sl-service-account
commented
10 years ago Alexa Linden commented at 2014-06-09T15:31:46Z
When filing a Jira please provide more details of your problem. Without this information your jira will be closed as Needs More information and will not be triaged.
Please include:
(!) Which version of the client are you using? (This can be found in Help, About Second Life)
(!) Your computer specs (from Help, About Second Life)
(!) A description of the problem, explaining what you do / see / don't see, including any error messages displayed
(!) Step by step Directions on how to reproduce your experience.
(!) A screenshot of what you are seeing. Please press the Information Provided button when you have added the additional information. Thanks!
Steps to Reproduce
Uploading meshes.
Actual Behavior
There seems to be a bug in the current switch distance formula that causes dimension along the local z axis to be ignored when the switch points are calculated.
This issue may have been there for a while. I noticed as early as November or December 2013 that some tall columns I was making had suprisingly low LOD. I didn't think about it then but recently I've been working on a series of windows that include both low and wide skylights and tall and narrow corner windows and the LOD difference between those became too obvious to ignore.
I suppose the picture says it all. The two windows there are made from the same model and absolutely identical, except that the one to the left was rotated 90 degrees (so the longest side was along the x axis rather than the z axis) before uploading and then rotated back afterwards. In the picture the window to the left is still at high res while the one to the right has switched to mid res model even though it's actually silghtly closer to the camera. The LOD difference between the two is actually so big that the right window switches to low resolution before the left window switched to mid resolution.
Expected Behavior
I expected objects with the same overall dimensions to have the same switch points regadrless of their orientation.
Other information
Attachments
Links
Related
Original Jira Fields
| Field | Value | | ------------- | ------------- | | Issue | BUG-6287 | | Summary | LOD Switch Distance bug? | | Type | Bug | | Priority | Unset | | Status | Been Triaged | | Resolution | Triaged | | Reporter | ChinRey (chinrey) | | Created at | 2014-06-07T10:31:21Z | | Updated at | 2016-10-02T14:18:36Z | ``` { 'Business Unit': ['Platform'], 'Date of First Response': '2014-06-09T10:31:45.570-0500', 'System': 'SL Viewer', 'Target Viewer Version': 'viewer-development', 'What just happened?': "There seems to be a bug in the current switch distance formula that causes dimension along the local z axis to be ignored when the switch points are calculated.\r\n\r\nThis issue may have been there for a while. I noticed as early as November or December 2013 that some tall columns I was making had suprisingly low LOD. I didn't think about it then but recently I've been working on a series of windows that include both low and wide skylights and tall and narrow corner windows and the LOD difference between those became too obvious to ignore.\r\n\r\nI suppose the picture says it all. The two windows there are made from the same model and absolutely identical, except that the one to the left was rotated 90 degrees (so the longest side was along the x axis rather than the z axis) before uploading and then rotated back afterwards. In the picture the window to the left is still at high res while the one to the right has switched to mid res model even though it's actually silghtly closer to the camera. The LOD difference between the two is actually so big that the right window switches to low resolution before the left window switched to mid resolution. ", 'What were you doing when it happened?': 'Uploading meshes.', 'What were you expecting to happen instead?': 'I expected objects with the same overall dimensions to have the same switch points regadrless of their orientation.', } ```