Re: Keystoning in projection

[P3D John Bercovitz <bercov@xxxxxxxxxxx>]

Wm. C. wrote:

> In fact I think you'd have a tough time creating a keystoned image with
> any manner of mirrors and/or prisms as long as you kept the film
> parallel to the screen.

Bill, I've been thinking on this some more.  I think this (film parallel 
to screen) is what you would call a necessary but not sufficient 
condition.  If you add another necessary but not sufficient condition, 
that the extended axis of the lens, after it works its way through the 
mirror system, end up perpendicular at the screen, I think you'd have it.

Here are some examples:  If you have the projector located down at 
operator height and you use an ordinary periscope to get the projection 
up to the height of an elevated screen, you'd have no problem.  However, 
if you now move the projector off to one side of the room and rotate the 
upper mirror of the periscope to put the image back on screen, you'd get 
your keystone back (not to mention the people on the screen would 
acquire a severe list due to image rotation).

Now for an exception.  If you made the axis of the projector parallel to 
the screen and used a twisted periscope to put the image on the screen 
(keeping the axis of the lens perpendicular at the screen) you would 
have a good image if you used a dove prism to rotate the image back to 
upright.

I wrote:

> Now try 5-perf format.  Screen height goes from 56" to 66".  New 
> figures are 15.5 degrees and 5'6" elevation gain.  Even worse.

The number of things I forgot to change when I went from the 7-perf calc 
to the 5-perf calc exceeds the number of things I _did_ remember to 
change.  So for the TDC with 4" lenses, the projector needs to be 76" 
above the center of the 66" screen which means the projector tilts at an 
angle of 15.1 degrees.  The things I forgot to change are that the 
projector has to move back to 24' from 20' and that the audience moves 
back from 7.5' to 8' from the screen.  So as I've said in the past, in
some ways 5-perf is more satisfactory giving much more pixel area on the 
screen and some more audience distance from the screen.

Marvin Jones wrote:

> Or better yet, if you tilted your screen to HALF the angle from the 
> projector to the viewers, wouldn't the keystoning built in on one side 
> be pretty much offset on the other?

Well, let's say that the projector is high as we've been discussing.  So
the upper part of the image is narrow due to keystoning.  Due to the slight
tilting of the screen, that part of the image will also be farther away
from the viewer so it will subtend a smaller angle at the viewer due to
this second effect also.  No net gain, I think.  However, the focus will be 
improved.

I think actually that 15 degrees is not so very bad although 5 or 10 degrees
would be better.  Maybe this is a non-problem (except for me with the low 
ceiling in my viewing room).

John B


------------------------------