# Question about geometricError in tileset.json

**URL:** <https://community.cesium.com/t/question-about-geometricerror-in-tileset-json/7933>\
**Category:** CesiumJS\
**Created:** [January 24, 2019, 9:42am UTC](https://community.cesium.com/t/question-about-geometricerror-in-tileset-json/7933 "2019-01-24T09:42:15Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![YAO\_Y](https://avatars.discourse-cdn.com/v4/letter/y/a587f6/32.png) [@YAO\_Y](https://community.cesium.com/u/YAO_Y)\
**Post date:** [January 24, 2019, 9:42am UTC](https://community.cesium.com/t/question-about-geometricerror-in-tileset-json/7933/1 "2019-01-24T09:42:15Z")

</div>

**1. A concise explanation of the problem you’re experiencing.**

Hi, recently I uploaded some model to the Cesium Ion and got the output 3dtiles. Since the tiles are hirachical and have children tiles, the geometricError should be consistently smaller and smaller from the root tile to the leaf tiles. If my understanding of the tileset and geometricError was right, then there may be something wrong about the geometricError in the tileset.json of the output 3dtile which contains some middle level tiles have smaller geometricErrors than their descendants. Is that reasonable? Also, I tried to load the 3dtiles and found some tiles were skipped when zooming in and out (shown in the screenshots attached). Can someone help me or explain this? Many thanks! The assetId is 14533 for you information.

screenshot 1: before zooming in.

 ![微信图片_20190124173810.jpg](https://global.discourse-cdn.com/cesium/original/1X/3a6a199e40554c60ff20774d4e190c02671cf8c4.jpeg)

screenshot 2: after zooming in with very little step, the level 4 tile becomes level 7/8/9 tiles immediately. Seems the level 6 was skipped entirely.

 ![微信图片_20190124173814.jpg](https://global.discourse-cdn.com/cesium/original/1X/85fc0a73a9887c5aa03d264625d3261995466374.jpeg)

**2. A minimal code example. If you’ve found a bug, this helps us reproduce and repair it.**

var tileset = viewer.scene.primitives.add(

new Cesium.Cesium3DTileset({

url: Cesium.IonResource.fromAssetId(14533)

})

);

**3. Context. Why do you need to do this? We might know a better way to accomplish your goal.**

**4. The Cesium version you’re using, your operating system and browser.**

Version 1.52.

---

<div class="post-metadata">

**Author:** ![omar](https://sea2.discourse-cdn.com/cesium/user_avatar/community.cesium.com/omar/32/1999_2.png) [@omar](https://community.cesium.com/u/omar)\
**Post date:** [January 25, 2019, 4:07pm UTC](https://community.cesium.com/t/question-about-geometricerror-in-tileset-json/7933/2 "2019-01-25T16:07:57Z")

</div>

The skipping is indeed a feature, not a bug! See [tileset.skipLevelOfDetail](https://cesiumjs.org/Cesium/Build/Documentation/Cesium3DTileset.html#skipLevelOfDetail). The idea is that if you’re already zoomed in close, instead of spending time requesting every child, it can skip to the one with the required resolution.

For the geometric error issue, you are correct in that the geometric error of a parent should always be greater than its children, so this sounds like a bug. I’ve open an issue for it:

> <https://github.com/CesiumGS/cesium-ion-community/issues/100>
>
> The tileset with asset ID 14533 produces geometric errors that are not strictly decreasing down the tree. See this forum thread...

If you can add more detail to it that would be great! Do you know which tiles have a greater error than their children?

---

<div class="post-metadata">

**Author:** ![YAO\_Y](https://avatars.discourse-cdn.com/v4/letter/y/a587f6/32.png) [@YAO\_Y](https://community.cesium.com/u/YAO_Y)\
**Post date:** [January 28, 2019, 1:38am UTC](https://community.cesium.com/t/question-about-geometricerror-in-tileset-json/7933/3 "2019-01-28T01:38:11Z")

</div>

Thank you for the reply! And if you open the tileset.json tree and looking down at every first child. you’ll find the geometricError from the root node to the last one is 1141.566 -\> 13.651 -\> 10.765 -\> 8.573 -\> 10.384 -\> 19.649 -\> 0.