Chuck might be wrong at times but he never pulls numbers out of the
ether (or anywhere else not credible). You really need to read the
article referenced here to fully understand what's going but if you just
want to cut to the chase and see the source of the data I'm quoting see:
<http://www.luminous-landscape.com/tutorials/resolution.shtml> and
scroll about 80% of the way down the page to find "Table 3".
Horizontally, table 3 shows different formats ranging from 4/3 to medium
format. Vertically it shows apertures from f/1 to f/22. Under each of
the four formats shown you'll find three columns with headings:
W=0,4 microns | w=0,55 microns | w=0,7 microns
Since one of the determinants of resolution is the wavelength of light,
the three columns are giving the wavelengths of light at the mid-points
of the blue, green and red channels. If you read the text under the
table it explains how to interpret the data. Using a 35mm frame size
(either film or digital) the maximum *theoretical* resolution obtainable
at f/11 is 16MP (in the green channel). But it's 30MP in the blue
channel and only 10MP in the red channel.
Caution: Reading this article will drag you all through the depth of
field and circle of confusion stuff all over again and even deeper (if
you can imagine that). :-)
Dr. Focus
On 10/26/2014 11:29 AM, John Hudson wrote:
Are your MPs per lens focal length per aperture numbers out the ether or
do they have some source to which curious minds can refer ?
For instance, you state that for "this 16MP camera ........... f16
knocks the sensor resolution down to 8 MP on the green channel and 5MP
on the red channel."
jh
On 10/26/2014 12:00 PM, Chuck Norcutt wrote:
Diffraction territory for this 16MP camera is about f/11 (for the
green channel, I'm assuming that the monochrom uses all colors). f/11
would slightly reduce resolution in the red channel but f/16 is
severe... it knocks the sensor resolution down to 8 MP on the green
channel and 5MP on the red channel.
I can't tell what the shooting distance is for the foreground but it
looks close enough to me that I wouldn't have tried it with a focal
length longer than about 35mm. At f/11 on 50mm a hyperfocal shot's
closest focus is about 11 feet (even at modest resolution). Using f/16
puts you seriously into diffraction territory for a 16MP camera. It
knocks the green channel resolution down to 7MP and the red channel
down to 5MP.
For this particular shot and focal length I have to agree that a
composite of 2 images would be required... unless you had a tilt lens.
Chuck Norcutt
On 10/25/2014 11:56 PM, Moose wrote:
On 10/25/2014 12:07 PM, Chuck Norcutt wrote:
Some hyperfocal work by Tina. Ask her how she did it. She used f/8
at 90mm. I suspect the lighthouse might be a bit soft at a larger
size. I'd have used f/11 and a shorter focal length if I could have
gotten close enough. But maybe she couldn't get closer.
OK, lets look at another one, Prospect Harbor Lighthouse. Just as you
suggest, shorter FL and f11. <http://www.pbase.com/image/157906308>
And yet, the near foreground is quite OOF. It appears to me that the
lighthouse and tree were also a bit OOF, and sharpening has been used on
them. The result is slightly etched, artificial looking, to my eye.
Two shots, one focused near, one far, feathered together in the water,
would get everything in focus and better looking, overall.
Cropping and different compositions are possible, as with the following
horizontal shot. But to get the chosen shot, hyperfocal technique won't
do it without slipping into diffraction limited territory - not that
anyone would see it at this size.
Two for One 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/
|