Jeff's mention of shift reminds me of a thought I had recently about the
effect of perspective correction in PhotoShop on image resolution.
Resolution goes down if you start pulling pixels apart. Those thoughts
made me contemplate maybe a PC lens was still justified. On the other
hand I happened to note DPReview's test of the new Canyon tilt/shift
24/3.5 Mk II the other day.
<http://www.dpreview.com/lensreviews/canon_24_3p5_tse_c10/>
At f/8 the test data shows that the lens resolves the the sensor limit
on the 21 MP sensor of the 5D Mk II but starts falling down rapidly
about 2/3 of the way out to the edge.
<http://www.dpreview.com/lensreviews/canon_24_3p5_tse_c10/page3.asp>
But that's with the lens unshifted. For what happens at max shift
scroll down on the same page to next graph labeled "Extended image
circle tests" which has the lens fully shifted. Here the lens performs
well in the center at f/8 and below but resolution at the edge of the
frame has typically fallen off to less than half the center value. So,
I think the question of shifting with software vs. a lens doesn't have
an easy answer. Of course, this lens tilts as well as shifts. PS
doesn't know how to do that trick yet.
Chuck Norcutt
Jeff Keller wrote:
> Besides the physics and modeling explaining the light fall off don't
> forget with large format, lens shift is common. So although a 90mm
> Super Angulon XL may not seem that wide, the center filter works for
> the 110 degree coverage of the entire image circle. I don't know how
> effective photoshop would be at removing vignetting when the image
> isn't centered.
>
>> From occasional visits to largeformatphotograpy.info it seems that
>> a good number of people still shooting large format are shooting
>> LARGE images and contact printing (11x14 up to about 16x20).
>
> In the end it seems the significant cost of a center filter motivates
> many to do without.
>
> Jeff Keller
>
>
> -----Original Message----- From: usher99@xxxxxxx
> [mailto:usher99@xxxxxxx] Sent: Monday, July 20, 2009 3:09 PM To:
> olympus@xxxxxxxxxxxxxxxxx Subject: [OM] [OT] Center filters
>
> Well, not so fast. The inverse square law accounts for only one of
> the cosines in the Cosine**4 law. I think you are describing the
> Slyusarev effect which lens designers utilize to increases the
> apparent size of one or both pupils for off-axis points and greatly
> mitigates the cosine**4 law.
>
> A second factor that causes light loss off axis is that the lens
> pupil is no longer round but elliptical when viewed from the
> periphery. It also strikes at an angle off axis. There is
> additional attenuation due to the Lambert effect. I think that makes
> four, but I am tired after a long day and a couple hours left to go
> and can no longer count.
>
> Mike (protg of Dr. Focus)
>
>
> see
>
> http://toothwalker.org/optics/vignetting.html
>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|