You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Showcase 30 extended with a GPU implicit variant as the new default
GPU marching cubes compute pipeline
Z-axis slab chunking: volumes of any size are split internally into slabs sized to fit device.limits().max_buffer_size; no public API change
upload_volume_for_mc now returns ViewportResult<VolumeGpuId>; returns Err(ViewportError::McBufferTooLarge) when even a single Z-layer of cells exceeds the device limit, allowing callers to fall back to CPU extraction