Timeline for Why is there a discrepancy between these two transform.up vectors?
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 1, 2017 at 13:41 | history | edited | Vaillancourt♦ | CC BY-SA 3.0 |
Clean up of the post.
|
| Nov 17, 2017 at 16:43 | history | edited | confuzzled | CC BY-SA 3.0 |
deleted 1044 characters in body
|
| Nov 17, 2017 at 16:41 | answer | added | confuzzled | timeline score: 1 | |
| Nov 17, 2017 at 6:12 | history | edited | confuzzled | CC BY-SA 3.0 |
reverted to show my last updated question before finding the error - added description of my failing
|
| Nov 17, 2017 at 5:50 | review | Close votes | |||
| Dec 1, 2017 at 13:41 | |||||
| Nov 16, 2017 at 23:31 | history | edited | confuzzled | CC BY-SA 3.0 |
found problem, own stupidity
|
| Nov 16, 2017 at 23:06 | history | edited | confuzzled | CC BY-SA 3.0 |
added more information as well as a minimal way to reproduce a similar error
|
| Nov 16, 2017 at 22:35 | comment | added | confuzzled | I'll see In a second. I tried adding a script to the empty objects I used as track points, these track points were set as children to the mesh object and its these transforms that are copied. The script I attached simply had called Debug.LogError(transform.up); in update. This output matches the output of the tracking object, but differs from the script attached to the rotating mesh object. I find it peculiar, shouldn't they be referencing the same block of heap allocated memory? Why is transform.up different from referenceobj.transform.up | |
| Nov 16, 2017 at 22:28 | comment | added | confuzzled | It was definitely in local mode, first thing I checked. I also experimented with various things to see if I could get any clue, even things I was sure would be pointless. TransformDirection, InverseTransformDirection, vector3.up, all sorts of faffing about. | |
| Nov 16, 2017 at 22:25 | comment | added | confuzzled | I'd rather not paste the scripts. The structure is as follows - the parent is an empty, there is a child object with a mesh that gets rotated in 90 degree steps, inside that script I have Debug.LogError(obj.transform.up); This gives a different result than expected. So I then did a track script that just copies the transform value of this object. the script is attached to an object that is child to the parent empty. In the script aside from copying transform, it also has Debug.LogError(transform.up); which does give expected results | |
| Nov 16, 2017 at 22:24 | comment | added | DMGregory♦ | Note that the orientation of the transform gizmo will vary based on your Transform Gizmo Toggles in the Editor Toolbar - in Local mode it will reflect the selected object's local transform. In Global mode it will stay aligned to the world coordinate system. Can you confirm which mode you have your editor in? Screenshots can often help in diagnosing geometric problems like this. | |
| Nov 16, 2017 at 21:58 | review | First posts | |||
| Nov 17, 2017 at 1:54 | |||||
| Nov 16, 2017 at 21:53 | history | asked | confuzzled | CC BY-SA 3.0 |