Since the visible effects of diffraction (loss of resolution) is in part dependent on the individual's eyesight, size of image and distance from the image, the drop in resolution at f4 may be a drop to a resolution that is still better than other lenses comparatively speaking and is perfectly acceptable or even not noticeable to some viewers. I start to see diffraction effects at f8. I don't see the effects at f4 but this is not to say they are not there or that someone else can see the resolution dropping off. I am ok with diffraction at f4 if it hasn't helped drop resolution to unacceptable levels.
How do you know that you do not see diffraction at f/4? You cannot magically remove it and compare.
You and Bob are now being argumentative (on different tangents). Is there not a difference between diffraction being present and diffraction being visibly detrimental to an image? I should add to the list above that diffraction depends on a person's pre-bias in and out of ridiculous online debates.
Bob says it doesn't matter and wouldn't answer questions about the Lenstip MTF charts but I drew my f4 comment from my direct study of images from the 5D3/24-70 II (which btw seem to differ from the MTF charts of the 24-70 II at Lenstip) as well as the study of many other camera/lens combos over the years.
At the link in my signature, there are many real-world RAW comparisons, one of which is a landscape shot with the 5D3/24-70 II combo @ 24mm and every full aperture starting wide open to f22. My observation being that at 100% view in center frame, f2.8 is as sharp as the lens gets. f4 and f5.6 were indistinguishable from f2.8 to my eye. By f8, there was a slight softening, and by f11, diffraction was clearly approaching detrimental levels, IMO of course. Nobody is saying that diffraction isn't present from the get-go or at f4 but nobody but geeks care about that.
What folks want to know is when and where does diffraction become a problem for their particular applications. I recommend that they use their eyes instead of charts.
Here's where I got some of my initial information about diffraction:
http://www.dpreview.com/forums/post/22067611
Mostly math related.
Lucky for me, Mike actually calculated the smallest desirable aperture (I am sure you will nitpick my description of 'desirable') as 12.64 which I have rounded up to f13. So, you will find an f13 shot in every comparison I have done since then.
Here's an interesting article posted via CR.
https://fstoppers.com/education/5ds-f11-and-confusing-circles-68177
What do you think about the info at these two links? (Not a trap.)
In the graph below, without diffraction, the curve would go straight up. It has zero slope at f/4 because of the diffraction. Actually, if you model diffraction and aberrations, then add them according to the "Kodak formula", you would get that at f/4 diffraction and aberrations have equal strength in this example; not only equal derivatives as functions of aperture.
This techno-babble is fine but mind numbing to most of us. And misused, can send newbies down incorrect paths. You are an accomplished aliasing detective. What do your eyes tell you about resolution? (Rhetorical)
Can you really detect the difference in a few lines of resolution by eyesight? (Context: your question above.)
--
Formerly known as Just Another Canon Shooter
--
Rick Knepper, photographer, shooting for pleasure. I use tools, not brands. It is better to have It and not need It than need It and not have It. Mystery Gardner: "Rick, you have a passion for photography but not a position. That's a good thing." Based on 2014 keepers, I shot the following percentages: 5D3=42%, D800=31%, 6D=25% & D3x=2%. Various RAW comparisons at Link below. Includes 5D3 vs D800E (new uploads), 5D3 vs. 6D, Zeiss lenses etc.
https://app.box.com/s/71w40ita6hrcfghojaie
