Vinten Radamec Free-d - Manual (Page 25)

Vinten Broadcast Ltd free - d ™ Installation Manual V1.4.4
Page 25 of 46
If the real and virtual objects cross the field-of-view at different rates, the view-
angle of the real lens and the view-angle of the virtual lens are different. If the
virtual object crosses the field-of-view faster than the real object, the view
angle of the virtual lens is smaller than that of the real lens. If the virtual
object crosses the field-of-view slower than the real object, the view angle of
the virtual lens is larger than that of the real lens. This may require alteration
to a studio lens calibration file, adjustment of the zoom and/or focus sensors
or the resetting of a reference point (eg, an end-stop).
Repeat the test with different settings of zoom and focus. If the view angles
match at some settings but not at others, the studio lens calibration may need
to be corrected. Note that at the wide-angle end of the zoom range lens
distortion may be significant, in which case it may not be possible to make the
real and virtual objects track accurately over the full width of the image. Note
also that, unless the objects are infinitely distant, a Y-offset error or a nodal-
shift error can result in a similar symptom.
7.9 Vertical view angle and aspect ratio
Repeat the previous test, but this time tilting the camera rather than panning it.
The registration between the real and virtual images should be just as good as
it was when the camera was panned.
If the performance is worse when tilting than when panning, suspect that
either the studio camera or the virtual set system is selected to the wrong
aspect ratio.
7.10 Nodal shift and camera Y-offset
Position a real object and a virtual object (ideally something like a thin vertical
pole) as close to the camera as you can whilst being able to focus on it with
the lens at its fully wide setting and with a tilt of zero. Ensure that the real
and virtual objects are co-sited by observing them from two different
directions, 90 degrees apart, and moving one or both as necessary (pan the
camera so that the objects are in the centre of the image).
Pan the camera from side to side whilst observing the relative positions of the
real and virtual objects. If the real and virtual objects cross the field-of-view at
different rates (and having already ascertained in the previous test that the
lens angles are the same) the real and virtual viewpoints must not be
coincident. If the virtual object crosses the field-of-view faster than the real
object, the viewpoint of the virtual camera is nearer to the object than the
viewpoint of the real camera. If the virtual object crosses the field-of-view
slower than the real object, the viewpoint of the virtual camera is further away
from the object than the viewpoint of the real camera. Check that the Y-offset