m43 vs APS-C vs FFrame: The issue of diffraction?

Discussion in 'This or That? (MFT only)' started by gdourado, Jul 18, 2012.

  1. gdourado

    gdourado Mu-43 Regular

    117
    Feb 23, 2012
    Lisbon - Portugal
    Hello,

    How are you?
    Today I was reading a blog about photography and then I started reading about diffraction.
    From what I understood, it is a phenomenon that makes the images softer and it is related with the lens build, aperture, sensor size and pixel size.

    From what I understood, in m43 cameras using the 12 mpx sensors, the pixels have a size of approximately 4.3µm, which makes diffraction an issue at apertures narrower than f7.1...

    The 16 mpx sensors apear to have a pixel size of 3.7µm and so, diffraction may appear at apertures narrower than f6.3...

    Now, I am not an expert in physics or electronics, but from what I understood, this issue is in direct relation to pixel size...

    If we currently have APS-C sensors with 24 mpx and FF sensors with 36 mpx, the pixel sizes have to be pretty small on those sensors, so diffraction would also be an issue at relative open apertures...

    It seems, altough, that the lens design of Full Frame lenses that are then used also on APS-C sensors, make diffraction less of an issue on those formats...

    This all got my head spinning pretty fast today, and since I cannot really make much of all this, I started this thread to hear your thoughts on this...

    Have you ever had diffraction be an issue on your images with m43?
    Are APS-C and FF system less prone to diffraction?
    What are your general thoughts on this?

    Hope to hear from you.
    Cheers!
     
    • Like Like x 1
  2. Hikari

    Hikari Mu-43 All-Pro

    Nov 26, 2010
    The problem of diffraction has nothing to do with pixel size, but format size. Format defines the smallest aperture before diffraction impacts the permissible circle of confusion. This is why as format increases, the smallest aperture that can be used gets smaller.
     
  3. silversx80

    silversx80 Mu-43 Veteran

    229
    Apr 27, 2012
    North Carolina
    Diffraction is only going to be a concern under significant cropping, or 100% viewing. Even then, it's not going to be hugely noticeable unless you're using apertures of f/16 or smaller.

    If I take a photo from the 4/3 format, which I framed well enough that I only need to crop the ends to fit at 16X20 print, I won't be able to determine if a 16MP sensor, 12MP sensor, or even the 8MP sensor was used.
     
  4. meyerweb

    meyerweb Mu-43 Hall of Famer

    Sep 5, 2011
    • Like Like x 4
  5. riverr02

    riverr02 Mu-43 Veteran

    258
    May 2, 2011
    New York
    Rafael
    Informative, well-written article- thanks for sharing it.

    R
     
  6. HarryS

    HarryS Mu-43 Top Veteran

    921
    Jun 23, 2012
    Midwest, USA
    Just understand when diffraction sets in, which is F16 for m43, and also realize that there are times you wanna go smaller aperture and live with the good things it can give you, like foreground/background in focus.
     
  7. NJH

    NJH Mu-43 Regular

    164
    Mar 8, 2012
    South West England
    Fundamentally this is Physics which places a hard limit on what is achievable. Some pretty whacky blue sky research out there to try and beat diffraction but it is still today the fundamental limit on what is achievable.
     
  8. NJH

    NJH Mu-43 Regular

    164
    Mar 8, 2012
    South West England
    For 16 Mp 4/3 sized sensor the Cambridge in Colour calculator says F8. This is using the advanced calculator, circle of confusion set to use pixels, 20/20 vision.
     
  9. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    I think that article is the source of much confusion. People skim the article, use the calculator, and jump to false conclusions. As well, this part of the article is incorrect:

    The one with smaller pixels will still have more resolution unless you stop down well past the point when diffraction begins to affect resolution, because diffraction isn't a hard limit.

    Here's an explanation I wrote up with some simulated images for demonstration purposes: What Does 'Diffraction-Limited' Mean?
     
    • Like Like x 1
  10. cmpatti

    cmpatti Mu-43 Veteran

    263
    May 8, 2011
    Berkeley, CA
    Reading these discussions there often seems to be a misimpression that once "diffraction sets in" or the "diffraction limit" of the lens is reached by stopping down, the lens becomes unusable. In fact, all it means is that there is some degradation due to defraction at smaller apertures compared to the lens's optimum aperture. For MFT, the optimum aperture of most lenses seems to be in the f/4-f/5.6 range, but I've seen some as high as f/8. That degradation may be minor or may be significant. The lens may be sharper stopped down beyond its "defraction limit" than it is when shot wide open (although the reduction in sharpness will generally be across the entire image when stopping down as opposed to mostly in the corners when open). Diffraction may be effectively irrelevant if you need the smaller aperture to get adequate depth of field for the image that you want: in that case you just have to live with whatever diffraction you get.

    All of this can be seen rather well on SLRGear's "blur index" widget, which allows you to graphically observe lens sharpness change at various apertures. If you play with the widget on a number of MTF lenses, you will see how lens sharpness tends to increase to a point as the lens is stopped down then decrease after it is stopped down further and get a sense of where the "sweet spot" typically is.
     
  11. ~tc~

    ~tc~ Mu-43 Hall of Famer

    Oct 22, 2010
    Houston, TX