Currently, the library provides tools for exploring an envelope. However, the parameters into the exploration process are manually determined. Excessively large jump distances can result in boundary lost exceptions, and excessively small jump distances can result in wasted resources. Determining proper parameters into the process will automate this process and avoid trial and error.
The solution, or at least part of it, is providing Envelope diameter approximation, which would provide a measure of how large the envelope is. This measure of size would inform how to scale the jump distance appropriately. Technically, this can be done for a single dimension, but that risks measuring a disproportionately large or small axis. An extension for this tool would be to provide an N-dimension parameter, where the user can specify how many axes to measure the diameter over. The algorithm would need to take an initial chord through the geometry, pick the center-point, and then rotate 90 degrees to find the next axis.
Currently, the library provides tools for exploring an envelope. However, the parameters into the exploration process are manually determined. Excessively large jump distances can result in boundary lost exceptions, and excessively small jump distances can result in wasted resources. Determining proper parameters into the process will automate this process and avoid trial and error.
The solution, or at least part of it, is providing Envelope diameter approximation, which would provide a measure of how large the envelope is. This measure of size would inform how to scale the jump distance appropriately. Technically, this can be done for a single dimension, but that risks measuring a disproportionately large or small axis. An extension for this tool would be to provide an N-dimension parameter, where the user can specify how many axes to measure the diameter over. The algorithm would need to take an initial chord through the geometry, pick the center-point, and then rotate 90 degrees to find the next axis.