Observation about Non-Linear Functions

[Lin Evans]

In applying the smooth non-linear functon to a couple of my older slideshows I discovered a feature which is really nice about the new non-linear functions for PZR.

If I click on the second keypoint (there could be a half dozen or more following) for a single slide, apply the non-linear function (smooth, in my case) to each (pan, zoom, rotate) then click on the "setup" for each and separate each keypoint pair, this attribute is carried forward for all keypoints for that slide.

That is the complete complement of keypoints following the second one are automatically updated with the smooth (or whatever has been chosen) attribute. This makes short work of updating older slideshows to reflect the new non-linear functions.

addendum to this observation: I find it easier to make all original actions for a single slide in linear then simply update the second keypoint with the "soft" non-linear for each (PZR) and separate the keypoints at this juncture which then applies this non-linear attribute to all following keyponts in this slide. For me it's faster than doing it at each keypoint on the original.

Congratulations to Sergey and Igor for this really nice feature.

Lin

[Lin Evans]

Hi Tom,

I suspect this is because you "may" not have applied the "separate" on keypoints for each of Pan, Zoom, Rotate or not have "glued" them back after changing the attribute back to linear. If, for example, you apply it to Zoom without applying to Pan and both Pan and Zoom values deviated from the original keypoint, the unusual effects will happen. As long as you are aware you can actually "use" this to make interesting effects, but I suggest that if you want to achieve the original linear you would get it by changing all three back to linear and "glue" all keypoints back to the original condition.

Only when we actually ignore the bounding rectangle when setting up our pan, zoom, rotate is it likely that we would have a "pure" pan, zoom or rotate. That is if we grab a corner of the bounding rectangle to size it and we move it even a tiny amount we have also affected the Pan as well. And if we are not very, very careful, we may also have changed the rotation from zero even if not intended. It's much safer to simply set the non-linear effect for pan, zoom AND rotate unless, of course, we have some reason to have one set to a linear and others set to non-linear.

Best regards,

Lin

I also think the non-linear movement is fantastic.

Sometimes I notice strange motion paths. For instance if change from a custom pan setting back to linear pan then the motion path is changed. I was expecting a return to the normal linear motion path. In some way this is really neat (curving motion), it almost looks like a b-spline/Bezier motion path. I am hoping that 5.n will have this feature.

Tom