Diffraction. How real is the problem?

I am experimenting with Macro at the moment. I have been experimenting with flash, and to get a reasonable DoF, I stopped down to F45.

The Raw was soft, but the Diffraction Correction , Capture One, seems to clean up the softness.

Any opinions.

Here is the original

Enlarged

Original

Corrected

Yes, photo stacking is the optimum solution. I use Helicon and it is pure magic. It even seems to account for some movement.

My problem was when I shot a set in windy conditions. I was curious to see if using F45 might save the day. Whatever C1 does, it yields a usable image. I might experiment at even smaller apertures, to see how far I can go in difficult situations.

A popular review phrase is that diffraction "sets in" at some increasing f/number - as if it suddenly becomes visible instead of lesser visibility at values such as f/2.8.

A less well-known factor is the sensor's pixel-pitch. For example, my Sigma SD9 has a 9.12 um pitch and shows very little diffraction at f/11 or lower. On the other hand, my Lumix DC-G9 can show diffraction even at f/5.6.

At f/45 the Airy disk diameter is 61 um - a whopping total of 10 normal sensor 6um pixels! ... mucha convolución ...

f16 is fine with apsc. i could shoot f14 with my m43 gear back in the day.

i can only see difraction when im shooting spiders at 10 x and view the image at 200 % with my apsc sensor. the 5um a7iv sensor is very crisp using 10x objectives. nothing sharpening cant fix but does add noise to my a6700.

An interesting definition of "fine" when f/16's Airy Disk diameter covers about three pixels on most sensors, Donald.

ChatGPT says that lens can go up to f/51 when focused at its closest focusing distance (about 0.29 m) - so not much further than f/45 ...

Which begs the question about whether with F45 showing at the readout closest focus, where I have lost 2 stops, the diffraction effect is really F22, which the camera would read for the same setting at infinity.

Good question. After all, the minimum aperture has a mechanical limit, so maybe the diffraction comes down to the actual diaphragm diameter and it's real distance from the sensor - then use Rayleigh's Criterion to determine the actual diffraction at the sensor especially at macro shooting distances.

As to the Criterion and the Airy disk, a clue here:

"A similar result holds for a small sensor imaging a subject at infinity"

en.wikipedia.org/wiki/Angular_resolution

Without going into diffraction math, I lean toward the idea that your actual max. f/number is indeed 22 at infinity and that the minimum aperture diameter is 105/22 = about 5 mm - no matter what the focusing distance. On the other hand, with the conventional FL/F# formula, the diameter would be about 2 mm ... pretty small. I once had a true macro lens that extended for close focus meaning that the actual distance to the sensor doubled approximately at 1:1 magnification ... meaning double the Airy disk size.

I wonder if, when you look at the aperture from the front of your lens at minimum diameter, does it's size change much, if at all, over the full range of focus?

Looks good to me! Didn't know Capture One had diffraction correction. I know Canon's proprietary RAW converter does, though. In any case, while diffraction can be a serious issue when the aperture is that narrow, often of far greater concern is either the significantly longer exposure time, where even the slightest motion can destroy far more detail than diffraction, or, to a lesser extent, the greater noise by using a narrow aperture and shorter exposure time (greatly ameliorated by modern noise filtering software, though).

However, focus stacking is the gold standard for high resolution deep DOF photography, but, once again, motion blur is an issue, and I don't know how well the software corrects for that by aligning each exposure (in-camera focus stacking probably doesn't do any alignment). But, even with aligning in software, it would be dependent of all elements in the scene moving the same amount between exposures, which will not necessarily be the case.

Still, like I said, the f/45 you used in the above photo worked out great!

I'm still not certain that the selected f/45 was real, i.e. was based on Rayleigh's Criterion and the actual aperture diameter and it's real distance from the sensor ...

... hoping that Nigel will respond to my earlier post ...

Isn't it quite likely that the focal length changed at such magnification? That is, at 1:1, maybe the actual focal length is doubled (210mm) so f/45 would result in an aperture diameter of 5mm.

Yes, but the other way round. My thought was that an actual minimum aperture diameter of 5mm (105/22) at about 210mm (105mm + extension) could result in a calculated f/number being shown f/45 in the Nikon viewfinder or on the LCD.

In other words, the real minimum diameter is 105/22 = 4.773 mm - a hard mechanical limit. Focusing such that (105 + extension) = 210 mm gets you a calculated f/44 which Nikon would probably show as f/45. Focusing such that (105 + extension) = say 150 mm gets you a calculated f/31.43 which Nikon would probably show as f/32.