Version: 2018.2

Description

Medium amount of mesh compression.

Vertex positions, normals/tangents and UVs will be quantized to respectively 16, 8, 10 bits per component. Skinning information will be compressed.

See Also: ModelImporter.meshCompression.