FreeCADView Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0002314 | Draft | Bug | public | 2015-10-28 17:53 | 2017-01-15 02:18 |
| Reporter | kobejohn | Assigned To | |||
| Priority | low | Severity | minor | Reproducibility | always |
| Status | closed | Resolution | no change required | ||
| Platform | Windows | OS | Windows | OS Version | 10 |
| Product Version | 0.15 | ||||
| Fixed in Version | 0.15 | ||||
| Summary | 0002314: Windows Rotation.toEuler() faulty | ||||
| Description | In Linux, I get what looks like reasonable results from Rotation.toEuler(). However in Windows, when angles are 90 or -90, the results of toEuler() are strange. I will put working code in the steps to reproduce. This can be done headless or in the FreeCAD console. | ||||
| Steps To Reproduce | import FreeCAD # (-90 < angle < 90) (not inclusive of the 90s) clearly works in both windows and linux rotation = FreeCAD.Rotation(-10, 20, -30) rotation.toEuler() # --> (-10, 20, -30) # angle = -90 or 90 seems to work in reality (the actual object / GUI) in both windows and linux, # but toEuler() is strange in Windows rotation = FreeCAD.Rotation(20, 90, 0) rotation.toEuler() # --> in linux (20.6, 90, 7.1) which looks like some angle wrapping? # it works in the GUI. Is it possible these numbers are a bug also? rotation.toEuler() # --> in windows (0, 90, 0) which is definitely not right # it works in the GUI. # beyond 90 degrees, I get different results from linux and windows although not blatantly wrong like (0, 90, 0) so I'm not sure how to evaluate whether either of them is correct. | ||||
| Tags | rotate, toEuler | ||||
| FreeCAD Information | |||||
|
|
When testing this on Windows for the 2nd test I got the same values as you on Linux. So, there is no difference in behaviour. But that's actually irrelevant because you ran into the so called "gimbal lock" which is a singularity. FYI, for Euler angles there exists many different notations and we use the XY'Z'' notation (which to be correct is not called Euler angles but Tait–Bryan angles). Here you will find some information about the gimbal lock: https://en.wikipedia.org/wiki/Conversion_between_quaternions_and_Euler_angles#Singularities |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2015-10-28 17:53 | kobejohn | New Issue | |
| 2015-10-28 17:56 | kobejohn | Tag Attached: Rotation | |
| 2015-10-28 17:56 | kobejohn | Tag Attached: toEuler | |
| 2015-10-29 10:31 | wmayer | Note Added: 0006509 | |
| 2015-10-29 10:35 | wmayer | Status | new => feedback |
| 2016-05-05 19:56 | wmayer | Status | feedback => closed |
| 2016-05-05 19:56 | wmayer | Resolution | open => no change required |
| 2016-05-05 19:56 | wmayer | Fixed in Version | => 0.15 |
| 2017-01-15 02:18 | Kunda1 | Tag Attached: rotate | |
| 2017-01-15 02:18 | Kunda1 | Tag Detached: Rotation |
