Sorry, Paul,
but you must be wrong. Within a circle that covers a rectangle of 24x36mm
you can draw a rectangle 27mm broad with a greater height than 24mm!
A mathematician better than you and me may tell us the useable height.
Greetings
Jochen (Hans-Joachim)
----- Original Message -----
From: "Paul Farrar" <farrar@xxxxxxxxxxxx>
To: <olympus@xxxxxxxxxxxxxxx>
Sent: Thursday, May 03, 2001 8:34 PM
Subject: [OM] 18mm is no substitute for a 24mm shift
> Since the 24mm shift and the 18mm have the same angle of view, I've
> heard people say that one can get the equivalent of a 24mm shift by
> using an 18mm and selecting an 18*27mm subframe. I've agreed in my
> own mind with this, without really thinking about it. But a few days
> ago, out of the blue, I had a theophany and realized it couldn't be
> true. Two days ago I tried it out with both lenses and found the
> differences were quite dramatic, especially for rise and fall
> (shifting across the narrow dimension). The 24mm shift has a
> tremendously greater coverage.
>
> The reason is this:
> Both lenses have the same ratio of (image circle radius)/(focal length),
> so the 24mm is like a scaled up 18mm. The 18mm fits a 24*36mm frame
> into the image circle so that the corners just touch the image
> circle. The 24 has a 33% large circle and moves a 24*36mm frame
> around within it. To get a 24mm equivalent with the 18mm, you take an
> 18*27mm subframe and shift it around. The problem is that the
> subframe is limited to within the 24*36mm full frame. This mean that
> you can only get a 3mm rise or 170f the frame height. On the 24,
> however, the 24*36mm frame is limited by the image circle, not the
> boundaries of a slightly larger frame. This means you can get a 10mm
> rise before your frame corners get to the image circle, or 400f the
> frame height. The difference is quite dramatic. The 18mm doesn't
> come close to the ability of the 24. For lateral shifts, the
> difference is much smaller, but detectable. This is because on the
> sides the 18mm full frame can get much closer to the image circle.
>
> You can get a bigger rise with the 18mm by turning it vertical and
> shifting a 24*16mm frame for a 10mm rise, or 63%. But you can do the
> same trick with the 24 shift and get a full 100% rise.
>
> What lens is equivalent to the 24 shift? At full rise, the top of the
> frame is 22mm above the axis, roughly the same as the focal length.
> For a subframe confined to a 24*36 full frame, this means you need a
> 12mm focal length, with a 12*18mm subframe. You will get more lateral
> shift with the 12mm. Only Voigtlaender makes a 12mm -- but it's
> cheaper than a 24mm shift, but by then you're getting close to Pocket
> Instamatic format size.
>
> The 24mm shift can do all sorts of tricks, not just prevent
> keystoning. For example you can control the elliptical distortion of
> 3D objects near an edge (present even on a perfectly "nondistorting"
> rectilinear). You shift the object towards the center of the image
> circle. You get a corresponding increase of distortion on the other
> side of the frame, but if there is a person's head, or some item of
> interest, you can choose for it to have low distortion.
>
> Unfortunately, it looks like the 24mm shift is no longer available
> new, at least in the US. It's not too hard to find used, but prices
> have risen a lot in the last few years.
>
> As a true Zuikoholic, you really neeeeeed this lens! Show the above
> to a sceptical spouse -- but, ummmm, don't mention my name,
> especially if you live near the Gulf Coast.
>
> Paul Farrar
>
> < 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 >
>
< 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 >
|