I wish Rhino tools were equipped with an analytical AI that could analyze the model while running certain commands—such as FilletEdge, FilletSrf, OffsetSrf, or NickEdge.
Many of these commands fail due to minor geometric issues, open or problematic surfaces, edge defects, or small errors that prevent proper execution.
If the tools could identify the real cause and, when safe, automatically apply corrective fixes, the workflow would become much faster and more reliable
How would it actually “analyze” such a thing? All the quality training data for such a system is locked in the proprietary work files of Rhino users and you can’t have it.
Is it possible that if software tools were equipped with a local generative AI and the user’s hardware had an NPU, the AI could analyze models locally and quickly, detect issues such as commands failing to execute properly (FilletEdge, FilletSrf, OffsetSrf, NickEdge), problematic edges or holes, and incorrect tolerances, and even suggest safe and limited corrections, without accessing users’ private files or requiring a server?
در تاریخ دوشنبه ۹ فوریه ۲۰۲۶، ۱۹:۴۱ JimCarruthers <notifications@mcneel.discoursemail.com> نوشت: