Choose the confidence that matches the evidence
Choose Exact only when you have a recorded time you are prepared to treat as one clock minute, after checking whether the record uses a 12-hour or 24-hour format. Choose Approximate for a rounded record, a family recollection, an estimate such as “around 7:30,” or any time whose likely error can be expressed as a plus-or-minus window. Choose Unknown when there is no defensible clock value.
The labels describe the source quality, not the importance of the person or chart. A time printed on a document can still have been rounded or transcribed, so the interface cannot certify that the record is historically exact; it can only preserve the confidence the user selects.
What each placement can support
Rising is the most sensitive supported outcome because it is tied to the local eastern horizon. It requires a time. The Moon can cross a sign during one local date, so an unknown-time Moon must be checked across the date. The Sun and planets usually move more slowly, but an ingress can still fall inside an approximate window or local date; “usually stable” is not permission to skip the boundary test.
| Placement | Exact time | Approximate time | Unknown time |
|---|---|---|---|
| Rising | One sign and degree | All signs and degree ranges in the window | No result; never guessed |
| Moon | One sign and degree | Stable range or all possible signs | Full local-date result: stable or possible signs |
| Sun and planets | One sign and degree | Stable range or all possible signs | Full local-date result: stable or possible signs |
How approximate and unknown ranges are evaluated
Approximate time creates a real interval around the entered clock label. The calculator resolves every valid civil-time sample in that interval, including both occurrences of a repeated daylight-saving label. It does not calculate only the centre and add a generic warning. Unknown time expands the interval to every valid instant belonging to the local calendar date.
The output is confidence-aware. One sign is reported only when all evaluated positions agree. If the interval crosses an ingress, all possible signs appear. A degree range describes only what the engine observed across the evaluated interval; it is not a probability distribution and does not identify the most likely birth minute.
Finding a better record without pressure
Possible sources include a long-form birth certificate, hospital record, civil registry extract, baby book or contemporaneous family note. Availability, wording and precision vary by country and institution. You do not need to obtain a document to use the calculators, and the project does not encourage sharing a record with the site or with strangers.
If sources disagree, keep the uncertainty visible. A remembered “about 8” should not become 8:00 exact. You can test an approximate window and compare whether Moon, Rising or a planet changes inside it. This is a calculation of possible placements, not birth-time rectification and not evidence that an event or personality trait proves a particular time.
Check date, place and timezone too
A precise clock value attached to the wrong place or timezone is not an exact astronomical instant. Confirm the country and region for same-name places, inspect the displayed coordinates and IANA zone, and resolve any repeated or nonexistent daylight-saving time shown by the audit. Historical offsets can include half-hour or other non-hour values.
If another calculator differs, compare its zodiac system, UTC conversion and missing-time rule before changing your input. The result-comparison guide provides a fixed order for that check.