Some observations I made comparing the two photos:
John Hudson wrote:
> Looking closely at John's photo...
http://www.taiga.ca/~gallery/subpages/gardner/gardner5.html
> ...the vertical lines of the architecture on
> either side of and beyond the altar appear to be parallel but neverthess
> they create the illusion, to me at least, that the walls are falling
> outwards. I do not sense anything in my photo...
http://persweb.direct.ca/jahudson/images/ranmoor.jpg
> ...that the walls are falling
> outwards.
First John Gardners photo is not perfectly aligned (or the church isn´t
symetrical build ;-), as can be seen on:
http://www.stud.uni-hannover.de/user/73239/gardner5_off_center.jpg
I have marked the center lines of the roof and of the rear wall.
In contrast to that, John Hudsons picture is perfect centered, as you can
see on:
http://www.stud.uni-hannover.de/user/73239/ranmoor_picture_center.jpg
> I am wondering if the effect of the shift lens is to create verticals but at
> the expense of the illusion that outside verticals appear to be falling
> outwards.
No that´s not the intended effect, but you should bear in mind, that an
picture taken with an shifted lens is in fact the cropped picture of an
superwide angle lens.
In case of John Gardners 35mm shift, of an lens comparable to an 24mm
lens. So one has to expect the same acute-angles and "dynamic"
perspective, this includes effects like objects in the picture edges
looking "blown-up" or being oversized.
Of course this"perspective distorsions" are perfectly normal, one is
just not used to it, because of the way the human visual system works.
This "distorsions" are further emphasized, because "shifted" pictures
use normaly the extrem part of the image the lens produces, as can be
seen on the picture:
http://www.stud.uni-hannover.de/user/73239/gardner5_picture_center.jpg
The center of the projection and the point were all lines converge is on
the bottom of the picture, while on John Hudsons picture the lines
converge in the center:
http://www.stud.uni-hannover.de/user/73239/ranmoor_picture_center.jpg
To counter this unwanted effects, one might shift not as much as is
needed to completely eliminate perspective "distorsions" and keep the
camera a little bit tilted.
Right angle oriented is not always best looking.
The old Greeks have know this perspective effects and have constructed
there temples to look more pleasing to the observer.
There temples were by far not as "right angled" as they look.
The colums are slightly conical and curved, the bottom with a bigger
diameter than the top.
The base, on which the columes were errected, had a slightly curvature,
as well as the roof.
A little, hardly noticeable distorsion please human perception and
expectation more than a "perfect" right angle construction.
So avoiding perspective "distorsions" is not the "ultima ratio" as
"orthodox" architectual photography tríes to teach.
Regards
Richard
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|
