If Poser had to decide how much of the next frame (in rendered form) is identical to the last frame, it would have to render the next frame to see what it looks like, and then make the decision about which pixels to change, and -then- figure out which objects need to be calculated this time. The latter determination would not be trivial, because anything can cast a shadow or a reflection into any part of the scene. So you would actually lose time that way! Some time could be saved if 100% identical frames are repeated, because that could be determined easily from the parameters. But a long 100% still sequence is rare in an animation, almost by definition. (Why animate in that case?)

My python page
My ShareCG freebies