Diffraction limited performance

[trendland]

Diffraction is a significant effect for radio fading in and out at the limits of the range (what I implied but could have been clearer on).

So Nodda Duma (just if I understand correct what you stated here) - you also would descibe the procedure of defraction [you used the term "fade out - fade in"] like it is shown in the following graph :

You agree ? But then a kind of 3mm roule can't be much precise! Would you also agree?

with regards

 

[warden]

I'm more than happy to offer the attached:

Ralph, I have the second edition of your wonderful book. I don't know why I didn't crack it open when I started thinking about diffraction. Thanks.

 

[warden]

Here's a fairly straightforward discussion on diffraction:

https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

I enjoyed that quite a bit, thank you. The "what it looks like" tool is interesting. It's fun to play with the advanced functions of that calculator to learn at what size you would see negative effects from diffraction.

 

[warden]

I appreciate and enjoy the mathematics and equations, but for those that do not or are getting lost the hyperfocus sweet spot and defraction limits are roughly

Format ___ Sweet Spot___Defraction Limit
35mm _____ f/8 ______ f/16
120 ________ f/11 _____ f/22
4"x5" ______ f/16 _____ f/32
5"x7" ______ f/22 _____ f/45
8"x10" _____ f/32 _____ f/90

As a practical solution this makes sense to me. The diffraction calculator linked to earlier in the thread suggests that I wouldn't be able to see the negative effects of diffraction with the 60mm Hassy lens, maxed out at f/22, printed and viewed normally, unless of course I print quite large. But the DoF difference between f/16 and f/22 isn't that great anyway so I'll consider that too. Thanks.

 

[Sirius Glass]

As a practical solution this makes sense to me. The diffraction calculator linked to earlier in the thread suggests that I wouldn't be able to see the negative effects of diffraction with the 60mm Hassy lens, maxed out at f/22, printed and viewed normally, unless of course I print quite large. But the DoF difference between f/16 and f/22 isn't that great anyway so I'll consider that too. Thanks.

Most, probably almost all lens manufactures limit their lens designs to stay away from the defraction limits. I have found that Hasselblad stays well away from the defraction limits.

 

[Nodda Duma]

No. I’m not sure why you’re showing a log plot but that is not representative of diffraction-affected signal strength. Look at the cross-section of the magnitude of the Bessel function (or sinc function). It’s the classic Airy disc cross-section in one-dimension with all its side lobes. You can research online. I just googled “diffraction signal theory” (no quotes) and found a ton of info that can explain it far better than I can typing with one finger on my smart phone.

If you’ll allow me the indulgence of repeating myself: when stopped down, diffraction isn’t the only thing that comes into play for determining what the blur is. However, the effects can be and are accurately modeled with the tools available to a lens designer. All of the discussion of approximations are just that: approximations. The closest this thread got to how it’s really calculated was the Bessel discussion which is way too mathy for photographers (no offense...it just requires a foundational knowledge of very complex math to derive the solution which even physicists outside of electromagnetics would struggle with).

If you really want to learn to model diffraction accurately, pick up the book “Fourier Optics” by Goodman. Diffraction is fully describe in there without approximation.

So Nodda Duma (just if I understand correct what you stated here) - you also would descibe the procedure of defraction [you used the term "fade out - fade in"] like it is shown in the following graph :
View attachment 221321

You agree ? But then a kind of 3mm roule can't be much precise! Would you also agree?

with regards

So Nodda Duma (just if I understand correct what you stated here) - you also would descibe the procedure of defraction [you used the term "fade out - fade in"] like it is shown in the following graph :
View attachment 221321

You agree ? But then a kind of 3mm roule can't be much precise! Would you also agree?

with regards

 

[trendland]

[QUOTE="Nodda Duma, post: 2170355, member: 61906" Look at the cross-section of the magnitude of the Bessel function (or sinc function).

If you’ll allow me the indulgence of repeating myself: when stopped down, diffraction isn’t the only thing that comes into play for determining what the blur is. The closest this thread got to how it’s really calculated was the Bessel discussion which is way too mathy for photographers (no offense...it just requires a foundational knowledge of very complex math to derive the solution which even physicists outside of electromagnetics would struggle with

If you really want to learn to model diffraction accurately, pick up the book “Fourier Optics” by Goodman. Diffraction is fully describe in there without approximation.[/QUOTE]


Nice play to tempt TRENDLAND to ever pic up a book again in his life.....Nodda Duma....!

seriously - I noticed there is obviously a failure to the graph and that is obviously in regards of
failure of log - function = no log function! But what you stated Nodda Duma redirects back
to the Bessel function - indeed!

Thanks much for clearing up a bit Nodda Duma!

with regards

PS : Not the first time I've been told to read a book! NO - don't be afraid!
Seriously many years ago (decades) an ex Zeiss employer (I guess it was Zeiss) I worked with him
in a photo store offered my a book (to read in one or two weeks) it was from basics about we discussing here! But to my dreadfull certainty it was brimming with mathematical
equations I've never seen before - so he explained me some basics from time to time!
But I realy remember : I told him that I will not read his book because it was just Issue 1
(The issue for EASY GOING = more basically) I was very frightening about to have next
ISSUE 2 - 4 to read AND TO DISSCUSE WITH HIM.....guess he had been fired from Zeiss
because he was much much overqualified for that job we met as colleguages!
But OK Nodda Duma - I will have some look at the Bessel function again and then have a look
to the following ....but to come to math reputation was not intended since 1985 - therefore allways easy jobs without maths!

 

[Nodda Duma]

To be fair I haven’t cracked my own copy open in 20 years!

 

[trendland]

To be fair I haven’t cracked my own copy open in 20 years!

This is indeed a last hope for all those who have not earned their doctorate at M.I.T. - isn't it Nodda Duma ??? - thanks for stating!

with regards

PS : Or for anyone for whom

was just too expensive?

 

[MattKing]

To be fair I haven’t cracked my own copy open in 20 years!

But then, to be fair, the math involved certainly hasn't changed in longer than that!

 

[abruzzi]

The closest I get to Bessel functions are sidebands in frequency modulation....