100-400 User opinons needed for faulty lens

Hi all,

Top image f5.6
bottom f9.5 (the sharpest for my copy)

I am going to return it for re-calibration as its still under warranty. Just wanted to see if anyone else has seen this.

MLU=ON
IR Remote trigger
camera on large beanbag with another on top (if you download the images and flip them you can see that the camera was perfectly stable)

These images are from top/left of frame but not at the very edge.
Notice the highlights around the window are defined quads in the f5.6 image.
The leaves have a definite ghost which I'm sure is not cam shake and its there on both images tho its a bit better at f9.5 due to longer DoF I guess.

Cheers

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

Did you have IS turned off for these?

Yes, sorry, forgot to say that IS was indeed OFF and I locked focus with AF then turned it to MF to take the shots so both images are using EXACTLY the same focus.

Cheer,
ED

Then obviously there's something very wrong. Send it to Canon for fixing and/or calibration. I've never seen such a thing with either of mine, other than when I mounted it on a tripod and left the IS on.

That's unacceptable. Let me know how it is after you get it back.

Both show f5.6 are these crops ?

Ah, just figured out why... they both show 5.6 because I loaded the 9.5 into the 5.6 one to do a transparent comparison in Photoshop. I then had to resize as they were too big to upload so I cropped them together and saved each individually. So they were both using the 5.6 EXIF data.

Yes they are 100% crops with no post processing. Shot with 350D and internal settings to default. ie sharpness, colour, etc

Ed

What was the shutter speeds, because to me it looks very clearly like you had some camera shake going on. Look at the leaves in the first picture. On some you can see doubles as if the camera shook while the shutter was open. I can also see it in the second shot.

The first at f5.6 is 1/1000s
the second at f9.5 is 1/350s

I thought it was cam shake when I first got the lens but as I said in the first post, its not on a tripod but securely supported with no chance of wobble. There is no way it could have produced such shake at 1/1000s.

- Also notice that the direction of the "shake" is the same in both images (up and right). I think it would be highly unlikely to get the same direction of blur on two consecutive shots. The only thing I can think of that would have repeatable shake would be the mirror and that was not a factor as MLU was ON.

- and if you look at the highlights on the windowsil they are not blured but are in a quad pattern.

Can you post a resized version of the original image, uncropped?

Here is the 5.6 full frame resized.
I don't have the 9.5 full frame

Ed

HOSTED PHOTO
please log in to view hosted photos in full size.

The full frame picture looks fine? Looks like the focus has locked onto the car. See no problem if thats the case.

The "ghosting effect in the first is because of the 2 imges overlayed yes ? the second I guess is just blur because its focused on the numberplate and at f5.6 400mm at that distance you have a shallow dof . how does the numberplate look at 100%.

Yes, the focus was in fact on the license plate. The problem is not so much the focus but the the aberrations which I think are making the image softer than it should be.

No, both images are crops as they came out of the camera

ED

The focus is on the first "O" and it is quite soft compared to other crops I've seen.

I think the softness is related to the aberations seen in the first post. It does get better at higher F-stops but I've seen crops that are sharper at f5.6 than mine at f9.5

This is the centre of the f5.6 image. Unfortunately I don't have the full frame f9.5 to compare. Its was a bit sharper but not alot.

Ed

HOSTED PHOTO
please log in to view hosted photos in full size.