You have to login first in order to be able to post messages.
Click here to login or register if you are a new user.

Diffraction on smaller aperture values

Showing posts 101 - 120 of 130
Page:
< 1 2 3 4 5 
6
 7 >
 of 7
Androniques
Alvera, at which aperture did you focus initially? Judging by your posts, I am sure you took all the measures to remove the possibility of any slightest shake of the camera. I am just trying to make sense of Parkinsonsniper's remark about the variation of the focus upon altering the aperture (in his case). You both have Nikon cameras+lenses, so I thought there could be the same trend (then in my case too, as D5000 is basically a simplified D90)... but it depends on what he meant.

Hambagahle, could you comment on how or where did you get the chart (seems you calculated it yourself)?

I compared the f-values corresponding to the (green, 550nm) airy disk of 1-pixel dimensions of D90 from the chart (f/4.0) and from this article (f/8). There is an obvious factor of 2 (2 stops is a lot!). Well, I guess it is explained by this quotation from that article: "airy disk will appear narrower than its specified diameter (since this is defined by where it reaches its first minimum instead of by the visible inner bright region)" It is though hard to believe that the visual perception "reduces" the diameter that much. In any case, the first minimum around the brightest spot is not the best practical estimate for the airy disk size in terms of diffraction effects apparent to the human eye (or camera sensor). Because two partially overlapping disks can be still resolved by their peaks, which can be further enhanced by sharpening algorithms. So, I would trust the apparent smaller airy disks as visually perceived by eye (for ex., one could define the disk size by a "more practical" metric of the diameter where the brightness is half of the peak value).

update: in fact, one should not bother about 1-pixel spots, but rather the ability to resolve two airy disks separated by at least one darker pixel... so, we again end up with a more optimistic estimate of limiting resolution, i.e. airy disks 2 pixels in size.
photos), Canon PS-A610 (rarely, very good at macro with ada...
Edited: 08/15/2012, 15:24:19 PM
Alvera

Originally posted by Androniques:
Quoted Message: Alvera, at which aperture did you focus initially?

The focusing system always focus with the biggest aperture.
This was my work flow: wake up in the morning - go to bath... ups! sorry, let's skip some steps... set f1.8, go to live view, manual focus with the live view (no chance to mistake with this with the zoom function, hope your camera is similar), tethered shoot 16 times with aperture change from PC, no camera movement.

Please go to the very end of my post http://alveraphoto.blogspot.ro/2012/08/diffraction-on-smaller-aperture-values_14.html to see the MTF chart for this lens. At 1.8 is not so sharp. But who use 1.8 in stock? :)
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Edited: 08/15/2012, 16:21:57 PM
Alvera
Just found this: "The lens showed a slight amount of focus shift when stopping down (residual spherical aberration)." Booooooooo!
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/15/2012, 16:42:00 PM
Hambagahle
Ok, so I calculated the chart myself, and checked my calculated values with others I consider reliable (Wikipedia, Cambridge in Colour, Luminous Landscape). That does not mean there is no mistake, so bear that in mind and be a little sceptical.

Pixel sizes were calculated simplistically from sensor dimensions in mm and pixels (according to Nikon) for D300. Nikon states same specifications for D90. I believe the Cambridge in Color assumptions under "Other Technical Notes" in the section called "VISUAL EXAMPLE: APERTURE VS. PIXEL SIZE" apply to my calculations too.

Just compared some of my values with those at Cambridge in Color (CiC), did not find any discrepancies. The formula I am using is Diameter(Airy Disk) = 2.43932 * Aperture Diameter (mm) * Wavelength (nm). Wikipedia says this gives the diameter of the first minimum around the Airy disk, and represents 85% of the light energy at that point.

I do not understand why CiC allows you to enter generic camera information for the Airy disk calculation; changing the camera information does not change value CiC calculates, and anyway diameter of airy disk depends only on aperture diameter and wavelength of light. This is the reason a 1 stop change in aperture will double/halve the diameter of the airy disk.

I agree 1-pixel spots probably not relevant, just provides a small check on the 2-pixel number.

I could not find any authoritative information that clearly explains how large the airy disk should be to affect the image quality, but I'd be very happy to find some. The reason I have 2-pixel size on the chart is the discussion on CiC; that discussion seems to be less precise and definitive than other material on CiC. Perhaps that means it's more complicated than we know - perhaps because of what the anti-aliasing filter does?

I'm sending complete table of values to Alvera.

Regards

edit: ..diameter of airy disk... REPLACES ...aperture of airy disk...
Edited: 08/15/2012, 18:00:08 PM
Androniques
Alvera, LOL
By now I only wonder why vignetting is not even on your avatar... ;~)
photos), Canon PS-A610 (rarely, very good at macro with ada...
Posted: 08/15/2012, 17:51:37 PM
Alvera

Originally posted by Androniques:
Quoted Message: Alvera, LOLBy now I only wonder why vignetting is not even on your avatar... ;~)

You are right my friend, a King's avatar should never show vignetting!!! I will write to support right now (10 emails/second).

Email to DT support: "Unfortunately, my royal avatar you have displayed below my royal name were not accepted by my royal person. Reason: - Poor optical performance due to low lens quality, such as lens fringing, chromatic aberrations, uneven sharpness in focus area - in this case: vignetting. If you have questions about the refusal reasons please ask Andromantic."
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Edited: 08/15/2012, 18:10:33 PM
Androniques
OMG, no way! I don't want to become a reviewer, even for avatars! :( I only wondered about the technical origin of the unevenly darkened corners while the avatars of all other members are evenly vignetted... it could actually be some way of distinction, after all, I just wanted to know! In fact I like the effect, even though it may overshadow your title :)
photos), Canon PS-A610 (rarely, very good at macro with ada...
Edited: 08/15/2012, 18:47:28 PM
Androniques
Hambagahle, thanks for your detailed explanation! I think we ended up already overdoing, in all practical terms, the analysis of the topic... but it is always good to learn something new and get deeper understanding of the tools we are using (lenses/cameras).

As for the "Cambridge in Color", I think their spot-on-grid presentation merely shows the relative sizes: airy disk (for 550nm) vs pixels. It is still a rough presentation, but it allows one to visually "feel" what to expect at different apertures for each particular sensor. As is usual with waves, the problem is how to define the edge of the gradually vanishing wave packet - any definition appears arbitrary, and the ultimate judge is your eye! :)
photos), Canon PS-A610 (rarely, very good at macro with ada...
Edited: 08/15/2012, 19:15:13 PM
Alvera
A chart from Hambagahle. Thank you.
http://alveraphoto.blogspot.ro/2012/08/diameter-of-airy-disk.html

Andromantic, this kind of unevenly darkened corners are only for Dorian Kings. There are people, of course, who think that golden objects, like my crown, must be photographed in special condition to achieve an evenly darkened effect. Don't trust them, urban legends...
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/15/2012, 19:19:19 PM
Parkinsonsniper
Hi everyone :)

I'm back...with something close to the solution. I worked on the CameraRaw. There are 4 different sliders in the sharpening tab.

Amount: Describes the amount of sharpness applied to the image (Hold down Alt key to see the image in BW)
Radius: (Airy disk size) the size (in pixels) of the local contrast, that will be applied to the hard edges of the image (not to the surfaces) (Hold down Alt key to see the High Pass filter)
Detail: The size of the local contrast, that will be added to the surfaces. (Hold down Alt key to see the High Pass filter)
Masking: The layer mask that will be applied to the sharpening layer (hold down Alt key to see the mask) Masking generally masks out the "Detail" of the surfaces.

If everything is clear till now...

First see the image in 100% (and even 200%) zoom. Then increase the Amount to %100 (or anything higher). Now move the radius bar from 0,5 to 3,0 slowly. You will see that there is a sweet spot that the image have the best sharpness. Check the edges for white spread lines. (there must be no whites around the edges)

Now hold down the Alt key and move the Detail slider to get the best result. That Detail slider loves to create extreme noise artefacts on surfaces so use it wisely :)) then use desired masking value for the sky or sea or other areas that don't need sharpness.

I generally add 10-20% Luminance Noise Reduction.

I think this technique will decrease the effects of diffraction and create an artifact free, perfectly natural looking sharp image. Maybe I should create a blog about this :)

Try and see if it works for you :)
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Edited: 08/28/2012, 05:01:34 AM
Alvera
You rulz!
How about the Radius value, must be set for every image or is sensor related?
Thanks for the effort!
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/28/2012, 02:19:18 AM
Parkinsonsniper
Hey thanks...You rulz too, I checked all of your tests. I can't do the test as good as yours, so I'm glad I lost the files.

For your question; it must be set for every image...actually, I'm a sharpness fan, so I sometimes use different radius values for different areas of the image. Because it depends on the size of the object in the picture.

   Beautiful Female Face   

Let's take this image as an example (oh my girlfriend is beautiful :))

The radius value needed for the eyelashes and finger edges are not the same. But I generally sharpen the smallest or most important part of the image and leave the rest to the post-post-process! :)
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Posted: 08/28/2012, 02:59:58 AM
Parkinsonsniper

Originally posted by Alvera:
Quoted Message: Just found this: "The lens showed a slight amount of focus shift when stopping down (residual spherical aberration)." Booooooooo!


I think my lens have the same problem! Both my Tamron and Sigma showed focus shift when I stopped down. I told you that we will learn new things during this journey! :))
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Posted: 08/28/2012, 03:03:37 AM
Alvera

Originally posted by Parkinsonsniper:
Quoted Message:  Let`s take this image as an example (oh my girlfriend is beautiful :))

Wow! This is your girlfriend?!! Forget about airy disk, let's talk about she!!!
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/28/2012, 03:19:59 AM
Alvera

Originally posted by Parkinsonsniper:
Quoted Message: I think my lens have the same problem! Both my Tamron and Sigma showed focus shift when I stopped down. I told you that we will learn new things during this journey! :))

...and we will stop taking photos because of so many problems :))
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/28/2012, 03:22:14 AM
Parkinsonsniper
No no...try to see the airy disk and diffraction in her eyelashes! :P try to concentrate to the art of photography. Stop staring at her eyes...I said stop it! :)))
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Posted: 08/28/2012, 03:28:51 AM
Alvera

Originally posted by Parkinsonsniper:
Quoted Message: No no...try to see the airy disk and diffraction in her eyelashes! :P try to concentrate to the art of photography. Stop staring at her eyes...I said stop it! :)))

lol... you are remote on my PC?!
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/28/2012, 03:39:20 AM
Parkinsonsniper
No :)) I know the effects of the composition, light and contrast (now I'm a smart-ass :))) and we are out of the subject! Let's go back to the airy disk :))
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Posted: 08/28/2012, 03:45:42 AM
Alvera

Originally posted by Parkinsonsniper:
Quoted Message: No :)) I know the effects of the composition, light and contrast (now I`m a smart-ass :))) and we are out of the subject! Let`s go back to the airy disk :))

Ok, back :) I think that airy disk is sensor related. see here at Examples chapter: http://en.wikipedia.org/wiki/Airy_disk
Nikon. 4x4 car for wildlife adventures. Inkscape for vectors. Love & p...
Posted: 08/28/2012, 04:20:24 AM
Parkinsonsniper
...and did you try my sharpnening hints to kick the airy-disk's back? :)
f2.8 - Nikkor 50mm f1.4 - Nikkor 50mm f1.8 - Tamron 90mm f2....
Posted: 08/28/2012, 04:28:26 AM
Showing posts 101 - 120 of 130 Page: < 1 2 3 4 5 
6
 7 >
 of 7