StrafesNET/strafe-project
6
Fork 4
Code Issues Pull Requests 1 Packages Projects Releases 2 Wiki Activity

comment about bad algorithm

Browse Source
This commit is contained in:
Quaternions 2025-01-16 07:35:02 -08:00
parent 57cc49dc1a
commit 28499800cb
1 changed files with 14 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
strafe-client/src/model_physics.rs
View File

@ -704,6 +704,20 @@ impl MinkowskiMesh<'_>{
}, },
} }
} }
// TODO: fundamentally improve this algorithm.
// All it needs to do is find the closest point on the mesh
// and return the FEV which the point resides on.
//
// What it actually does is use the above functions to trace a ray in from infinity,
// crawling the closest point along the mesh surface until the ray reaches
// the starting point to discover the final FEV.
//
// The actual collision prediction probably does a single test
// and then immediately returns with 0 FEV transitions on average,
// because of the strict time_limit constraint.
//
// Most of the calculation time is just calculating the starting point
// for the "actual" crawling algorithm below (predict_collision_{in|out}).
fn closest_fev_not_inside(&self,mut infinity_body:Body)->Option<FEV<MinkowskiMesh>>{ fn closest_fev_not_inside(&self,mut infinity_body:Body)->Option<FEV<MinkowskiMesh>>{
infinity_body.infinity_dir().map_or(None,|dir|{ infinity_body.infinity_dir().map_or(None,|dir|{
let infinity_fev=self.infinity_fev(-dir,infinity_body.position); let infinity_fev=self.infinity_fev(-dir,infinity_body.position);