mirror of
https://github.com/Kelsidavis/WoWee.git
synced 2026-05-06 09:03:52 +00:00
c9c4a15e9a
Log showed: models=2, instances=3, draws=0 — models loaded and instances created but zero draw calls. The M2 renderer skips instances where cachedIsValid is false, which depends on the GPU vertex buffer being valid after upload. The per-model waitAllUploads/pollUploadBatches calls inside the loading loop may have corrupted the upload batch context (started at beginUploadBatch but flushed mid-loop). Now all models upload in a single batch with one final waitAllUploads+pollUploadBatches at the end of rebuildObjects. |
||
|---|---|---|
| .. | ||
| asset_extract | ||
| auth_login_probe | ||
| auth_probe | ||
| blp_convert | ||
| dbc_to_csv | ||
| editor | ||
| asset_pipeline_gui.py | ||
| backup_assets.sh | ||
| diff_classic_turtle_opcodes.py | ||
| gen_opcode_registry.py | ||
| generate_ffx_sdk_vk_permutations.sh | ||
| m2_viewer.py | ||
| opcode_map_utils.py | ||
| validate_opcode_maps.py | ||