FAQs for Terrain datasets
-
Question:
What triggers the terrain to be updated on the fly when, for instance, adding a new breakline?
Answer:
Editing referenced data inside an edit session creates tile-based dirty areas in the terrain. It doesn't rebuild automatically; you have to execute the process. The Terrain Properties dialog box in the Catalog window and the Layer Properties dialog box both report whether a terrain needs to be built. There's also a dirty area renderer available for the layer, accessible from the Symbology tab. After editing participating feature classes, run the Build Terrain geoprocessing tool or use the Build button on the Update tab of the Terrain Properties dialog box in the Catalog window.
-
Question:
How can I keep the terrain's current surface in ArcMap updated based on extent changes when the layer is turned off? This is in support of using interactive analysis/editing tools on the terrain while displaying things other than the terrain.
Answer:
Keep the terrain layer turned on, but turn off the layer renderers through the Symbology tab.
-
Question:
How can I copy a feature class that participates in a terrain dataset? The user interface does not permit it. You must copy the terrain to get another feature class.
Answer:
Use Export.
-
Question:
How do I convert 2D features to 3D?
Answer:
Use the Interpolate Shape tool to generate a 3D feature class from a 2D feature class and the terrain surface.
-
Question:
Does build performance for a terrain dataset improve when a multipoint feature class is embedded?
Answer:
Using embedded feature classes does not improve performance. Embedded feature classes offer two primary benefits: help reduce storage requirements, since you can remove the source multipoint feature class after the terrain is built, and ensure BLOB-based attribute values are properly handled relative to edits/updates.
-
Question:
If I append a multipoint feature class to a terrain dataset using the Add Points To Terrain geoprocessing tool, will the binary attributes also be appended?
Answer:
Yes.
-
Question:
How many points can the file geodatabase hold efficiently?
Answer:
File geodatabases can store points reaching levels in the billions for display and analysis of terrains datasets. However, consider using ArcSDE databases for advanced editing which can also store billions of points.
-
Question:
Is ArcSDE versioning on a massive dataset supposed to increase performance?
Answer:
No. It is for improved editing workflows and what if scenarios.
-
Question:
Are there any special considerations, specific to terrains, when working with SQL databases, for example, tuning or configuration?
Answer:
No.
-
Question:
Does embedding multipoints have any impact on versioning/unregistering in ArcSDE?
Answer:
No. You only need to be aware of conflict resolution behavior for embedded data when reconciling between different versions.
-
Question:
Why are terrain datasets not supported in ArcScene?
Answer:
ArcScene requires that the elevation source data reside in memory. Terrain datasets utilize an on-the-fly query and display structure to work efficiently with millions or billions of points. Consider converting a terrain or portion of a terrain to either a TIN or raster-based surface for use in ArcScene.
-
Question:
When building a terrain, there is a changing fraction value to indicate progress. What determines the size of the denominator displayed in the progress window?
Answer:
The fraction indicates the number of tiles that will be generated and built for the terrain dataset. The number of tiles and size of each tile is determined by the given point spacing and the extents of the data area.
-
Question:
What does each phase indicate during the build process?
Answer:
The first phase scans through all tiles. The tiles whose extents are completely covered by data have pyramid levels built. During this first phase, it is determined which tiles are completely or partially void of data. The second phase composites and builds pyramid levels for empty tiles. The third phase builds the partial tiles.