Local civil time
The wall-clock time recorded at the birthplace.
BaZi method · evidence guide
Time zones, solar terms, true solar time, and Four Pillars boundaries
Two BaZi calculators can receive what looks like the same clock input and still disagree because the hidden assumptions around that clock may differ. Sibyl exposes those assumptions instead of pretending every disagreement has one universal answer.
The wall-clock time recorded at the birthplace.
The unambiguous global instant after civil-time resolution.
The local clock time adjusted for longitude and the difference between clock time and solar time.
The solar-term moment used to determine the Year and Month Pillars.
Sibyl compares the resolved birth instant in UTC with the solar-term instant in UTC for year and month boundaries. A longitude-adjusted true-solar wall clock is not compared directly with a UTC solar-term instant.
An invalid calendar date fails. A DST gap is a local time that never occurred and must fail rather than silently slide forward. A DST overlap can describe two different global instants, so Sibyl exposes the earlier and later occurrence for explicit selection.
Historical clock offsets can differ from today’s. Sibyl currently uses a 1970–2038 coverage window for its support labels; this does not guarantee historical timezone accuracy within that window. The runtime tzdata edition is unverified, including for dates inside the window. Outside it, support is marked provisional rather than treated as proof of accuracy; outside-window dates may still yield a provisional result. Check the original local time records when the distinction matters.
Longitude correction can move the displayed true-solar wall time across civil midnight. That can affect the Day or Hour Pillar while the original civil record and resolved global instant remain distinct evidence. Sibyl uses its existing equation-of-time approximation with longitude correction. This is not NOAA’s algorithm and does not provide a high-precision guarantee.
The same Li Chun instant applies globally. Shanghai, Tokyo, New York, and London display different local clock times for that event, but the underlying instant is the same. Sibyl's year and month boundary logic works from that global instant. Its solar-term values come from lunar-javascript 1.7.7, with adapter wall times interpreted in Asia/Shanghai. This evidence is adapter-derived and provisional. HKO-based tests check selected reference cases, not every date or event returned by the adapter.
Sibyl rounds the absolute distance to the nearest whole minute and asks for review when the rounded result is at most five minutes. A distance of 5 minutes 29 seconds rounds to five; 5 minutes 30 seconds rounds to six. This is a Sibyl product review rule, not an external accuracy guarantee or a strict 300-second threshold. When review is required, Sibyl withholds a definite chart rather than guess which side of the boundary applies.
Sibyl treats Zi-hour as 23:00–01:00 and changes the Day Pillar at midnight in the adjusted local solar time. The hour pillar follows the same selected day and two-hour branch rules. These are selected traditional and product conventions; IANA and HKO do not establish these BaZi rules. Another calculator may choose a different rollover convention.
These references do not audit Sibyl, certify its adapter or establish traditional BaZi interpretations. The coverage labels, rounded-minute review threshold and day/hour conventions above describe Sibyl’s current implementation.
Sources checked ; this records the reference check, not a new accuracy or scientific-validation review.
This page explains how Sibyl handles birth times and calendar boundaries. It does not calculate Day Master strength, chart structure, Useful God, favourable elements, luck cycles, health, wealth, relationship outcomes, or a universal BaZi school consensus.