Canon 200-400 f/4 reviewed by "Chasseur d'Images"

Started Jun 20, 2013 | Discussions thread
Steve Balcombe Forum Pro • Posts: 12,813
Re: Canon 200-400 f/4 reviewed by "Chasseur d'Images"

Jerry Fusselman wrote:

Steve Balcombe wrote:

Jerry Fusselman wrote:

Steve Balcombe wrote:

At f/32 diffraction will totally dominate. Any difference which is not seen at larger apertures is just a blip - a statistical fluke or a bad measurement.

Just as a general comment, it's hard to imagine that the internal TC would be any better than the 1.4x MkIII which is stunningly good on the MkII big whites. The point of the internal TC isn't better IQ, it's instant (and weatherproof) availability.

The results at f/32 make sense to me. As I explained two weeks ago in the "Strange thing about the built-in 1.4x teleconverter" thread, the internal 1.4x considerably reduces focal length when focusing close. It probably reduced focal length (compared to shooting at infinity) at whatever distance "C d'I" was using in their tests. The lens's chip must know about this, so to keep the aperture constant at relatively closer focusing distances, it has to close down the aperture more. If Canon did not do this, they should have (though perhaps not when shooting wide open). Smaller physical aperture size, more diffraction. No surprise there.

It would be easy to verify my hypothesis by stopping the lens down somewhat, engaging the internal 1.4x converter, and watching the size of the working aperture (from in front of the lens) shrink a bit while focusing closer. My hypothesis is that the degree of shrinkage when focusing closer is greater with the internal 1.4x than with an external 1.4x. If I am right, then the results at f/32 from C d'I are merely a consequence of their mistaking marked aperture numbers for actual aperture numbers. That is, with the external 1.4x, there was less diffraction at f/32 because the size of the aperture was larger. Another way to check is exposure levels in the images both ways.

That would make sense if it wasn't for the fact that no difference was observed at other f-stops - only at f/32.

The effect of diffraction is much more noticeable at f/32 than at f/22, so it is possible to insignificant at f/22 and significant at f/32. I which I could read the report easily.

Yes but the _difference_ wouldn't suddenly become more apparent at f/32.

I'm surprised you doubt the potential improvement of an internal 1.4x.

Perhaps you haven't seen how good the MkIII Extenders are.

It's a widely-held assumption that the 200-400's built-in extender must be better because it's specifically made for the lens. But the 1.4x MkIII is also optimised for a very small range of similar lenses, and in practice we know it to be phenomenally good on those lenses. Where is the scope for improvement?

Would you or someone else give me a citation of testing showing no degradation at all? That would amaze me, if true.

Same answer as below - the Chasseur d'Image review reports sharpness with the internal TC same as without, then with the external TC "no significant difference for most apertures, f/32 being the exception".

To be honest I don't actually believe that - and The Digital Picture's test shows a slight degradation but it is very slight and I see no significant difference between internal and external.

Think of the huge number of extra options the lens designers had about where to locate the extra elements.

And they did take advantage of this - the block diagram shows that they are not the rearmost elements. But I'll wager that was for practical reasons - the design allows the elements to be smaller so they can be switched in and out without the bulge becoming enormous.

I kinda want to accept this wager with you, if it was easy to do. The problem is, you might be too knowledgeable for a wager to be wise on my part.

LOL! No, I'm not.

I'm just looking at the anecdotal evidence - the image quality doesn't appear to be better in practice, and the elements are by far the smallest in the lens.

Anyway, perhaps you can tell me what is wrong with my thoughts on additional reasons for making it not rearmost:

  1. Higher image quality compared to an external multiplier due to more degrees of freedom for the lens designers.

As previously discussed.

2. Much closer minimum focus distance---I think I have verified this.

Has there been a discussion about this? If so I've missed it. But I would point out that a conventional TC gives the same MFD as no TC, so how could an internal TC improve on that?

3. Better in combination with an additional teleconverter.

Don't know. I remember reading a claim, years ago, probably on this forum, that two stacked (external) 1.4x TCs can give better IQ than a 2x TC but I would have no idea where to find the thread now.

Elsewhere, perhaps two weeks ago, I read that the internal converter is better in several ways than an external,

I'd be interested to see that.

so I think that the jury on point 1 is still out. Besides, image quality is more than just sharpness, and the OP only summarized sharpness for the internal-vs-external comparison. Other areas to compare include bokeh, CA, etc., as well as autofocus speed and accuracy.

Agreed, but I think we will have to wait for more comprehensive reviews to learn that level of detail - if indeed anybody ever goes to those lengths. When stuff is as good as this, any slight differences become academic.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow