perf(sleepers): the instances of the sleepers and timbers of a track sent to the gpu once, as it's laid, and kept there in a buffer of their own; in each pass the runs of them in sight turned by the modelviews of their cells into the instance buffer of the frame by a compute pass, one dispatch and one barrier a pass, and drawn in one instanced draw for each model, part and level of detail, in place of the cpu turning every instance every frame

world/TrackSleepers.cpp

@@ -125,11 +125,15 @@ std::pair<double, double> shared_zone( TTrack const &Track, int const Path ) {
 
 namespace {
 
+// the instances made so far, of all the tracks
+std::uint64_t makings { 0 };
+
 // the beams as instances of the parts of the model of the track: those of the fastenings with them, or left for later and the beams
 // kept for them meanwhile
 void instance( TTrack &Track, std::vector<beam> &&Beams, bool const Fastened ) {
 
     auto &laid { Track.m_sleepers3d };
+    laid.made = ++makings;
     for( auto &instances : laid.instances ) {
         instances.clear();
     }

world/TrackSleepers.h

@@ -103,6 +103,7 @@ struct laid {
     std::vector<glm::mat4> whole; // transforms of the timbers of a model without parts, drawn whole
     std::vector<glm::mat4> plain; // transforms of the plain sleepers, whole models, relative to the origin of the track
     std::vector<chunk> chunks;
+    std::uint64_t made { 0 }; // which making of the instances they are, none other of any track the same: a copy of them tells by it
     bool fastened { false }; // the fastenings are laid, else they're laid as the track comes near where they're drawn
     std::vector<beam> beams; // kept for the fastenings till they're laid
 };