Free, open-source 3D sculpting for Linux, macOS and Windows: dense meshes, remeshing, painting, built-in AI (image/text to 3D), Send to Unreal, game-ready LODs, Gaussian splats and OpenUSD.
C++ 3d3d-sculptingcppdear-imguidigital-sculptingfbxgame-developmentgaussian-splattinggltfimage-to-3dmcpopenglopenusdsculptingtext-to-3dunreal-engine
Merge remote-tracking branch 'origin/main' into perf/an-gated
islandmedd8fdf2ca22d1
diff --cc src/app/App.cppindex 16e8e488,894382cd..212a75e5--- a/src/app/App.cpp+++ b/src/app/App.cpp@@@ -1948,15 -2041,7 +2042,16 @@@ void App::beginStroke(float x, float yhaveLastDab_ = false;doc_.sculptor.draftStrokes = draftStrokes() && asyncStrokeAllowed(); // (AsyncStroke.cpp)doc_.sculptor.adaptiveDabs = fastBigBrushes();+ // Big brushes where the CPU has room: the exact normals the analytic ones defer land at the pen-up. On a 4-core+ // floor machine that cost more than it saved (E-core floor, 120 px release 0.5 -> 1.3 s), and at 120 px even on+ // the 3060 (release +190 ms for -45 ms lag); from 160 px on 6+ threads it wins (3060 240 px p50 428 -> 283 ms,+ // release +50). SF_ANALYTIC_NORMALS=0/1 forces it (A/B).+ static const int analyticNormals = [] {+ if (const char* e = std::getenv("SF_ANALYTIC_NORMALS")) return e[0] != '0' ? 1 : 0;+ return !simulatedFloorTier() && std::thread::hardware_concurrency() >= 6 ? -1 : 0; // -1: by brush size+ }();+ doc_.sculptor.analyticNormals = analyticNormals == 1 || (analyticNormals == -1 && radiusPx_ >= 160.0f);+ fusedDabs_ = 64.0;if (!doc_.sculptor.beginStroke(doc_.mesh,!strokeSculpt_ ? std::string("Paint") : recipe ? recipe->name : brushLabel(strokeBrush_),strokeSculpt_ ? doc_.activeLayer : -1)) {