1. Reminder: Please user our affiliate links to get to your favorite stores for holiday shopping!

Diffraction and Post Processing

Discussion in 'Open Discussion' started by Clint, Dec 7, 2013.

  1. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    There have been a lot of comments about how diffraction will, for a simplified statement, destroy the sharpness of a photo.

    While diffraction does effect a final photo, poor post processing can easily do more harm to a photo than diffraction and appropriate processing will bring out the best of what the captured pixels have to offer.

    I have posted two images, both shot with the EM-1, Panasonic Lumix G Vario 7-14mm f/4.0 lens at 14mm, ISO 200, 1/80, and f/11 and one image was shot less than 1 second later than the other. The Olympus E-M1_PB091162.jpg photo has had appropriate post processing while the Olympus E-M1_PB091163.jpg photo has not. Note that according DXO Mark test, these images were shot in what DXO Mark has defined as red zones in their charts, which many are making the assumptions that the photos would be garbage due to diffraction.

    Olympus_E-M1_PB091162. Olympus_E-M1_PB091163.

    There are also two 100% crops posted. One crop is from the image that was appropriately post processed, Olympus E-M1_PB091162.jpg, and the other crop is from the same photo but with the same processing steps as from the Olympus E-M1_PB091163.jpg.

    Olympus_E-M1_PB091162crop. Olympus_E-M1_PB091162pp_crop.

    It should be pretty obvious that I could easily make a sound argument on how f/11 destroyed the photos. But it would be an unsound argument , the proof is in the Olympus E-M1_PB091162.jpg and Olympus E-M1_PB091162crop.jpg photos.
     
    • Like Like x 6
  2. jeffg53

    jeffg53 Mu-43 Veteran

    270
    Aug 22, 2012
    Sydney, Australia
    Jeff Grant
    Would you mind sharing what post processing you did? Why did you take a second shot? Why not process the first twice?
     
  3. mattia

    mattia Mu-43 Hall of Famer

    May 3, 2012
    The Netherlands
    Post-processing can make or break a photo, but I still wonder whether F11 is appropriate for a shot like this. Focus about 2 feet away from the lens at f5.6 at 7mm, and pretty much everything (from 1ft, which is not far off the minimum focus distance, to infinity) will be in focus. At f7.1 and a hyper focal distance of a hair over 1.5 feet it's everything between 8 inches from the lens to infinity. F11 gains you 2 inches of in focus foreground, no more than that.
     
  4. denniscloutier

    denniscloutier Mu-43 Veteran

    204
    Dec 24, 2011
    I regularly use f16 when I'm shooting macro. There will be some softening due to diffraction, but on the other hand at f8 I can't get a decent amount of the subject in focus so it is a trade off I'm willing to make.
     
  5. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    Jeff,
    Post Processing Changes, I could go back to Lightroom and look and see if I can find all of the adjustments and report them. But there are changes such as tone curves and others that would be hard to duplicate, and the adjustments for this image are very specific to the image. So the actual settings will have little benefit to anyone, even me. Would you stil like them?

    Why did I take a second shot? Basically it was from getting to know the camera and lens. In this case I was using Digital ESP and wanted to see if a focus point change would affect the exposure. So I moved the focus point several target points over and reshot - after reviewing the photos I saw there no exposure change.

    Why not process the first twice? - The detail images did come from the same photo, one was done with few changed settings and the other was done with the final changes. I did not reprocess the first image twice as it had already been finalized before this post, and I was to lazy to reset it since I had an almost identical image shot within 1 second of the first. So I used the two separate images as the examples.
     
  6. jeffg53

    jeffg53 Mu-43 Veteran

    270
    Aug 22, 2012
    Sydney, Australia
    Jeff Grant
    Thanks Clint, I'm not after all the changes that you made. I don't use LR anyway, What I really wanted to know is how you could take an image showing diffraction and the make it disappear as you appear to have done. That's why I asked why not do it on the same frame, but now I understand that you did.

    I am coming to grips with diffraction having recently arrived from the world of medium format and f/16
     
  7. hkpzee

    hkpzee Mu-43 All-Pro

    Sep 5, 2011
    Hong Kong
    Patrick
    I would definitely agree with your argument about diffraction and post-processing. I just don't think the sample you used is supportive of your argument. The shot without post-processing is pretty much what I expect from the setting, with diffraction reducing the sharpness somewhat. However, too much sharpening has been applied to the post-processed shot, such that there is halo all over the image. Just look at the edge of the sail on the boat you were on. If you look at the 100% crop of the post-processed image, the lines are all jagged. These samples seem to suggest that there is only so much you can do to salvage the softness resulting from diffraction!
     
  8. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    mattia

    First, depth of field is the area that is consider acceptably sharp for a certain size enlargement, typically 8 x 10. Acceptably sharp is a very subjective statement so the calculated DoF settings can become quite controversial. However if I was to accept the calculators acceptably sharp definition, I was only making an 8 x 10 print, and changing f-stops only changed the depth of field of the photo (within the acceptably sharp subjective definition), then you would be correct. But f-stop changes affect other lens factors as well and printing a photo larger than an 8 x 10 technically would require different settings than most DoF calculators are designed for.

    Long response

    I like to know and understand what my equipment is capable of throughout the range of settings. So I shoot with new gear throughout the f-stop and focal range until I think I know the strong points and weakness. These photos were part of that process. And then more experince with the equipment will proviue me more insight.

    There are many things that change throughout a lens at various settings and they do not follow simple curves of getting better or worse as the setting change in one direction or another. While one item may get better there may be something that offsets that change or even accentuates the change.

    Some lenses I have used have acutance shaped similar to a Mexican sombrero at some f-stops but not at others. That would be very sharp in the center, moving out to a softer points and then back again to very sharp and back again to soft. Other lenses can exhibit focus curvatures at some f-stops which might be used to keep subjects in focus despite the different distances from the focal plane. Some lenses distortion will become extremely negligible by one or two stops down but almost immediately and significantly increase at the next stop closed down.

    Some things I've learned about the 7-14 - at 14mm -
    • CA is consistent at all f-stops for each focal length,
    • from f/8 - f/22 vignette is minimized at all focal lengths and f/11 may provide the minimum amount
    • the lens has an almost defined center circle of very decent acutance at 14mm and f/5.6, with a quick transition to softness that is pretty close to the same all the way to the edges,
    • at 11-12mm distortion is minimized, f-stops have little effect on distortion
    • f/11 is probably the highest f-stop to use and still maintain accutance at a fair-good level for a print up to 24" (across all focal lengths)
    • the lens is sharper at f/11 than most lab test would have one believe, very fine details or light contrast areas are affected, but still useable
    • f22 produces a soft image from edge to edge that may give or enhance a sense of fantasy, fairy tale, or high key photo

    It is things like these that I learn so I can take advantage of the best or worst the lens has to offer.
     
    • Like Like x 3
  9. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    Here is another crop. Apparently when I exported the previous crop I had the LR output sharpening checked and I had no intent of applying more sharpening. When I went back and looked I noticed there is CA around the sail, which was enhanced when I post processed the photo, and there was quite a bit of jpeg artifacting as well.
    Olympus_E-M1_PB091162crop3.
     
  10. hkpzee

    hkpzee Mu-43 All-Pro

    Sep 5, 2011
    Hong Kong
    Patrick
    Ah, that explains the jaggedness of the lines... Just curious, how much sharpening and clarity were applied in your PPed image?
     
  11. monchan

    monchan Mu-43 Veteran

    254
    Jul 7, 2012
    Tottori City, Japan
    Thanks Clint,

    Between this thread and mine on my landscape difficulties,
    It has been my biggest week of learning in a long time.
    Thanks for taking the time and start another thread.
    Cheers.
     
    • Like Like x 3
  12. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    That is, I think the bottom line to the issue of diffraction. At f/8, there is a touch of diffraction softening with most lenses, but it is hardly perceptible. In some cases it is even necessary to address lens aberrations adequately. Beyond f/8, best not to go there unless you need to do so to get the shot you want (eg, more DOF in a macro without focus stacking).

    I think the OP makes two good points: 1) Diffraction effects are relatively subtle until you get to extremely high f-numbers. 2) Sharpening can have more impact on apparent detail than many realize.
     
    • Like Like x 2
  13. hazwing

    hazwing Mu-43 All-Pro

    Nov 25, 2012
    Australia
  14. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    That's a key point. Crop factor figures into diffraction issues as well. Specifically, diffraction effects at f/16 in the 4/3 format are similar to diffraction effects at f/32 on full frame, etc.

    Sent from my SCH-I535 using Mu-43 mobile app
     
  15. T N Args

    T N Args Agent Photocateur

    Dec 3, 2013
    Adelaide, Australia
    call me Arg
    Your posts are helpful Clint, cheers mate. The idea that we can "take advantage of the worst a lens has to offer" is refreshing, at least I think so.
     
  16. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    Hey, when you can't afford everything, find an means to use what you have to the best extent possible. And gee, even after you can afford anything, how you gonna get those unique shots of you don't know your equipment. There is no perfect lenses nor camera.
     
    • Like Like x 1
  17. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    Actually most of the image enhancement was not from sharpening. I could take this image into Photoshop and with several sharpening techniques enhance the perceived sharpness even more.

    The settings asked for
    Clarity, +12
    Sharpening, 61, Radius .9, detail 6, masking 76

    Most of the benefits come from slight adjustments in many areas, contrast, highlights, whites, blacks, tone curves, hue, saturation, luminance, and then the final one Remove Chromatic Aberration which was really the toughest to get adjusted - and actually if this was going to be a print for customer, I'd go back and tweak a little more.
     
  18. Clint

    Clint Mu-43 All-Pro

    Apr 22, 2013
    San Diego area, CA
    Clint
    monchan,

    Thanks and I have to congratulate you on posting what you did with the questions that you did, and then coming back for more! I almost started another post titled "Be careful what you ask!" LOL

    You do have a good eye and can produce some nice work. Gordon's 11+ tips post was great and should be a lot of help, and a reminder for those of us that have been shooting many years.
     
    • Like Like x 1
  19. pisanoal

    pisanoal Mu-43 Regular

    63
    Oct 29, 2012
    Isnt that only true if both sensors have the same megapixel count? I thought diffraction had to do with the pixel size relative to the circle of confusion (or whatever its called). So that a detail of a certain radius is percieved as sharp only if it is smaller or very close to the size of a pixel on the sensor, and if you took a 16-mp 4/3 and a 16 mp FF, your individual pixel area on 4/3 would be half that as on FF (of course i dont know the diffraction equation off the top of my head, and dont really want to look it up at the moment). The conclusion I got from this is that if you pack pixels into a sensor, you are going to hit diffraction earlier. Of course the trade off is less pixels = less possible detail, but it seems to me that there would be a maximum resolution for a given sensor size and anything past that you would get no or deminishing results (at least for a given f-number. I guess thats a bit off topic, but id be interested on what people who know more about this have to say on it.