Settling after a rig change
While the camera blends to a new Rig, its automatic follows wait, then ease back in. Two motions at once would read as a jolt.
A blend is already a camera move. If the automatic follows (turning to face the way you run, recovering pitch, following slopes) kept steering during one, you'd see two motions with different targets at the same time, and it would read as a jolt even though each is smooth on its own.
So while a Rig blend runs, those follows wait. When it lands, they ease back in over the Rig's Follow Ramp Time.
Adding more quiet#
Follow Settle Time on the Rig adds extra quiet after the blend. Use it when a shot should arrive and simply sit for a moment before it starts steering itself again. It's zero by default, so out of the box you get the blend's own length and nothing more.
What doesn't wait#
Three things carry on through a blend, on purpose:
- Falling. A fall is over in a second or two; framing it late means never framing it at all.
- A recentre the player asked for. They pressed the button.
- A point of interest's shot. An authored shot takes priority.
Go deeperBlend the results, not the settings
During a Rig change, every value the camera uses is a blend between the two Rigs: two arm lengths, two fields of view, two framings. That's simple until a value passes through a limit on its way to the screen, such as Max Camera Move.
Blend first and then apply the limit, and the camera sets off at the speed of the whole gap, hits the limit part way, and sits on it for the rest of the blend. Nothing jumps, but the speed drops to zero in one frame and the shot looks like it changed its mind.
Apply each Rig's limit to its own answer and blend those instead, and the shot travels from where the old Rig sat to where the new one sits at one steady speed, arriving with everything else. That's what Blendable does, everywhere a limit or threshold sits in the path.
Go deeperWhy the waiting runs through the follow ramp
The three exceptions above aren't scaled by the follow ramp in the first place, which is why the waiting is done by holding that ramp at zero rather than with a separate switch. A switch would need its own list of what counts as a follow, and that list would drift out of date the first time someone added a new one.
The rule that keeps it honest: anything an automatic behaviour measures has to be held at rest while that behaviour is switched off. A value that keeps running, or freezes at its last reading, gets used the moment a blend switches its feature back on, carrying a reading earned under a different Rig.