Astrology software
Every calculation this app performs, and what it refuses to do
Ten calculation modes, eighteen divisional charts, 82 event questions — and a published study of whether the timing engine works. Written so a practitioner can check every claim.
In short. astrocult computes a full Parashari chart set, KP as a genuinely separate system, and a Vimshottari event scan across 82 classical questions. What it will not do is attach a date or a probability to a life event — because we tested that and published the result.
The difference from most Jyotish software is not the number of features. It is that every number here is either arithmetic you can verify against an ephemeris, or a classical rule shown with its reasoning attached. Nothing in between is presented as a forecast.
1 · What it computes
Ten independent calculation modes, each with its own reference frame rather than a shared one bent to fit:
- Janma kundli. Lagna, the nine grahas, whole-sign houses, nakshatra and pada, retrogression by instantaneous motion so stations are marked accurately, and combustion.
- Eighteen divisional charts. D-1, D-2, D-3, D-4, D-5, D-7, D-8, D-9, D-10, D-12, D-16, D-20, D-24, D-27, D-30, D-40, D-45 and D-60.
- Vimshottari and Yogini dasha. Vimshottari to three levels — Maha, Antar, Pratyantar — with boundary dates and the corresponding age at each.
- Shadbala and Ashtakavarga. The six-fold strength measures and the bindu tables, computed rather than approximated from sign position.
- KP as its own system. Krishnamurti ayanamsa and Placidus cusps, not the app's Lahiri and whole-sign houses reused under a different label. Star lord, sub lord and cuspal sub lord follow from that.
- Gochar, Sade Sati, Varshphal, Sudarshana Chakra, Ashtakoot milan, Muhurta, BNN, numerology, Panchang, Chogadiya and Hora. Each with its own panel and its own stated basis.
- Birth-time rectification. A minute-by-minute table of lagna and every varga, so you can see exactly where each divisional chart changes sign.
2 · Precision, and how to check it
Positions are computed in-app rather than fetched, so the figures below are the ones that matter. Every one of them is verifiable against any standard ephemeris — that is the point of publishing them.
| Quantity | Method | Agreement |
|---|---|---|
| Sun | Truncated VSOP87 | ~1″ |
| Moon | Meeus lunar theory | ~10″ |
| Time base | ΔT correction applied | — |
| KP ayanamsa | Calibrated to KSK 1900.0 epoch | within ~3′, 1900–2100 |
| Lagna | From city centroid | sub-degree |
| D-60 sensitivity | Shifts with birth time | ~every 2 min |
The KP row deserves the honesty it gets in the app itself: at roughly 3 arc-minutes of drift by 2025, that is about 6% of the narrowest sub-division. Star and sub lords are unaffected — but a cuspal sub lord sitting within 3′ of a boundary may differ from software built directly on Swiss Ephemeris. The app says so rather than hiding it.
3 · The Events section, in full
82 classical questions — marriage, first job, property, higher study, litigation, chronic illness, and so on. Each one is not a keyword but a rule set: a primary house, weighted supporting houses, obstructing houses, natural karakas, and the divisional chart the tradition assigns to it. Marriage, for instance, resolves to the 7th house, the 2nd and 11th as weighted support, the 6th and 12th as obstruction, Venus and Jupiter as karakas, and D-9 for refinement.
What a result actually shows
- Which grahas carry the significations, and by which rule — lordship, occupation, aspect, or natural karaka — rather than an unexplained list of names.
- Named periods, not dates. "Mars–Venus" with the dasha's own boundary dates and your age across it, because those are arithmetic. All three levels are checked before a period is listed.
- Whether Jupiter and Saturn confirm by transit, and in what dignity — exalted, debilitated, retrograde — with the house they transit from.
- A convergence dot, not a score. It counts how many independent classical indicators point at the same period. That measures the rules agreeing with each other. It is not a probability, and the app says so on the legend.
- An explicit "no promise" state. If no planet rules, occupies or aspects the primary house for that question, the app says the classical significators are absent and lists nothing — rather than lowering the bar until something appears.
The part no other software offers
You can enter events that have already happened. The app then shows what these same rules say about them — which period was running, and whether its lords are significators for that event in your chart. It reports the tally plainly: if your real marriage falls in a period the rules never flag, you see that before you weigh anything still ahead.
4 · Where this differs from other Jyotish software
Not on feature count — plenty of packages compute more vargas or more dasha systems. The differences that matter are about what the software is willing to claim.
| Question | Commonly | Here |
|---|---|---|
| Event timing | A date range with a confidence percentage | Named periods with dasha boundary dates; no probability attached |
| Basis for the ranking | An undisclosed score out of 100 | A count of classical indicators that converge, defined on screen |
| Published validation | None | 111 charts, 866 dated events, result published including the null |
| Checking it against your life | Not offered | Backtest panel, plus a multiple-comparisons warning before you look |
| KP | Often Lahiri and whole-sign relabelled | Krishnamurti ayanamsa with Placidus cusps, as KP requires |
| Rectification | Returns one “correct” time | Returns Inconclusive when candidates score alike, and says why |
| When significators are absent | Lists periods anyway | Says the classical significators are not present, and lists nothing |
5 · What it deliberately will not do
These are design decisions, not missing features. Each one was in the product at some point and was removed on evidence.
- No date for a life event. The engine that produced dates was tested against 866 of them and matched about as often as marking off the same amount of calendar at random. Dates were removed rather than kept for the sake of the demo.
- No score out of 100, no confidence bands. A numeric confidence is a claim about likelihood, and there are no grounds for one.
- No tuning the reading to your past. Fitting parameters per subject was tested; it did not improve prediction. Your entered events are used only to show you what the rules say about them.
- No medical, legal or financial advice. Periods flagged for health or litigation mark where classical combinations warrant care. They do not say anything will happen.
- No deciding your birth time for you. Rectification computes positions and scores candidates against events you supply. Choosing among them stays with the practitioner.
6 · Reading it in Hindi or English
Every panel, label, note and caveat exists in both languages — not a machine translation layer over an English product, but written text in each, with Devanagari typography set properly rather than squeezed into a Latin line height.