eu07/maszyna
17
0
Fork 0
mirror of https://github.com/MaSzyna-EU07/maszyna.git synced 2026-11-02 23:16:22 +01:00
Code Issues 157 Wiki Activity

perf(gl46): leave out instance buckets before the culling, lod bounds once per model, shared walks of the shadow cascades

Browse Source
This commit is contained in:
Hirek
2026-10-22 20:13:25 +02:00
parent b343e98e22
commit 5d3f857c39
7 changed files with 250 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
scene/scene.h
View File

@@ -269,6 +269,13 @@ public:
mutable glm::dvec3 recordorigin { 0.0 }; // the records hold locations relative to it
mutable float lodrangesquared { 0.f }; // past it no submodel of the model of the bucket is drawn
mutable float rangesquared { std::numeric_limits<float>::max() }; // past it no instance of the bucket is drawn: the largest range of its instances, within lodrangesquared
// bounds of the visible instances of the records, for the tests of a whole bucket: a sphere around their locations
// (recordcenter, recordlocationradius), the radius of the sphere which holds the instances whole (negative when none
// of them is visible), and the largest radius of an instance
mutable glm::dvec3 recordcenter { 0.0 };
mutable float recordlocationradius { 0.f };
mutable float recordradius { -1.f };
mutable float recordradiusmax { 0.f };
mutable unsigned int recordsversion { 0 };
// place of the records in the record buffer of the gpu culling: first record, room, and the generation of
// the buffer it is good for. gpustale: the records changed since they were sent