Skip to content
This repository has been archived by the owner on Mar 28, 2024. It is now read-only.

[BUG-231904] Joint offsets not applying consistently on the same collada file #9276

Closed
2 tasks
sl-service-account opened this issue Mar 9, 2022 · 7 comments
Closed
2 tasks

Comments

@sl-service-account
Copy link

sl-service-account commented Mar 9, 2022

What just happened?

This tail attachment was made for a dinkie cat. It uses different tail joints from the actual cat. In tests it's always uploaded with joints in the Rigging tab of the uploader and also looks ok in the uploader when the joint offsets visualiser checkbox is ticked. However...it is pot luck whether the joints actually get applied or not. So I did some tests as I really need this tail. Always joint offsets is enabled. I'm reiterating that in case you think I am forgetting to tick it.

What were you doing when it happened?

Uploading a tail

What were you expecting to happen instead?

I expected the tail to be short and match the joints of the tail rig it was made on and not default to original tail bones on the cat rig

Other information

These are my test results for the exact same collada file each time (attached).
Beta Grid
good
bad
bad
good
good
bad
good
good
bad
bad
good
good

Main grid
bad
bad
bad
bad
bad
bad
good <----- successful upload

It seems like it's completely random whether the joints get applied or not when worn. I have tried both Firestorm and Second Life viewers

Attachments

Links

Related

Duplicates

Original Jira Fields
Field Value
Issue BUG-231904
Summary Joint offsets not applying consistently on the same collada file
Type Bug
Priority Unset
Status Closed
Resolution Duplicate
Reporter Etheria Parrott (etheria.parrott)
Created at 2022-03-09T19:17:15Z
Updated at 2022-04-29T23:31:17Z
{
  'Build Id': 'unset',
  'Business Unit': ['Platform'],
  'Date of First Response': '2022-03-09T14:51:48.056-0600',
  "Is there anything you'd like to add?": "These are my test results for the exact same collada file each time (attached).\r\nBeta Grid\r\ngood\r\nbad\r\nbad\r\ngood\r\ngood\r\nbad\r\ngood\r\ngood\r\nbad\r\nbad\r\ngood\r\ngood\r\n\r\nMain grid\r\nbad\r\nbad\r\nbad\r\nbad\r\nbad\r\nbad\r\ngood <----- successful upload\r\n\r\nIt seems like it's completely random whether the joints get applied or not when worn. I have tried both Firestorm and Second Life viewers",
  'ReOpened Count': 0.0,
  'Severity': 'Unset',
  'System': 'SL Viewer',
  'Target Viewer Version': 'viewer-development',
  'What just happened?': "This tail attachment was made for a dinkie cat. It uses different tail joints from the actual cat. In tests it's always uploaded with joints in the Rigging tab of the uploader and also looks ok in the uploader when the joint offsets visualiser checkbox is ticked. However...it is pot luck whether the joints actually get applied or not. So I did some tests as I really need this tail. Always joint offsets is enabled. I'm reiterating that in case you think I am forgetting to tick it.",
  'What were you doing when it happened?': 'Uploading a tail ',
  'What were you expecting to happen instead?': 'I expected the tail to be short and match the joints of the tail rig it was made on and not default to original tail bones on the cat rig\r\n',
  'Where': 'Beta Grid: Mesh Sandbox 2\r\nMain Grid: Rain Cloud',
}
@sl-service-account
Copy link
Author

Bloodsong Termagant commented at 2022-03-09T20:51:48Z, updated at 2022-03-09T20:58:19Z

it sounds like the 'random' rig dominance thing going on.  ie: https://jira.secondlife.com/browse/BUG-231579

 

this is a sterling example of why 'random' rig dominance is BAD.  creators wasting hours trying to 'fix' broken rigs.  customers complaining when it doesn't work, but only for them...  sometimes it does, sometimes it doesn't....  no one can find a pattern to solve this problem.

 

@sl-service-account
Copy link
Author

Lucia Nightfire commented at 2022-03-09T20:58:10Z

Is there a reason why mPelvis was included?

Wouldn't it be fighting your scaled rig deform?

@sl-service-account
Copy link
Author

Etheria Parrott commented at 2022-03-09T22:52:42Z, updated at 2022-03-09T23:13:11Z

I add mPelvis to anchor the tail so it doesn't pop out when the dinkie sits down. I've always done this with tails I make for them. I have lots of varieties. Lion, pig, cow, devil and so on. Really a lot but that would be probably a year ago or more

 

 

@sl-service-account
Copy link
Author

Etheria Parrott commented at 2022-03-17T17:00:53Z

Looking at the linked Bug report, this has been regarded as a feature request. It used to work so I think it's a bug. Could someone take a deeper look please? The unpredictability at upload is not fair on our pockets or our time.

@sl-service-account
Copy link
Author

Etheria Parrott commented at 2022-03-29T13:26:15Z

Can we have an update on the investigation into this bug please?

@sl-service-account
Copy link
Author

Dan Linden commented at 2022-03-29T18:35:02Z

Hi Etheria.

Going with the supposition that dominance is determined based on the mesh id and that the lowest mesh id wins, then if the UUID of your dinkie outfit happens to be low, then there's a small chance that the UUID of the tail you upload will be lower and beat it.

I need to have the same mesh you are currently wearing before wearing the tail. Where can I get a copy of the dinkie outfit you are wearing on the Agni grid (or Aditi grid, if you prefer)?

@sl-service-account
Copy link
Author

Dan Linden commented at 2022-04-29T23:30:54Z

Hi Etheria.

Thank you for the sample dinkie animesh.

The "Dinkie cat body v1.9 rev4" asset id starts with b, which is 11 on the hexadecimal scale of 0 to 15, so rather high. Most of the tails you upload will be lower and thus look broken. I uploaded several tails and confirmed that those with asset ids higher than b worked correctly.

"Dinkie cat body v1.9 rev2" starts with 4, so has a better chance to be lower than most of the tail assets you upload.

Ideally, the asset for the "Dinkie cat body" should be 0 to save you from needing many uploads for the tail. You can see that asset id by watching the cache folder while you attach the Dinkie cat body for the first time (or after a cache clear).

This is not a fix, but only a confirmation that this is the same issue as BUG-231579.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

1 participant