Methodology
How we compute your chart
Most astrology sites will not tell you which ephemeris they use, which ayanamsa they apply, or which house system they draw. That makes their accuracy claims impossible to check, which is convenient for them. This page states the method exactly, gives reference values you can compare against a published almanac, and is honest about where our figures are approximate.
The calculation, step by step
1. Your birth time becomes a moment in UTC
A date, a clock time and a place are not yet a moment in time — they need a time zone. We resolve the zone from the coordinates of your birth place and apply the historic rules from the IANA time zone database, so wartime daylight saving, half-hour offsets and the many local rule changes of the twentieth century are handled rather than assumed away. The offset is resolved in two passes, which matters for births that fall inside a daylight-saving transition.
2. Planetary positions come from a professional ephemeris
We use astronomy-engine, an open-source implementation of the VSOP87 planetary theory and the ELP2000 lunar theory. It agrees with NASA JPL reference ephemerides to well under an arcminute across our supported range of 1900 to 2100. Rahu and Ketu are computed as the mean lunar nodes, which is the usual Vedic convention; some software uses the true node, which oscillates around the mean by up to about a degree and a half.
3. Tropical positions are converted to the sidereal zodiac
The ephemeris gives tropical longitudes, measured from the March equinox. Vedic astrology uses the sidereal zodiac, measured against the fixed stars. The difference between them is the ayanamsa, and it is a convention rather than a measurement — different schools fix it to different reference points and disagree by up to about a degree and a half. We default to Lahiri and let you switch, because the honest thing is to show that the answer depends on a stated choice.
4. Houses are whole-sign
The entire sign containing your ascendant is the 1st house, the next sign is the 2nd, and so on. This is the classical Vedic default and the reason every house in our charts spans exactly thirty degrees. Software offering Placidus, Koch or Equal houses is applying a Western system, and its house placements will disagree with ours and with most Indian astrologers.
5. Dashas come from the Moon’s exact position
The vimshottari cycle is 120 years divided among nine lords. Which one you were born into, and how much of it had already elapsed, is set by exactly where the Moon sat within its nakshatra at your birth. Sub-periods divide each mahādaśā in proportion to the sub-lord’s own allotted years. For the period already running when you were born, we reconstruct the true pre-birth start before dividing — otherwise every sub-period in your first mahādaśā would be wrong.
Reference values you can check
These are computed at page-build time by the same functions that generate your chart, so they cannot drift away from what the calculator actually does. Compare them against any published table.
| Moment | Lahiri | Krishnamurti | Raman |
|---|---|---|---|
| 1 Jan 1950, 00:00 UT | 23°09′16″ | 23°09′29″ | 21°43′16″ |
| 1 Jan 2000, 00:00 UT (J2000) | 23°51′08″ | 23°51′21″ | 22°25′08″ |
| 1 Jan 2026, 00:00 UT | 24°12′55″ | 24°13′07″ | 22°46′55″ |
The standard published Lahiri value at J2000 is 23°51′12″. Ours differs from it by about four arcseconds — roughly a thousandth of a degree.
The three ayanamsas, and how honest each of ours is
We implement three. Two are computed from their own published definitions. One is an approximation, and we would rather say so than let you assume otherwise.
Lahiri — Chitrapaksha — the Indian government standard
Fixed so that the star Spica (Chitra) sits at exactly 180° of sidereal longitude. Adopted by the Indian Calendar Reform Committee in 1955 and used by the great majority of Indian astrologers.
Our implementation: Verified to 3.6 arcseconds at J2000 and 18 arcseconds at 2026 against published tables.
Krishnamurti — KP New — for Krishnamurti Paddhati
Defined from a 1900 epoch value of 22°27′37″ advancing at 50.2388475 arcseconds per year. It sits within a few arcseconds of Lahiri, so charts rarely differ.
Our implementation: Computed directly from its published definition, so exact to it. Runs 12 arcseconds above Lahiri at J2000.
Raman — B. V. Raman's system
Uses an earlier reference epoch than Lahiri and so runs about a degree and a half smaller. A planet in the first degree and a half of a sign under Lahiri will usually fall in the previous sign under Raman.
Our implementation: Approximate — a constant 1°26′ offset from Lahiri rather than an independent derivation. Matches published values to 7 arcseconds at 2026, but treat it as good to roughly ±3 arcminutes generally.
The Raman figure deserves the plainest statement. It is derived as a fixed 1°26′ offset from Lahiri rather than from Raman’s own epoch, because published sources put the gap anywhere between 1°25′ and 1°30′ and we could not verify a single authoritative constant. Treat it as good to roughly ±3 arcminutes. That is fine for deciding which sign a planet occupies unless it sits within a few arcminutes of a boundary — and in that case the chart will tell you the degree, so you can judge for yourself.
What limits accuracy is your birth time, not our arithmetic
It is worth being blunt about the scale of things. Our computational error is measured in arcseconds. The uncertainty introduced by a birth time recorded to the nearest fifteen minutes is about four degrees of ascendant — some fourteen thousand arcseconds, or roughly three thousand times larger.
- The ascendant moves about 1° every 4 minutes, so it is by far the most time-sensitive part of a chart.
- The Moon moves about 13° per day, so its sign is usually safe even with no birth time at all.
- Every other planet takes weeks or years to change sign, so those placements are essentially immune to a wrong birth time.
This is why our tools let you proceed without a birth time and then tell you which parts of the result to trust. Any site that presents a lagna to the arcsecond from a birth time you guessed is selling precision it does not have.
What we do not compute
As much a part of an honest methodology as what we do. We currently do not offer ashtakavarga, shadbala, yoga detection, Jaimini or KP sub-lord analysis. Some of these are on the way; none of them are hidden behind a payment. We also do not compute anything whose output would be a prediction of a specific event, because that is not a claim the method supports.
We do not sell remedies, gemstones or poojas, and we never will. That is not only an ethical position — it removes the commercial incentive to overstate what a chart says.
Common questions
Why does this page exist when no other astrology site has one?
Because the claim that a site is accurate is worthless unless it can be checked. Every number on this page can be verified against a published almanac or against other astrology software. If our figures were wrong, this page is how you would catch us — which is the point of publishing it.
Which ephemeris do you use, and why not Swiss Ephemeris?
We use astronomy-engine, an open-source implementation of the VSOP87 planetary theory and the ELP2000 lunar theory, which agrees with NASA JPL reference ephemerides to well under an arcminute for every body we compute. Swiss Ephemeris is excellent and slightly more precise, but it is licensed under the AGPL, which would require us to impose licensing conditions we do not want on a free site. The difference between the two is far smaller than the error in any recorded birth time.
How accurate is accurate, in practice?
Our planetary positions are good to better than an arcminute, and our Lahiri ayanamsa matches published tables to about four arcseconds. But the limiting factor is never the astronomy — it is your birth time. The ascendant moves roughly one degree every four minutes, so a birth time recorded to the nearest fifteen minutes carries about four degrees of uncertainty in the lagna. That is thousands of times larger than our computational error.
Why do I get a different chart on another website?
Almost always the ayanamsa, the house system, or the time zone. If the difference is around 24 degrees, the other site is using the tropical zodiac rather than the sidereal. If it is about a degree and a half, one of you is using Raman. If the planets agree but the houses differ, the other site is using Placidus or another Western house system rather than whole-sign. If everything is off by an hour, it is a daylight-saving disagreement about your birth date.
Do you handle historic daylight saving time?
Yes. Time zone offsets are resolved through the IANA time zone database via the browser's own Intl implementation, which means historic rules are applied correctly — including India's brief wartime DST in the 1940s and the many changes in European and American rules. We resolve the offset in two passes so that times falling inside a DST transition are handled rather than silently shifted.
Can I verify this myself?
Yes, and we would like you to. Compare our ayanamsa figure — shown on every chart you generate — against any published Lahiri table. Compare planetary positions against Jagannatha Hora, Swiss Ephemeris, or a printed panchang. If you find a discrepancy larger than a few arcseconds that is not explained by a different ayanamsa or house system, please write to us.
Found an error? Tell us — corrections to this page are the most useful mail we receive. Or go and generate a chart and check the ayanamsa figure yourself.