Dion Moult 51dc31a50b wgpu backend: 10× perf — megadraw, async HiZ, parallel cull, motion mode
Closes the perf gap to the GL backend on real BIM benchmarks. On a 10-
sidecar / 380k-instance corpus at a fixed --camera the wgpu binary went
from 110.6 ms to 11.6 ms (vs GL's 23 ms — half the frame time, but
note GL is doing extra work the wgpu backend hasn't ported yet; see
the caveats list at the bottom). Bundled because the pieces interlock
and shipping any of them without the others reintroduces the same wall.

1. Cross-mesh vertex pulling (single mega-draw per model)
   The previous one-drawIndexed-per-(mesh × LOD-bucket) loop was costing
   ~13ms on a 27k-mesh scene. CPU now emits a flat visible_draws[]
   (16 B per visible (mesh,lod,instance)) plus a prefix_sums[] table.
   WGSL binary-searches prefix_sums by @builtin(vertex_index) to find
   the entry, then manually fetches the mesh-local index from a
   storage-bound indices[] and pulls the packed 12 B vertex. No
   setIndexBuffer; the shader reads everything from storage. Bind
   group grew from 4 to 7 entries (vertices, meshes, instances,
   indices, visible_draws, prefix_sums, per-model uniform) — well
   under WebGPU's mandatory 8 storage / 12 uniform floor.

2. Async HiZ readback via ping-pong staging buffers
   Sync wait via wgpuInstanceProcessEvents was costing ~37 ms on a
   real scene (GPU drain). Two staging slots now ping-pong: frame N
   kicks a non-blocking mapAsync on slot K, frame N+1's first action
   is one processEvents drain. Pyramid is 1-2 frames stale — matches
   the "slightly-stale depth, fine" pattern the GL backend already
   documents. encodeHizResolve returns -1 (skip) if both slots are
   in flight; cull keeps using the most recent pyramid.

3. Cull reorder: contribution before HiZ
   HiZ projection is ~10× more expensive than the contribution
   check, yet most contribution-survivors would be HiZ-rejected
   anyway on dense scenes. Computing projected_px first lets
   contribution short-circuit ~80% of HiZ tests with no rejection-
   quality loss. Saved ~34 ms on the dense bench.

4. Motion-mode contribution threshold
   AppSettings::motionMinPixelRadius parity. While the camera is
   changing (orbit/pan/zoom/--benchmark sweep), drop instances
   below 10 px instead of 2 px. Halves visible_objects during
   motion with no perceived quality loss.

5. Parallel cull (std::async across models)
   Per-model cullModelCpu split into Compute (CPU-only, thread-safe)
   + Upload (main-thread wgpu queue writes). std::async fan-outs the
   compute across models; main-thread joins and uploads. Wall-clock
   cull on the 10-model corpus drops from ~17 ms single-threaded to
   ~9 ms across cores.

6. --no-hiz CLI flag + per-phase benchmark timings
   Benchmark now also prints "per-frame avg ms: cull=X
   hiz_readback=Y" so future regressions can be attributed without
   guesswork. --no-hiz toggles the master switch from the CLI.

Honest caveats — wgpu is currently faster mostly because GL is doing
work we haven't ported yet:
  - Edge silhouette pass (stage 9) will add ~3-5 ms back to wgpu.
  - GL's HiZ uses the BVH so it rejects whole subtrees (1.7k vs
    our 358 rejects on the same scene). BVH for HiZ is future work
    (task #13 / a new task) — until then we draw more sub-pixel
    geometry that's behind closer surfaces. Visually correct, perf
    cost paid. Stage 4+5 are unaffected.

Verified pixel-identical on basic.ifc through every change. Real-scene
visual diff against GL pending the --screenshot flag on the GL minimal
(task #10's other half).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-27 19:46:13 +10:00
2026-05-25 16:34:19 +10:00
2025-12-19 18:53:14 +05:00
2026-05-08 16:20:26 +02:00
2026-01-26 17:10:03 +05:00
2026-05-18 19:24:25 +02:00
2023-09-14 09:29:20 +02:00
2026-02-18 09:09:46 +11:00
2026-05-08 16:20:26 +02:00
2026-05-08 16:20:26 +02:00
2026-05-08 16:20:26 +02:00
2026-05-08 16:20:26 +02:00
2026-04-13 19:38:25 +05:00
2026-04-13 19:40:25 +05:00

IfcOpenShell

IfcOpenShell is an open source (LGPL) software library for working with Industry Foundation Classes (IFC). Complete parsing support is provided for IFC2x3 TC1, IFC4 Add2 TC1, IFC4x1, IFC4x2, and IFC4x3 Add2. Extensive geometric support is implemented for the IFC releases IFC2x3 TC1 and IFC4 Add2 TC1. Extending with support for arbitrary IFC schemas is possible at compile-time when using C++ and at run-time when using Python.

In addition to a C++ and Python API, IfcOpenShell comes with an ecosystem of tools, notably including IfcConvert (an application to convert IFC models to other formats), Bonsai (an add-on to Blender providing a graphical IFC authoring platform), and many other libraries, CLI apps, and more. Support is also provided for auxiliary standards such as BCF and IDS.

For more information, see:

Development is sponsored through your generous donations!

Open Collective Contributors

Contents

Name Description License Service
bcf Library to read and write BCF-XML and query OpenCDE BCF-API modules LGPL-3.0-or-later PyPI Anaconda-Server Badge
bonsai Add-on to Blender providing a graphical native IFC authoring platform GPL-3.0-or-later Official GitHub Unstable Chocolatey
bsdd Library to query the bSDD API LGPL-3.0-or-later PyPI
ifc2ca Utility to convert IFC structural analysis models to Code_Aster LGPL-3.0-or-later
ifc4d Convert to and from IFC and project management software LGPL-3.0-or-later PyPI
ifc5d Report and optimise cost information from IFC LGPL-3.0-or-later PyPI
ifcbimtester Wrapper for Gherkin based unit testing for IFC models LGPL-3.0-or-later
ifcblender Historic Blender IFC import add-on LGPL-3.0-or-later*
ifccityjson Convert CityJSON to IFC LGPL-3.0-or-later PyPI
ifcclash Clash detection library and CLI app LGPL-3.0-or-later PyPI
ifcconvert CLI app to convert IFC to many other formats LGPL-3.0-or-later* Official GitHub
ifccsv Library and CLI app to export and import schedules from IFC LGPL-3.0-or-later PyPI
ifcdiff Compare changes between IFC models LGPL-3.0-or-later PyPI
ifcedit CLI wrapper for ifcopenshell.api IFC model mutation functions LGPL-3.0-or-later PyPI
ifcfm Extract IFC data for FM handover requirements LGPL-3.0-or-later PyPI
ifcmax Historic extension for IFC support in 3DS Max LGPL-3.0-or-later* Official
ifcmcp MCP server for querying and editing IFC building models LGPL-3.0-or-later PyPI
ifcopenshell-python Python library for IFC manipulation LGPL-3.0-or-later* Official GitHub PyPI Anaconda Anaconda Docker AUR AUR Unstable Pyodide WASM Wheels tag
ifcpatch Utility to run pre-packaged scripts to manipulate IFCs LGPL-3.0-or-later PyPI
ifcquery CLI tool for querying and inspecting IFC building models LGPL-3.0-or-later PyPI
ifcsverchok Blender Add-on for visual node programming with IFC GPL-3.0-or-later GitHub
ifctester Library, CLI and webapp for IDS model auditing LGPL-3.0-or-later PyPI

The IfcOpenShell C++ codebase is split into multiple interal libraries:

Name Description License
ifcgeom Internal library for IfcOpenShell LGPL-3.0-or-later*
ifcgeom_schema_agnostic Internal library for IfcOpenShell LGPL-3.0-or-later*
ifcgeomserver Internal library for IfcOpenShell LGPL-3.0-or-later*
ifcjni Internal library for IfcOpenShell LGPL-3.0-or-later*
ifcparse Internal library for IfcOpenShell LGPL-3.0-or-later*
ifcwrap Internal library for IfcOpenShell LGPL-3.0-or-later*
qtviewer Internal library for IfcOpenShell LGPL-3.0-or-later*
serializers Internal library for IfcOpenShell LGPL-3.0-or-later*
S
Description
No description provided
Readme 524 MiB
Languages
C++ 57.4%
Python 36.2%
C 3.9%
Gherkin 0.7%
JavaScript 0.3%
Other 1.3%