Distribution-centre sequence: text description

The silent, 30-second video shows three labeled parcels moving through a fictional distribution centre. P01 travels to outbound lane D01, P02 to reject lane R01, and P03 to outbound lane D02. P03 pauses in an accumulation area before continuing. The raised orange arrows show intended routes; they do not contact or push the parcels.

How the screen is arranged

The upper-left panel is a wide Blender Cycles view derived from the reopened USD. The lower-left panel is a synchronized plan diagram with parcel IDs, the scan plane and destination lanes. That diagram omits the scanner shell so a parcel stays visible at the crossing. It is explanatory artwork, not a view through the actual scanner enclosure.

The right panel gives the time, parcel destinations and positions, the active amber/red/green lamp, route-indicator angles and the most recent measured event. Parcel IDs and state labels supplement color. The film uses Cycles; the article’s before/after stills use NVIDIA RTX.

Events in the repaired scene

Scroll sideways for more columns on a small screen.

These times measure parcel-centre crossings of declared planes, estimated within authored-frame brackets. They are not first-contact times or a simulation of physical handling.

P03 holds from 12 to 15 seconds and has moved one metre onward by 16 seconds. The status tower shows amber before 12 seconds, red from 12 until 15 seconds, then green. At the video’s 13.5-second view, P01 is moving into D01, P02 is continuing along the main conveyor and P03 is held. The panel identifies the hold in text and shows the red lamp.

The route-indicator repair

In the earlier delivery, DIV01 remained angled toward D01 while P02 passed straight through its junction, approximately 14.755–15.791 seconds. In V1 the indicator returns to straight before P02’s passage and holds that state until P02 clears. The matched article stills show the old rectangular vane and the revised arrow at 15.273 seconds. V1 changes both the shape and the schedule. The parcel’s successful destination timing alone had not exposed this route-state concern.

Evidence for this description

The event values come from the saved V1 scripted results. The parcel hold and lamp states agree with the retained producer state record; the finding and repair ledger identifies the separate geometry and visual reviews. The recorded review compared all 721 supplied state rows with USD readback and inspected ten decoded frames. That supports the supplied states and sampled views; it does not mean every video frame received visual approval.

Full scene deliveries and executable trial inputs are not included. See the runtime and reproduction scope.