Research-run intake is paused. No files are accepted by upload or email. The hosted tool is open: Start free trial · Ask a question

Sample: a past research-engine check

Research result, 2026-10-16 · real lot: the municipal sports-complex parking lot in South Dakota · customer details withheld · the check service is paused
Research — Runnax Engine study. This is an example of a Runnax Engine research check that recomputed an AGi32 file. That check is paused and no files are taken. It is not what the hosted tool returns; see “What the hosted tool gives you” below.

In this check the customer's AGi32 file of a parking lot, 15 poles at 25 ft with 75 W LED area lights, was recomputed by Runnax Engine: it read the fixtures, aiming and calculation points straight from the file and recalculated every point on its own, in about 4 seconds.

The numbers held, and the lot met the customer's targets. Runnax Engine matched AGi32 within 0.03 % on every zone. Average illuminance was 1.24 fc against the 1.0 fc target; Avg:Min 3.13 against 3.3. One thing flagged before the plan went out: the light loss factor.

1 · Zone by zone

ZoneAGi32 avg / minRunnax Engine avg / minDifferenceTargetMeets it
Parking area, horizontal at grade (271 points)1.237 / 0.395 fc1.237 / 0.395 fc0.01 %Eavg ≥ 1.0 fc, Avg:Min ≤ 3.3Yes
Vertical at 5 ft, facing 71° (495 points)0.553 / 0.000 fc0.553 / 0.000 fc0.03 %none given–
Vertical at 5 ft, facing 251° (495 points)0.485 / 0.000 fc0.485 / 0.000 fc0.03 %none given–

Foot-candles, direct light. Point by point, the largest difference on the parking area is 0.3 % of its average. The vertical grids in this file run into the corners of the lot, which is why their minimum is 0. Research figures, 2026-10-16.

2 · What the file assumes

Light loss factor 1.0. No lamp or dirt depreciation, so these are initial levels. At a typical 0.9 the average drops to about 1.11 fc: still above 1.0 fc, with less room. Some reviewers ask for the factor to be stated on the plan.

3 · Inputs on record

What the hosted tool gives you

The hosted tool at sim.runnax.com does not read AGi32 files. On an uploaded site image you draw the areas to light, then either place the poles or draw where poles may and may not go and let it search for the fewest new poles (it says whether every layout with fewer poles was tried); upload or choose luminaire IES files and enter the illuminance requirement; it calculates point-by-point horizontal illuminance for every area, tries the candidate luminaires and aimings you allow, and returns the combination that meets the requirement, or says none does. You get a report page, a PDF and a download package. Checked against AGi32: 473 of 473 points within 0.13 %, horizontal illuminance on given poles, a research run of our engine on a real parking-lot project, not yet repeated in the hosted tool; reported, not independently reviewed (Verification). Results are for evaluation, not a field measurement or a standards certification.

Direct light only. Not a code-compliance statement and not a construction approval; the engineer of record signs the design.

Start free trial → · Ask a question