Polyline width doesn't work properly

CesiumJS
1

Example:

Solution?

2

The inconsistency in width of the polyline is caused by the sharp turns. To fix it, you can change the point position so that the turns are less sharp, for example.

 cartographicDegrees: [44.264969, 34.835854, 0, 32.164949 ,37.913873 , 0, 37.617014, 40.458882, 0,36.694637, 41.815041, 0],
3

They are unfortunately not the same geometries, difference between points makes more than 100 km difference. If you consider it as borders of some farm or country this difference is too much to omit. It doesn’t resolve the issue. I am not so into webgl vertex shaders but I believe resolution can be at cesium side at the point which calculates triangulation of line at intersection of 2 lines. Cesium seems to be forcing to linewidth at all points which creates exampled anomaly. In real life cases you cannot ignore more than cms of errors.

Solution can be like in these blogs, they don’t force intersections to be at same width:

4

Thank you for pointing this out and for the detailed explanation. You can record this issue in the CesiumJS github repository! https://github.com/CesiumGS/cesium/issues

5

Yeah, have been doing it but they say you need to write issues to community forum. As a group, could you make up your minds and decide on where to write the issues? Check recent responses on cesium github from mr. amatoromo and omar something please. Seems like they don’t want to see bug reports in github repo.

6

Can you follow up with the link to the issue that you wrote on the CesiumJS repo? I can’t seem to find it.

7

No, I cannot :slight_smile: Few weeks ago I’ve written a bug as issue there and they were like “it’s for community forums not for here” and I was like “it is a bug” and they were like “it is for community forum”. Now I’ve written here and you are like “it is for github issues” and I am like “what the?”. Is it just to waste my time? Here, the issue is clear and all details were written. I don’t want to go there come here go there again. I guess when an issue seems to be hard to fix, that happens.

8

My apologies for the confusion. We actually have a Github issue to track this problem here: https://github.com/CesiumGS/cesium/issues/5231

Currently we don’t have an estimated timeline for fixing this issue, but if you have ideas for how to do it, we would love to see a pull request or your comments on the issue thread.