Diffraction Limit

Started 11 months ago | Discussions thread
Detail Man
Forum ProPosts: 14,969
Like?
Re: I've stopped worrying ...
In reply to Anders W, 11 months ago

Anders W wrote:

Detail Man wrote:

Anders W wrote:

Detail Man wrote:

LTZ470 wrote:

tinternaut wrote:

Really, I don't care. In the highly unlikely event I need to shoot at f22, I'll simply accept the hit I take in terms of diffraction. As it stands, I really need to go beyond f12 (anything beyond that is probably because I'm too lazy to buy an ND filter), and the effects are minimal enough for me.

BINGO!

Correct, completely irrelevant as the amount of photos shot above f/4 on a FZ200 are probably less than 1% of all the FZ200's that have been manufactured...

Don't know of anyone who prints larger than 8X10 from a Superzoom...do you?

Don't know of anyone who looks at a Superzoom photo at 100% for viewing purposes...except when PP'ing a photo in LR that would be the only time for me...

Same goes for Oly 75mm f/1.8 and folks shooting it above f/5.6...maybe 1% of all the photos? even less?

And the amount of folks that bought an Oly 75mm f/1.8 to shoot at f/8?...lol...thats the real joke...less than .0000000000000000000000000000000000000000000001%

"Hey man I'm ordering me a Oly 75mm"..."Cool, your going to really love it, mine is sharp at f/1.8"..."Yeah, thats awesome but I buying mine for landscapes to shoot at f/8"...LMBO...

So if you are printing above 8X10 with a Superzoom then diffraction softening/details will be a concern...otherwise we can all go back to sleep...

I actually thought we were talking about super critical issues concerning the end of the world as we know it...lol...

Irrelevance at it's best...

Perhaps we have been chasing a mythical dragon ? "Diffraction Limited" meaning one specific thing, the F-Number at which lens-system diffraction actually causes "real world problems" appears to be a bit higher than the kids (as well as their supervisors) "typing from scripts" at the known to be limited Panasonic Technical Support offices located at a phone-bank in Virginia cited.

So DM, what would be the corresponding f-stop if instead we were talking about a 16 MP MFT sensor (e.g., the E-M5) ...

Assigning a value of 700 nM for Wavelength, and 3.73 Micron Photosite Pitch of the Oly E-M5:

( 4 * 3.73 E-6 ) / (700 E-9) = Nmax = 21.31

From the already referenced text here: http://www.dpreview.com/forums/post/51858399

Assuming an optical low-pass ("AA") filter yielding the first zero magnitude response at 2/3 of the spatial sampling frequency, that number reduces to F=14.21. In the sometimes speculated case of no optical low-pass ("AA") filter whatsoever existing in the E-M5 image-sensor, that number further reduces to F=10.66.

Thanks!

Sure, for what it is worth, anyway. I assumed an adjustement factor of 2 (for 2x2 Bayer) - but it is my impression that most fancy de-modaicing algorithms interpolate wider than that (with some particular weighting-factors), up to the 4x4 periodicity of Bayer arrays.

and how does the result correlate with what your eyes see in the test series I link to below (please click on "view original"):

http://www.dpreview.com/forums/post/52052886

F=16 looks "pretty good" to my (imperfect and inherently subjective) eyeballs. F=22 does not.

My way of verbalizing it based on my own perception would be that f/8 is fine (and rather hard to tell apart from f/4 and f/5.6, f/4 most likely being the point where the lens peaks based on the MTF-50 criterion). F/11 is perceptibly worse, yet decent and something I might still consider using in emergency. F/16 is poor and f/22 pretty useless.

Yes, peering a bit closer at F=11, it surely falls short of F=8.0. Sound like F=11.314 wins the day. (If) we were to exptrapolte that (aesthetic value judgment) of F-Number downwards from F=11.314 (normalizing for the FZ200's 1/2.3" form-factor image-sensor size), I get a maximum of F=4.012. Such "numero-aesthetic extrapolations" might just warm the heart of our visually empirical friend LTZ470...

Your use of Lightroom meaningfully resonates only with fellow Abode-ites who live and breath those same (silent, proprietary) processes. As you know, "sharpening" can have the effects of "pumping" the output MTF response (a bit). As you also know, I am not a big fan of Adobe "sharpening" tools.

As I think you are aware though, sharpening comes at a cost even if you use better tools than LR's. I'll be happy to send you the RAWs via mail or PM so that you can do your own processing and post your results.

I'll be happy to send you the RAWs via mail or PM so that you can do your own processing and post your results.

That's OK. I do have DxO Optics Pro 7.51 (a 64-bit sweathog that allegedly will not choke my WinXP 32-bit OS). It supports the E-M5, GH5, and the LX7. At some point, I may back up my present partition and give it a whirl. So far, I cannot really get too excited about doing so, though.

I proposed on this thread using DxO Optics Pro to process LTZ470's "Blue Angel air-show" ORFs that I have (with accompanying OOC JPGs) - but he is a JPG-shooter only, and why try to improve upon "Olympus perfection" ? There appears to exist very low (or no) interest on that score.

.

I calculate the "critical" (minumum) Photosite Aperture dimension to be:

Pa = ( Pa / ( K * Pp ) ) * ( Fz ) * ( W * N )

where:

Pa is Photosite Aperture;

K is the Bayer-arrayed and de-mosaiced "fudge factor";

Pp is Photosite Pitch;

Fz is the fraction of the spatial sampling frequency (which is equal to the reciprocal of Photosite Pitch) at which the first zero magnitude response occurs in the composite Optical ("AA") Filter combined (convolved in the spatial domain, multiplied in the spatial frequency domain) with the Photosite Aperture);

W is the Wavelength;

N is the F-Ratio of the lens-system.

.

Solving for the simple case of a 100% Fill Factor (Photosite Aperture equals Photosite Pitch), setting the value of K to conservative value equal to 2, and setting the value of Fz to 1/2 (the strongest possible "AA Filter" (resulting in a zero magnitude response at the Nyquist spatial frequency), the identity presented above simplifies to the following form:

Pa = ( W * N ) / 4

Re-arranging to solve for the maximum F-Ratio (N) as a function of Wavelength (W) and Photosite Aperture (Pa):

Nmax = ( 4 * Pa ) / W

http://www.dpreview.com/forums/post/51858399

.

For the DMC-FZ200, Pa ~ 1.5 Microns. For a worst-case Wavelength (W) of 700 nM, it appears that (in the base case), diffraction "extinction" is not an issue until F=8.571 (which exists, in fact, above the maximum F-Number adjustment value of F=8.0 for the FZ200).

The above case being for an optical low-pass ("AA") filter yielding a zero response at the Nyquist (1/2 of the spatial) frequency itself, a more likely situation is one where the optical low-pass ("AA") filter yields a zero resonse at (around) 2/3 of the spatial sampling frequency. In that case, the result of the above calculations being applied result in a maximum value equal to F=5.714.

Probably little reason to "sweat it" about F-Numbers lower than F=6.3. It might seem that a "stuperzoom" lens-system that is constricted by design to maintain a constant F-Number of F=2.8 might (perhaps) suffer from opticall aberrations that are not (in any event) very effectively controlled - especially as the Focal Length increases to longer than "wide-angle".

Thus, the down-side of the FZ200's around 1.5 Micron sized Photosites would seem to be that the F-Number of the camera system probably should not be raised above around F=5.6 in attempts to mitigate the various optical aberrations which just may be present in such an interesting, yet likely somewhat ambitious, lens-system design.

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Yep.New
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow