E-PL2 F# settings for lanscape

Discussion in 'Olympus Cameras' started by klrman, Aug 29, 2011.

  1. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    I was playing around the other day with my E-PL2 in "A" mode on the dial and ran the F numbers (stops?) from 3.5 all the way to 22 to see if I noticed a difference compared to the iauto mode which settled on F9. I had my Camera on a tripod and F9 in iauto mode was the sharpest images always when shooting the same landscape scene. Is that normal? I though the higher the F number, the more balanced the detail would look everywhere. I did use center focus.
     
  2. Ned

    Ned Mu-43 Legend

    Jul 18, 2010
    Alberta, Canada
    If you go too high in your f-stops then diffraction will start to set in and soften your image. If your lens is an f/3.5 maximum aperture to begin with, then f/8 is a pretty common sweet spot for a Standard-Grade Zuiko lens. Fast Zuiko lenses are normally sharpest at about f/5.6 (high-grade) or f/4 (super high-grade).

    The general rule of thumb for Zuiko Digital lenses (I don't know how this compares with m.Zuiko) is about 2 stops down from wide open is the sharpest. This varies slightly from lens to lens. With many other lens manufacturers, you may find the sweet spot about 3 stops down from wide open.
     
    • Like Like x 5
  3. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    Thanks Ned, you're always reliable for a high quality answer, very much appreciated. Yes, I used the standard kit 14-42 kit lens. Good to know this. I'm afraid to ask, but what would a 14mm super high grade zuiko pancake lens go for and would the difference in detail be noticeable or just barely noticeable?
     
  4. nickthetasmaniac

    nickthetasmaniac Mu-43 All-Pro

    Jan 11, 2011
    There isn't a 14mm hi-grade pancake from Zuiko at the moment - there's the Panasonic 14/f2.5, which is very good, but probably not 'high-grade'.

    Then there's the new 12mm f2 from Olympus, which is excellent but not a pancake. It's listed on B&H for $800USD.
     
  5. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    Thanks! Do you think that lens would give a noticeable improvement in clarity and detail for an amateur like me?
     
  6. DekHog

    DekHog Mu-43 Top Veteran

    579
    May 3, 2011
    Scotland
    Not really anything you would see in your day to day photography unless you pixel peep - optically better, yes, but not to the point that you would look at the same image taken with both lenses and gush over the one taken with the prime. The beauty of them (for a lot of people) is the speed and physical size..... and the street cred.... :biggrin:
     
  7. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    Very good to know thanks DekHog. I like the size, but not spending $800 for lens right now that won't make me gush seems like a good idea at the moment :biggrin: I think I need to just work more for now on knowing my camera and getting some good captures. Thanks again.
     
  8. MajorMagee

    MajorMagee Mu-43 All-Pro

    Feb 24, 2011
    Dayton, OH
    Diffraction effects are controlled by the aperture f stop (that makes it independent of the focal length), the number of data pixels your saving, the sensor's individual detector cell dimensions, and the wavelength of light your concerned about. For the PEN at 4032 by 3024 this works out to be between f2.8 and f4 for green light. After that you need to reduce the recording dimensions (resolution) to avoid being able to detect diffraction effects in the image.

    For the Micro Four Thirds Sensor it has detector unit dimensions of:
    Width 17.3mm / 4032
    Height 13mm / 3024

    As the f stop goes up the size of the diffraction caused Airy Disc goes up so that you need to increase the unit dimension (divide by smaller recording resolution since the chip is a fixed size) to avoid having the information spread over more than one piece of recorded data.

    For Green Light At 550 nm
    _f___Airy_Disk__Acutance___Width____Height
    1.2_____1.6_____100.0%_____4032_____3024
    1.4_____1.9_____100.0%_____4032_____3024
    1.8_____2.4_____100.0%_____4032_____3024
    2.0_____2.7_____100.0%_____4032_____3024
    2.8_____3.7_____100.0%_____4032_____3024
    4.0_____5.3______81.0%_____3264_____2448
    5.6_____7.5______57.2%_____2307_____1730
    8.0_____10.7_____40.1%_____1617_____1213
    11______14.7_____29.2%_____1000_____750
    16______21.3_____20.1%_____812______609
    18______24.0_____17.9%_____721______541
    22______29.3_____14.6%_____590______443


    To maintain 100% Acutance at higher f stops you need to either drop your resolution, or down sample in post processing. For example if you only plan to use the final image at 1024 x 768 on the web you can shoot at f11 without any concern for diffraction effects being detectable in the image as they will always be contained within one pixel.
     
  9. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011

    Thanks for the great post! I think I have to re-read your info a few times to absorb what you are saying! I feel even less than a beginner now but that's fine by me and I love to learn.

    Just so I get it straight in my head.... when you say
    2.8_____3.7 100.0%_____4032_____3024
    would that mean that anything above f3.7 @ 4032x3024, the image would start to deteriorate or become more fuzzY?
     
  10. MajorMagee

    MajorMagee Mu-43 All-Pro

    Feb 24, 2011
    Dayton, OH
    The Airy Disk is the area that a point of light get's spread over after the information passes through an aperture causing the diffraction effect.

    At f2.8 the Airy Disk size is 3.7/1000 = 0.0037 mm and each location saving a pixel of data is 17.3/4032 = 0.0043 mm

    At f4 the Airy Disk caused by diffraction effects increases to 5.3 / 1000 = .0053 mm which is now bigger than the area saving the pixel data so some of it bleeds across into it's neighbor's space. To avoid the fuzziness that results you can drop the resolution to 3264 and then the area capturing one pixel of data becomes 17.3/3264 = .0053 mm and we're back to matching the size of the Airy Disk for one point of light with the Pixel Dimensions.

    If you don't do this the Accutance will start to fall off, in this case to 81% of the best case sharpness where every pixel contains only it's own discrete information. As you can see by the table by f8 you have 60% of your green light image information bleeding over into the neighboring pixels if you're still shooting and viewing the image at 4032 x 3024.

    Another way to think about this is that because diffraction spreads the photons from a point source of light over an increasingly larger area as the aperture blades close down you can't resolve as many points in the same amount of sensor area. At f22 the greatest number of green points of light you could resolve would be 590 x 443 on a micro four thirds chip.

    A bad lens can make this worse, but the greatest lens in the world can't change the physics behind this to make it any better.
     
  11. MajorMagee

    MajorMagee Mu-43 All-Pro

    Feb 24, 2011
    Dayton, OH
    This is another reason why chasing higher and higher megapixel sensors, only makes sense if you're making the chip physically bigger at the same time because your not really gaining anything unless you're shooting wide open.
     
  12. Djarum

    Djarum Super Moderator

    Dec 15, 2009
    Huntsville, AL, USA
    Jason
    This sort of speaks for the perfect case. The problem is most lenses don't perform well stopped down.

    What I have done in the past is write down the sharpest settings for lenses that have been reviewed by slrgear.com. I've found this invaluable. In some cases stopping down to F8 is required to utilize the sharpness of the lens, while compromising in terms of diffraction.
     
  13. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    Good tip thanks! I also got from that site this tip from the 14-42 II lens review that I never knew.

    "After a long conversation with an Olympus tech rep we found out that the lens needs to be reset with the camera, as both units have firmware. Instructions are to shut off camera, lock and remove lens, re-install lens but do not unlock. Turn on camera. Go to reset in menu and follow instructions. Then you can unlock the lens and use. I did all of the above and the lens works very well."
     
  14. anidel

    anidel Mu-43 Regular

    166
    May 13, 2011
    Twickenham, Uk
    Meaning to reset the whole camera to its default values while having the lens locked?
     
  15. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    While I was driving home today I was wondering the exact same thing. I'm not really sure if I did the right thing, but that's what I did and reset to the default values with the lens locked in the way they mentioned by unlocking it first etc. I guess that is what they meant although it sounds too simple? Didn't hurt my images and I just mentally changed a few things in the menu afterwards such as LF instead of LN and so on.

    Would be nice if someone chimed in hear who knows about this stuff? Another thing.... I have yet to notice the difference in image quality between LF and LN or for that matter LSFN and LN. LN actually looks sharper every time I take a shot. Any thoughts?
     
    • Like Like x 1
  16. anidel

    anidel Mu-43 Regular

    166
    May 13, 2011
    Twickenham, Uk
  17. klrman

    klrman Mu-43 Regular

    91
    Aug 11, 2011
    lol.. oh well, I did the reset thing and it didn't hurt any. Default has lens set to infinity every time the camera is turned on and the images are crisp and clear so might as well leave it alone.
     
  18. anidel

    anidel Mu-43 Regular

    166
    May 13, 2011
    Twickenham, Uk
    Indeed :)