I'm not totally certain of this, but for non-bezier curves, (so linear and catmull) it may be more convenient if, by default, the contextual steps range each control point lives in was shortened to span from the steps of the expressions right before and right after. For example:
[a:b:c:d:1,2,3,4]
Here:
a would span steps range (-1, 2)
b would span steps range (1, 3)
cwould pan steps range (2, 4)
d would span steps range (3, 18) (assuming 20 sampler steps)
For the first control point, the min step is the enclosing interpolation's own min step, and for the last one, the max step.
IIUC this should be a better default value for nested interpolation:
[a:(b:1,2):c: 4, 9, 14]
I'm not sure whether the default steps range of catmull-rom control points should be processed in exactly the same way, if they should take the step of 1 control point further, or if they should stay as they are in main ATM.
To override this behavior, you can just specify a prompt editing step:
[a:[b:0]:c:1,2,3]
Here, b would span steps range (0, 3) instead of (1, 3).
I'm not totally certain of this, but for non-bezier curves, (so linear and catmull) it may be more convenient if, by default, the contextual steps range each control point lives in was shortened to span from the steps of the expressions right before and right after. For example:
Here:
a
would span steps range(-1, 2)
b
would span steps range(1, 3)
c
would pan steps range(2, 4)
d
would span steps range(3, 18)
(assuming 20 sampler steps)For the first control point, the min step is the enclosing interpolation's own min step, and for the last one, the max step.
IIUC this should be a better default value for nested interpolation:
I'm not sure whether the default steps range of catmull-rom control points should be processed in exactly the same way, if they should take the step of 1 control point further, or if they should stay as they are in main ATM.
To override this behavior, you can just specify a prompt editing step:
Here,
b
would span steps range(0, 3)
instead of(1, 3)
.