Transparency
Our methodology
What Nakshara computes every chart with, and how we check it: Swiss Ephemeris, the Lahiri ayanamsa, and a plain account of the reference-chart cross-checks, including what they do not prove.
Most kundli apps never say what engine their calculations run on, or what that engine was checked against. This page does: what software Nakshara uses to compute planetary positions, which ayanamsa it applies, and how that computation was cross-checked against published references.
We don't manufacture an accuracy claim, and we don't cite a credential we don't have. Every number below is pulled straight from our test code, and if that code changes, this page changes with it.
The calculation engine: Swiss Ephemeris
Nakshara computes planetary positions with Swiss Ephemeris, the library most widely used in professional astronomical-astrology software, developed by Astrodienst. Specifically, we use its Moshier analytical ephemeris rather than the JPL-derived mode that depends on external data files.
That choice is deliberate: the Moshier ephemeris is built directly into the library, so it needs no external data files, and the result is identical across every server and every environment it runs on. Its accuracy for the Moon is well under one arcsecond, several orders of magnitude finer than the 3°20' pada boundaries our nakshatra calculations resolve.
The ayanamsa: Lahiri, and only Lahiri
Western astrology uses the tropical zodiac. Vedic astrology uses the sidereal zodiac, anchored to the fixed stars. The gap between the two is the ayanamsa, and it grows by roughly 50 arcseconds (0.014 degrees) every year.
Several ayanamsas exist, including Raman, Krishnamurti (KP), and Fagan-Bradley. Nakshara always uses the Lahiri ayanamsa, the official standard adopted by India's Calendar Reform Committee in 1955, used by most panchang publications, kundli software, and published reference charts. We don't offer a choice of ayanamsa, because sticking to one standard is what makes our numbers comparable to another tool's in the first place.
How we checked it
Before building anything on top of the calculation engine, we wrote automated tests that check it against published references. These tests run against four distinct kinds of ground truth, every time the code changes:
1. Six public figures’ birth charts
The Moon rashi and nakshatra for Mahatma Gandhi, Jawaharlal Nehru, Albert Einstein, Sachin Tendulkar, Narendra Modi and Indira Gandhi are checked against the values published on celebrity.astrosage.com (Indira Gandhi’s birth data comes from astro.com’s Rodden-rated databank, which grades the reliability of the birth-time source). This check runs independently in two separate test files.
2. Two published ayanamsa values
The Lahiri ayanamsa at J2000.0 against the value published in the Indian Astronomical Ephemeris (23°51′11″), within a 0.0045° tolerance. And against N.C. Lahiri's own Chitrapaksha ayanamsa definition epoch (21 March 1956) at 23°15′, within a 0.006° tolerance.
3. Seven astronomical events
Seven moments drawn from NASA’s eclipse catalogue and the public astronomical record: two total solar eclipses, a total lunar eclipse, the 2020 Jupiter-Saturn great conjunction, a Mercury transit across the Sun, a Venus inferior conjunction, and a Mars opposition. For each, our engine computes the separation between the two bodies involved and checks it against the known astronomical value at that instant.
4. Hand-worked Ashtakoota and dasha arithmetic
Expected Ashtakoota results (5 pairs, 4 Manglik cases) and Vimshottari dasha results (3 cases) were worked out on paper from the classical rules before the engine was ever run, so the engine can’t grade its own homework.
Reference charts: engine vs. published
These six charts live in our test suite and run every time the code changes:
| Name | Engine result | Published value | Source |
|---|---|---|---|
| Mahatma Gandhi | Karka / Ashlesha ✓ | Karka / Ashlesha | celebrity.astrosage.com |
| Jawaharlal Nehru | Karka / Ashlesha ✓ | Karka / Ashlesha | celebrity.astrosage.com |
| Albert Einstein | Vrishchika / Jyeshtha ✓ | Vrishchika / Jyeshtha | celebrity.astrosage.com |
| Sachin Tendulkar | Dhanu / Purva Ashadha ✓ | Dhanu / Purva Ashadha | celebrity.astrosage.com |
| Narendra Modi | Vrishchika / Anuradha ✓ | Vrishchika / Anuradha | celebrity.astrosage.com |
| Indira Gandhi | Makara / Uttara Ashadha ✓ | Makara / Uttara Ashadha | astro.com (Rodden-rated) |
All six charts match the published rashi and nakshatra exactly. This is proof of the placement calculation, not a claim about the accuracy of any predictive reading built on top of it.
What this doesn't prove
Honesty is the whole point of this page, so here are the limits. The reference charts use publicly published birth times, not birth-certificate-verified ones, though rashi and nakshatra both stay stable across a wide enough window (roughly 13 degrees and 53 minutes respectively) that ordinary time uncertainty doesn’t change the result.
The Ashtakoota rules are deliberately calibrated to AstroSage’s documented conventions in a dedicated ruleset (`guna_tables.py`), including the Dhanu/Makara half-sign split in Vashya and Bhakoot dosha’s cancellation conditions. That’s a design decision, not a claim that it’s the one correct reading among classical traditions. The mid-range Yoni scores (the friend/neutral split) follow the classical Yoni table rather than AstroSage, because AstroSage’s live matcher blocks automated checking.
None of this has been audited by an outside body or a professional astrologer. It’s our own test-driven process, and it’s visible in the code, not a credential we’re asking you to take on faith.
FAQ
Questions about our methodology
Has Nakshara been checked by a professional astrologer? +
No. Our checks are automated tests run against published astronomical and astrological references, not an outside audit. We don't claim a professional astrologer has certified this.
If my birth time is uncertain, will my result change? +
Rashi and nakshatra can absorb several minutes of uncertainty, since each rashi lasts about 2 hours and each nakshatra about 53 minutes. The ascendant, though, changes roughly every 4 minutes, so an accurate birth time matters a lot more for ascendant-based calculations.
Why might Nakshara give a different result than another kundli app? +
The most common reason is a different ayanamsa: an app using Raman or KP ayanamsa places rashi/nakshatra boundaries roughly 0.4 to 0.6 degrees from Lahiri, which can flip the result for a birth time near a boundary. Rounded birth times or an incorrect historical time zone are the other usual culprits.
Will this methodology ever change? +
We have no plan to change the engine or the ayanamsa; they're the foundation every calculation sits on. If that ever changes, this page changes with it.
See your chart for yourself
The same engine, the same ayanamsa, behind our free birth chart and kundli matching.
Accurate astronomical calculation · no spam · Privacy