Cesium Pipeline Producing slow unusable data for big models

Cesium ion
1

Hi,

I’m using the Cesium In house pipeline and it’s producing big unusable data, could you show me how you would create this Melbourne Model using the Cesium Pipeline:

At the moment I’m using: .\model-tiler.exe -I PHOTOGRAMMETRY --input-list ‘C:\file_list.txt’ -o ‘C:\output’ --texture-format WEBP --compression-level 8 --gzip

With a list of OBJ’s and the tileset.json is coming out at 40 MB’s and the model is loading slow.

2

Hi Oliver,

With the most recent on-premise tiler update (v4.2.2), we added the ability to split the tileset.json file into smaller files for better loading time. You can see the new changes we have made by going to the downloads page and clicking on “What’s new?” under the 3D Tiling Pipeline section.

The splitting feature is on by default, but you can modify the number of tiles to split at with the flag decimation.tilesPerFile. The default value is 1000.

3

Hi @dzung , I know this was a while ago - but I can’t seem to find any documentation for the decimation.tilesPerFile flag within the on-premise documentation. Has this been removed?

4

Hi @TomPovey - The option is hidden as an advance option for most users. You can show this using -h --show-hidden.

5

That’s great, thanks for the quick reply Shehzan