Portal V
Editorial note, added October 15, 2026. The paragraph and the tool below are the example as published on April 15, 2026, left exactly as written. Four of its claims did not survive the rebuilt study. I’m keeping each one beside its correction, because a historical example that quietly fixes itself is worse than one that admits it was wrong.
- Claim: “IBC caps that distance at 200ft for sprinklered buildings.” Correction: no single number is the rule; even my own later notes disagree (Thesis #41: 200 ft sprinklered “in most occupancy types”; Thesis #22: 100 ft). The limit comes from the adopted code and amendments, the occupancy and the sprinklers, and it is only one test: a complete review also checks occupant loads, exit capacity and exit separation, as the Tennessee Fire Marshal’s IBC-based plan-review guidance makes explicit.
- Claim: the tool measures travel distance. Correction: it measures straight lines to the nearest exit in an empty rectangle. Travel distance is walked: a wall or a locked door changes the path, not the line. That is the argument of Thesis #41, published the next day; the rebuilt study’s cards now read “farthest sampled point” and “samples beyond line.”
- Claim: the red shows what the 200 ft limit means. Correction: the red came entirely from the “Max travel distance (IBC)” slider’s 30 ft default. The largest room the sliders allowed, 100 × 80 ft, has a diagonal of about 128 ft, so a 200 ft line could not turn a single sample red.
- Claim: a second exit “eliminates the red,” and “two exits placed well can cover almost anything.” Correction: it pulls the far corner in, not all the way. In a 60 × 40 ft room against a 30 ft line, an exit on the opposite wall cuts the farthest sampled point from 62.6 ft to 35.4 ft, but the middles of the long walls are still more than 30 ft from either exit.
The rebuilt study’s four examples, each sampled at 2,400 points (exits sit halfway along the short walls):
| Room · exits · line | Farthest sampled point | Samples beyond the line |
|---|---|---|
| 60 × 40 ft · one · 30 ft | 62.6 ft | 54% |
| 60 × 40 ft · opposite pair · 30 ft | 35.4 ft | 8% |
| 120 × 80 ft · opposite pair · 30 ft | 70.7 ft | 70.2% |
| 60 × 40 ft · opposite pair · 20 ft | 35.4 ft | 47.3% |
Doubling the room doubles the distances; tightening the line changes only how they are read. The percentages are shares of sample points, not of occupants, and the sampled distances run slightly under the exact ones (63.2, 36.1 and 72.1 ft for the first three rooms) because the grid measures cell centers, not the boundary.
What survives: exit placement reshapes the distance field, and a second, opposite exit is the cheapest way to pull the far corner in. That is the relationship a building model should carry for a portal: its position, the space it serves and the connections that make it reachable. Moving an exit should immediately change the distance field; adding partitions should change the route graph; the door symbol and schedule are outputs of those relationships. What doesn’t survive: treating a straight-line heat map as a code check.
Exit placement determines how far anyone in the building has to walk to safety. IBC caps that distance at 200ft for sprinklered buildings. This tool shows exactly what that limit means — and where adding a second exit eliminates the red.
Purpose: make the relationship between exit count, room size, and IBC travel distance limits tangible. One exit in a big room fails code. Two exits placed well can cover almost anything.