Dr. Diffraction says it's all in the diffraction numbers. At 8MP on a
4/3 size sensor you get most of the resolution at f/8 but full
resolution of the red channel means limiting the aperture to about
f/6.3. And that's for an ideal (non-existent) lens. So a more
practical limit is probably about f/5.6... assuming the lens itself
performs well at f/5.6.
But for those who have to see to believe you'd have to make and closely
compare prints of about 11x14 or even larger. With prints especially
since the printing process itself tends to soften the image and reduce
the final resolution.
Small apertures can be important to DOF, of course, and especially at
macro distances. But it helps to realize that the increased DOF does
not contain greater resolution. If the aperture is smaller than the
diffraction limit all that's happening is that the previously sharper
foreground pixels are being smeared to equal the lower resolution of the
fuzzier background pixels. All may look well to the eye with apparently
even sharpness but maximum resolution has not been achieved.
Also remember that the focal ratio indicated on the aperture ring is
only valid at infinity. At macro distances the focal ratio is much
greater than the indicated setting. At a magnification of 1:1 a lens
must be extended by its own focal length beyond infinity focus.
Therefore at 1:1 a 50mm lens is actually 100mm from the image plane and
an indicated aperture of f/32 is doubled to an effective aperture of
f/64. If we accept that 8MP on a 4/3 sensor is achieved at f/8 we must
realize that each additional stop decreases resolution by a factor of 2.
Therefore, by f/22 the actual resolution is only 1MP and by f/64 it's
down to only 1/8MP... yes, that's 1/8th. That's only enough to make a
sharp image of about 300x400 pixels (or a bit smaller than the images on
Jan's page)
Dr. DOF
On 12/18/2012 12:53 AM, Moose wrote:
> On 12/15/2012 6:11 PM, Jan Steinman wrote:
>> ... I did a study of the various OM macro lenses on the E-300
>> (http://www.biofos.com/cornucop/opage_1.html).
>
> Ah yes, I remember that study. As you note, the changes in lighting make it
> very hard to compare the lenses. Light
> direction and degree of diffusion make such a big difference with 3D surfaces
> like that, especially reflective ones.
>
> What I don't recall noticing back then is that the limited camera resolution
> is almost certainly masking the onset of
> diffraction limiting of lens resolution, leading to possibly erroneous
> conclusions about optimum aperture. Easier to see
> now, with experience of a higher resolution 4/3 sensor.
>
> Dr DOF would say all is in the numbers, but I have to see to believe. ;-)
>
> Resolving Moose
>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|