Currently I’m working with polygons consisting of 246 vertices that move over time. The polygons do not change in size or shape, only position. Currently the project accomplishes this by separate CZML polygons that define the positions of each vertex for every second the polygon needs to appear, like so:
{
"id":"id1",
"polygon":{
// color stuff here
"positions":{
"interval":"2020-09-21T03:53:54.671000000002095Z/2020-09-21T03:53:55.671000000002095Z",
"cartographicDegrees":[
14.763680326900012, 27.029629646718668, 0,
15.04683390517481, 27.080578360029904, 0,
...
]
}
},
{
"id":"id2",
"polygon":{
// color stuff here
"positions":{
"interval":"2020-09-21T03:53:55.671000000002095Z/2020-09-21T03:53:56.671000000002095Z",
"cartographicDegrees":[
14.743772057474061, 27.089709345645197, 0,
15.02706418867925, 27.140718219843077, 0,
...
]
}
},
etc...
This is resulting in very poor performance with only a handful of polygons visible at a time. What’s the best way to improve the performance here? Is there any difference between specifying a polygon for each interval like we do now vs. specifying on polygon and just using references to refer to the dynamic vertices? Any other possible tips I could try?