I'd agree with Randy here. If the 18mm prime is not designed for the film
back then it would have an equivalent focal length of, in your example of 27mm
(though that is just based on your example. )
However if it is a prime lens designed for the chip, then it is an 18mm lens,
so no conversion necessary.
However again, the solve you would get should work either way, but the relation
between the point cloud and camera will vary and will begin to be an issue for
objects that should be close to the camera.
That is you can double a focal length and move an object twice as far away in
nuke and it should look the same, except for objects very close, when
distortion artefacts creep in, and I don't mean lens distortion
characteristics here before any one jumps in, I mean if you stick a sphere
right next to the lens and have a super wide angle, that isn't the same as
an object further away with say a standard lens size. (eg 50mm for 35)
So you can track and try 2 solves with 2 different lens lengths, and if there
are objects close to camera you should see a difference in shape of any test
spheres etc, otherwise it should still lock.
I couldn't find anything quickly online to confirm your film back sizes,
but someone has created a whole set of presets for cameras (Nathan?) which are
very useful on nukepedia.
H
_______________________________________________
Nuke-users mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
