 of them, and am trying out the third. All 3 lenses seemed to show the same problem. With the zoom at 200mm, focused at infinity (at least 1500 feet where I test), and with IS turned on (mode 1), at f/5.6, and shutter speed faster than 1/1000th, the rightmost 1/4 of...Show more →

maybe they sent you the same copy all three times

Oct 26, 2012 at 12:58 AM

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #7 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

jcolwell, thank you for your reply. Just so I can be sure we are doing apples-to-apples, are you sure your good experience includes images with these criteria:

200mm zoom setting;

Distance > 1,000 feet;

Shutter speed faster than 1/1000th;

Fine-grained, sharply-focused image detail throughout the rightmost 1/4 of the frame?

Another reason I ask is that after checking on the web I have found a few references to soft-on-the-right problems from other users, vaguely described. It occurs to me that my list of criteria might not be that commonly used in other peoples' normal shooting styles. If so, maybe a widespread problem might have escaped broad notice, especially with a lens which is otherwise so outstanding.

Oct 26, 2012 at 04:59 AM

RCicala
Offline
• • • •
Upload & Sell: Off

p.1 #8 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

You mention the problem occurs with IS turned on and at f/5.6. Is the problem still present with IS off and at f/2.8?

The reason I ask is that a decentered field-flattening element causes an odd problem: the lens looks fine at f/2.8, but when you stop down one side doesn't sharpen up while the rest of the lens does. If your problem isn't apparent at f/2.8 that could be an issue.

The second reason is an IS element is rather shaped like a field flattening element. I'm speculating here but I would guess a decentered IS might have the same effect.

The odds of seeing that 3 times in a row is rather amazing - you wouldn't happen to have a record of the SNs you received, would you? I think B&H usually puts the SNs on the receipt. If they were all similar SNs it might be a bad batch.

Oct 26, 2012 at 06:01 AM

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #9 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

More info. Tried the same tests using a borrowed 5D III. On that camera the IS does not seem to produce the major problem observed when I used it on the 5D II. Repeated tests all showed the right side of the frame much sharper than previously.

However, in the previous tests, when I turned IS off and shot from a tripod, there was nevertheless a difference (noticeable with 100% enlargement) between the left side of the frame, which was tack sharp (quite like the center), and the right side, which was, at best, acceptably sharp for most purposes. That residual difference is now seen in all the images taken with the 5D III with the IS on. In short, the 5D III with IS on is behaving quite like the 5D II did with IS off and shot from a tripod�which is not perfect in the case of this lens, but near perfect on the left and in the center, and good on the right.

Based on this experience, I suspect that on a 5D II body many or all 70-200 f/2.8 L II lenses may fail to focus properly across the frame with IS on when used at 200mm and when focusing at infinity. I haven't tested shorter focal lengths or nearer focus distances.

The 5D III seems to correct whatever is causing that problem (apparently IS-related) with the 5D II, but of course does not get rid of smaller problems that might be caused by decentering of some optical elements or from some other source that is not IS-related.

Oct 26, 2012 at 12:32 PM

skibum5
Offline
• • • • • •
Upload & Sell: Off

p.1 #10 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

splathrop wrote:
More info. Tried the same tests using a borrowed 5D III. On that camera the IS does not seem to produce the major problem observed when I used it on the 5D II. Repeated tests all showed the right side of the frame much sharper than previously.

However, in the previous tests, when I turned IS off and shot from a tripod, there was nevertheless a difference (noticeable with 100% enlargement) between the left side of the frame, which was tack sharp (quite like the center), and the right side, which was, at best, acceptably sharp for most purposes. That...Show more →

maybe the mount on your 5D2 is a trace tilted compard to the sensor plane and that combined with the lens having something a trace tilted added up to a problem?

Oct 26, 2012 at 01:53 PM

RCicala
Offline
• • • •
Upload & Sell: Off

p.1 #11 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

skibum5 wrote:
maybe the mount on your 5D2 is a trace tilted compard to the sensor plane and that combined with the lens having something a trace tilted added up to a problem?

+1

I hate for us to take an example from one 5DII and one 5D III and paint with such a broad brush that we extrapolate to all 5DII and all 5DIII.

It got me curious enough to grab a couple of 5DII and 70-200 IS II and try to repeat the experiment. Of course, I may have done it differently, but saw no difference right and left side with either pair, IS on or IS off.

It's an interesting problem, but I'm not sure it's going to involve all copies of a lens and camera - all these occurred with one copy of the 5D II. Occam's razor would suggest this copy of the 5D II rather than all 5D IIs.

Oct 26, 2012 at 02:14 PM

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #12 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

RCicala, thanks for your comment, and also your effort in trying the experiment. Apologize if I put you on a wild goose chase. You may well be right about the problem being just this camera body. I'm tending in that direction myself, for the reason you suggest.

What I can't explain, however, is why if the body is causing the problem, it doesn't show up at all when using other lenses, or why the problem largely disappears on that suspect body when you turn off the IS. Maybe I'm naive to suppose the IS runs independently in the lens, without feedback from the body. I assume you know if the IS system interacts with the body? I confess I don't even know whether its mechanism is based on optical feedback or gyros, or what. Never gave it any thought because I never before used a lens with IS.

Or maybe there is something about the way the IS operates that picks up on and amplifies some physical misalignment in a way that the other optics in the lens do not�thus the appearance and disappearance of the problem when the IS is activated/deactivated.

Oct 26, 2012 at 08:45 PM

RCicala
Offline
• • • •
Upload & Sell: Off

p.1 #13 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

Someone please correct me if I get some of this wrong. IS isn't my strength.

But the IS system is powered from the camera, but otherwise operates independently with it's own sensors and electromagnets, basically just reacting to movements of the lens. However, it is an optical element so if it's out of sorts it can affect the image.

From taking them apart, I can say the 'cage' of the element is plastic. There could be something that keeps it from centering properly. When IS is turned off, the element is locked in the center of it's cage. When on, it moves around, the idea being it moves to bend the light path, correcting for shake.

There's a nice diagram here: http://www.usa.canon.com/cusa/consumer/standard_display/Lens_Advantage_IS

The point is the stabilizer is not magic, it's an optical group refracting light like any other group. If it's decentered when turned on, it would have an effect like a decentered element.

My initial comment was based on an odd finding we've had with lenses that have a 'field flattening' element. Most decentered elements cause problems in all 4 corners. A decentered field flattening element causes one side of the image to be soft.

The elements in an IS unit are optically somewhat like field flattening elements, so my thoughts that it might be the problem.

Another thought, though, to finally get to my point: if your mount is slightly decentered in comparison to the image sensor, that might give the same effect as if the element was decentered.

One more bit of thinking out loud: a camera with a tilted mount will only show that with sharp, very wide-angle lenses. I don't know but it's possible an off center mount might only show with sharp telephoto lenses.

Just speculating out loud, of course. What is that old saying - "No man can speak day after day and refrain from saying more than he knows". I resemble that remark sometimes.

Roger

Oct 27, 2012 at 12:20 PM

splathrop
Offline
• • •
Upload & Sell: Off

p.1 #14 · Diagnosing 70-200 f/2.8 L II problem (same for 3 copies)

Roger, thanks for all the thoughtful comments. You have given me great input, but I remain baffled.

Nothing so far discussed really fits all the facts very well, but the evidence is as described, and reliably reproducible.

In summary, there has been a problem with using IS on 3 successive copies of the 70-200mm f/2.8 II when mounted on my 5D II. The most recent copy tried functions flawlessly with IS off (although slightly soft on the right compared to the left). Turning IS on improves the left and center of the frame just as it should, and makes a mess on the right. That is a result similar to the one delivered by the previous two copies. All my other lenses (none of which uses IS) function flawlessly across the entire frame, supporting enlargements > 30-inches at a quality level entirely satisfactory for gallery sales. This includes even the 135 L when used with the 1.4 Extender II (although the edges of the image in that configuration do show the expected slight degradation, but symmetrically). It also includes the Zeiss 100mm MP, which reliably delivers the razor sharp performance across the frame that it is noted for.

In addition, mounting the zoom on a (borrowed) 5D III fixes the problem, with the IS functioning as it should, but leaving the slight right-side softness still evident.

Possible explanations have been advanced:

1. The 5D II has a decentered or misaligned mount. But why would other lenses go unaffected? Why would the 70-200 function properly with IS turned off?

2. The IS unit in the zoom is flawed. But why would the same symptoms have recurred in 3 separate copies?

3. There is a previously-missed incompatibility with the design of the zoom and the design of the 5D II, one that shows up only when using the zoom at maximum extension while focused at infinity on a subject that challenges resolution across the entire width of the frame. But tests by Roger using a different copy of the 5D II and a different zoom refute that.

By elimination, I suggest the only likely explanation left is that there is some insufficiency (but not failure) in the electrical circuit connecting camera to lens to power the IS system in the lens. The IS system is clearly operating; it can be heard working, and it creates the expected improvements across the left 2/3 of the image, while degrading the right 1/3.

Any thoughts? How could that even produce the peculiar symptoms seen without causing the entire image to degrade?

Oct 28, 2012 at 03:47 AM


FM Forums \| Canon Forum \| Join Upload & Sell