Interpolate, factoring in axis cell values

[bradsm87]

I noticed today that with non-linear Y axis values, using the interpolate vertically function, it interpolates presuming axis row spacing is equal even if they aren't. It's be great if the feature could factor axis values.

[Vaughan]

Reasonably often a non linear axis wants linear data lined up against it such as tps as engine load or resistance to temperature, in these situations a linear interpolation taking into account axis values will result in incorrect or unhelpful values in the table.

While it would be helpful in other situations to have it look scale the interpolation values with the axis the kind of axis that would want a scaled interpolation is often already setup with evenly spaced increments.

I would argue that it's more useful to be able to make a linear interpolation across a non linear axis than to make an interpolation that is scaled to the axis.

Thank you for the suggestion though.

[Adamw]

Maybe Im understanding it wrong, but if you interpolated the table data using the same polynomial as a non linear axis had then wouldnt it effectively give you same result as having a linear axis with linear interpolation?  So in other words exactly the same as just keeping the first and last axis value and deleting all breakpoints in-between?

As an example if I interpolated this table from the far left cell to the far right cell using the same curve as the axis, then it would give me the result below.  I would get exactly the same control output from the table if I deleted the 5 & 20 columns.

[bradsm87]

4 hours ago, Vaughan said:

Reasonably often a non linear axis wants linear data lined up against it such as tps as engine load or resistance to temperature, in these situations a linear interpolation taking into account axis values will result in incorrect or unhelpful values in the table.

While it would be helpful in other situations to have it look scale the interpolation values with the axis the kind of axis that would want a scaled interpolation is often already setup with evenly spaced increments.

I would argue that it's more useful to be able to make a linear interpolation across a non linear axis than to make an interpolation that is scaled to the axis.

Thank you for the suggestion though.

I agree for many situations . Why not have two separate options then?

2 hours ago, Adamw said:

Maybe Im understanding it wrong, but if you interpolated the table data using the same polynomial as a non linear axis had then wouldnt it effectively give you same result as having a linear axis with linear interpolation?  So in other words exactly the same as just keeping the first and last axis value and deleting all breakpoints in-between?

As an example if I interpolated this table from the far left cell to the far right cell using the same curve as the axis, then it would give me the result below.  I would get exactly the same control output from the table if I deleted the 5 & 20 columns.

But often on 3d tables you will want to interpolate some but not all rows or columns and still have the resolution to make changes in other areas. One example where I would have liked it for a starting point was is the E-Throttle target table to be able to make a non-linear curve with finer resolution in the bottom 15%, not a lot of resolution around the middle where it's quite straight and linear, and finer adjustment again near the top end but if I want to use the interpolate function, I'm kinda forced to have a linear Y axis.