-
Notifications
You must be signed in to change notification settings - Fork 3.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Model matrix not getting applied to tileset correctly #7036
Comments
The reason for the bad positions in the second image is The summary here is |
Same issue in this forum thread, retiling with Cesium ion fixes it. |
|
@engineerhe just to narrow down the issue here, can you confirm if tiling your source data with Cesium ion produces a tileset that works as expected? |
@OmarShehata,Thanks for your patient answer,I followed the steps below to solve the problem. |
@OmarShehata,i have solved it . Thanks! |
@engineerhe can you describe what the solution was? |
Sorry, I didn't see the message until now.The solution is very sample , you just switch obj23dtiles tool into branch 2.0. |
|
This bug goes back to at least 1.42.
The
modelMatrix
property seems to have some weird interaction with the ContextCapture tileset posted here. The tileset uses theCESIUM_RTC
extension but is in local coordinates.Default view at center of the Earth:
With model matrix applied
Debugging Sandcastle demo
The good thing is when upgrading the tileset with 3d-tiles-tools (links to the
2.0-tools
branch) the tileset is correct. Also needs to be running on a branch that combines #7035 and #6805. Hopefully it is easy to track down.Debugging Sandcastle demo
The text was updated successfully, but these errors were encountered: