Originally posted by lithos I was hoping you come along and explain how the trapezoidal prism works...if you could elaborate on how it's different to a normal pentaprism, if there is any difference, that'd be great. Does the trapezoid flip the image the same amount of times as a normal prism?
Well, I don't know actually.
Forced to think about it, my wild guess would be that it works a little different from a traditional roof pentaprism with 3 reflections. Looks like it has the same 3 reflections but no roof. I.e., the left/right flip would have to come from the optical system in the ocular and that the prism itself wouldn't actually invert the image orientation.
Also, unlike a pentaprism, the image seems to be focussed (more) before entering the prism, so that it gets smaller along its path and the prism surfaces can be made smaller for 2nd and 3rd reflection surfaces. And magnified by the optical system in the ocular upon exit of the prism.
I searched and found a patent from 1978 hold by "Asahi Kogaku Kogyo Kabushiki" (is this Pentax?) which describes the principle:
Patent US4171888
Seems like Pentax did their homework some 32 years ago
Some difference to the patent: The mirror doesn't seem to be 45° exactly (more like 40°). There is a noticeable tilt in the entire optical viewfinder system from mirror to prism exit. I would say that this is a sophisticated piece of engineering.
However, the 645N is even 6mm flatter and needs a larger mirror. Therefore, I guess that this isn't new technology for Pentax at all. The viewfinder in the 645N is described as "Keppler telescope finder", a variant described in the above patent as well. This variant was shown in the patent with mirrors but I believe, the 645N finder has a prism as well.