Highlighted pillars are the ones that moved. Neither column is a trick — they are the same engine, run once without the correction and once with it.
This is the big one. A whole country keeps a single standard meridian, so if you were born east or west of it your local noon is not twelve o’clock. The arithmetic is plain — four minutes for every degree of longitude away from that meridian. Spain sits west of Greenwich but runs on Central European Time, so Madrid is more than an hour out. Xinjiang keeps Beijing time from 2,000 km away.
Mid-January, so no summer time. In summer a country on daylight saving shifts a further hour.
An hour of error moves the hour pillar by a whole branch, because each branch covers two hours. Half an hour moves it whenever the birth sits near the boundary — which is a quarter of all births.
Korea ran summer time in 1948–1960 and again in 1987–1988. Anyone born in those windows has a clock reading an hour ahead of standard time, and a calculator that does not know the history will place their hour pillar one branch late. The same trap exists in most countries; the rules changed often and were rarely tidy. This is a lookup problem, not a doctrinal one — a tool either carries the timezone history or it does not.
The month pillar turns when the sun reaches a particular apparent longitude. That moment drifts by hours from year to year, so it does not land on a fixed calendar date. A tool that reads the month off a table will put anyone born within a day of the boundary into the wrong month — and the month pillar is the one most readings lean on hardest.
The branch that begins at 11pm belongs to the next day’s cycle in one reading, and to the current day in another. Schools genuinely disagree here, and both positions have a long pedigree. If two charts differ only in the day pillar and the birth was late at night, this is usually why — and neither tool is broken. It is worth knowing which convention a reading assumed.
Older family records often carry the lunar date. Typed into a field expecting a Gregorian date, it produces a chart that is close enough to look plausible and wrong by weeks. If a chart feels off by about a month, this is the first thing to check.
For the first three, there is a correct answer and it is arithmetic rather than opinion. For the fourth, there is a real difference of schools. For the fifth, ask which calendar the date came from.
If you want to see your own chart with the birthplace taken into account, the calculator is here. It is free, it asks where you were born, and it will tell you how many minutes the correction came to.