The Vimshottari dasha divides a notional lifetime of 120 years among nine planetary rulers, each taking a turn in a fixed order. The calculator above works out which turn you were born into, how much of it was left, and every period after it.
All of that comes from a single measurement: how far the Moon had travelled through its nakshatra at the moment of your birth. Everything else is arithmetic.
How the whole system falls out of one number
The twenty-seven nakshatras carry nine ruling planets in a repeating cycle, and each planet is assigned a period length in years. Ketu 7, Venus 20, Sun 6, Moon 10, Mars 7, Rahu 18, Jupiter 16, Saturn 19, Mercury 17. Add them and you get exactly 120, which is where the name comes from.
Your birth nakshatra names your first ruler. The Moon does not usually enter a nakshatra at the instant you are born, though, so you do not get the whole period. You get whatever was left. If the Moon was 40% of the way through Rohini, which the Moon rules, your first mahadasha is Moon and it runs for 60% of ten years, so six years.
That leftover is called the balance of dasha at birth, and it is the only genuinely computed quantity in the system. After it, the sequence is mechanical: the next planet in the cycle takes its full term, then the next, and so on for the rest of the 120 years.
Sub-periods work by the same proportion applied one level down. Inside any mahadasha, each planet gets a slice sized by its share of the 120-year total. Jupiter's antardasha inside a Saturn mahadasha runs for nineteen years times sixteen over one hundred and twenty, which is a little over two and a half years. Apply the rule once more and you get pratyantardashas, which is why the calculator can nest three levels deep without any new logic.
The year length, which is where the real disagreement is
Here is the thing almost no calculator will tell you, and it matters more than any other choice on the page.
A dasha period is quoted in years. To turn nineteen years of Saturn into a pair of calendar dates, you have to decide how many days a year has. Two answers are in genuine circulation.
365.25 days, the solar year. This is what most modern software uses and what most published astrologers have worked with. It is defended from the astronomical siddhantas, the Surya Siddhanta among them, and from the Bhagavata's treatment of the Sun's motion as the basis of time reckoning.
360 days, the savana year. Defended from the many scriptural periods quoted in round multiples of 360, and from the tidy correspondence with 360 degrees of zodiac.
The two diverge by 5.25 days for every year that passes. After one year the tables differ by five days and nobody notices. After thirty-five years they are roughly six months apart. After seventy years they are a full year apart, and the mahadasha a chart says you are in may not be the one another chart says.
This page lets you switch between them and states which one produced the table you are looking at. Switch it and watch every date move.
There is no way to average the two, and no way to tell from a printed table which was used. If you have ever compared your dasha dates against a relative's chart or an older printed horoscope and found them months apart, this is the first thing to check, before assuming either is wrong.
What your birth time is actually worth here
The Moon crosses one nakshatra in about 24.3 hours. So an hour of uncertainty in your birth time is about four per cent of a nakshatra, which is four per cent of your first ruler's period.
Under Ketu or Mars, seven years, four per cent is about three and a half months. Under Venus, twenty years, it is nearly ten months. And here is the part that gets missed: that shift does not wear off. Every mahadasha boundary for the rest of the 120 years moves by the same amount, because they are all measured forward from the same starting offset. A ten-month error at birth is still a ten-month error when you are eighty.
The calculator asks how sure you are of the time and reports the shift as a number of days rather than as a warning.
There is a second, worse case it also checks. If your uncertainty spans a nakshatra boundary, you do not have a shifted table. You have a different table: a different first ruler, and therefore a different sequence of every period that follows. The page says so plainly and names both candidates, because at that point no amount of interpretation of the printed dates is worth anything.
A worked example, and what it reveals about other calculators
Take a chart where the birth data is public and well attested: Gandhi, born 2 October 1869 at 08:36:19 in Porbandar. AstroSage publishes the chart from Kundli Sangraha and gives the Moon in Cancer, Ashlesha nakshatra, fourth pada, with a balance of dasha of Mercury 2 years 2 months 9 days.
The Moon placement is not in dispute. This page agrees: Cancer, Ashlesha, pada four, Mercury opening the sequence.
The balance is where it gets interesting. Porbandar sits at 69°40′ east. In 1869 there was no Indian Standard Time, which was not introduced until 1906, so the birth time recorded there was local mean time. At that longitude local mean time is UTC+04:39.
Feed this calculator that offset and the balance comes out as Mercury 1 year 6 months 19 days. Feed it the modern +05:30 instead, and it returns 2 years 2 months 15 days, which lands within six days of AstroSage's published figure on a seventeen-year period. That is agreement to a tenth of a per cent, and it is a useful check: it means the two engines compute the dasha itself the same way.
Which tells you exactly where the eight-month discrepancy comes from. It is not the astronomy and it is not the dasha arithmetic. It is a time zone applied thirty-seven years before it existed. Fifty-one minutes of error moves the Moon half a degree, and half a degree of Moon is eight months of Mercury.
This page resolves the offset that was actually in force for your date. Bombay kept UTC+04:51 until 1955 and Calcutta kept UTC+05:53:20 until 1948, neither of which the standard time-zone database can express, so both are handled as explicit exceptions.
How the first mahadasha is cut, and the mistake that hides there
The first mahadasha is short, because only its remainder was lived. Its antardashas have to be shortened too, and there is a right way and a common wrong way to do it.
The wrong way is to scale them: take the shortened mahadasha, apportion it among the nine planets by the usual proportions, and print the result. It looks reasonable and it is wrong, because it silently moves every internal boundary.
The right way is to build the antardashas on the full period as if it had run from the start, then discard the ones that finished before you were born. The one that was running at your birth stays, with its true start date, which is genuinely earlier than your birthday. The calculator marks that row rather than quietly clamping the date, since clamping it would misstate the period's length.
You can see the difference on any chart with a small balance, and the two methods can put an antardasha boundary years apart.
A concrete case makes it obvious. Suppose your balance is one year of Venus out of twenty, meaning nineteen years of that mahadasha had already run before you arrived.
Under the scaling method, your one remaining year gets carved into nine slices in the usual proportions. Venus opens its own sequence, so the first sub-period printed is Venus, lasting one year times twenty over one hundred and twenty: exactly two months.
Under the correct method, the Venus sub-period of a full Venus mahadasha runs three years and four months from the notional start. That finished nearly sixteen years before you were born, so it does not appear in your chart at all. Walking the real sequence forward to the nineteen-year mark lands you in the Ketu sub-period, the last of the nine, which began at 18.83 years and runs to the end of the mahadasha.
So the two methods do not give you slightly different dates for the same sub-period. One says you were born under Venus, the other says Ketu. There is no arithmetic that reconciles them, and a chart cannot tell you which method produced it.
Reading the table without overreading it
The dasha is the most widely used predictive framework in Indian astrology, and it is worth being clear about what this page does and does not offer.
What it offers is arithmetic, done carefully and shown in full. The Moon's sidereal longitude, the fraction of the nakshatra elapsed, the balance to six decimal places, the days-per-year constant used, and the time-zone offset resolved for your date. Every one of those is checkable, and if you disagree with the result you can find out precisely where the disagreement begins.
What it does not offer is any statement about what a Rahu mahadasha will do to your career, or whether a Saturn period is something to brace for. There is no shortage of places online willing to tell you that, frequently attached to a remedy for sale. This page reports periods and dates, and stops.
The one thing worth saying about interpretation is structural. Because the dasha is generated from a single number, the entire predictive apparatus built on it is only as good as your birth time. That is a strong claim to hang on a figure your family may have rounded to the nearest half hour, and it is the reason the uncertainty is reported here in days rather than glossed over.
Why another calculator disagrees with this one
In descending order of how much difference it makes:
The year length. 365.25 versus 360, worth up to a full year over a lifetime. Almost never disclosed. Check this first.
The birth time. Worth months, permanently, and it never washes out.
The time zone applied. For Indian births before 1955, a site assuming +05:30 can be up to fifty-one minutes wrong, which is months of dasha.
The ayanamsa. Lahiri and Raman are about 1.45° apart, and a nakshatra is 13°20′. So roughly one birth in ten sits close enough to a boundary for the two systems to give different first rulers. That is not a shift in the dates. It is a different table entirely, and the selector above lets you test your own case.
The starting rule. This page uses the standard Parashari construction, in which the sequence begins from the Moon's nakshatra. A few traditions start Vimshottari from the lagna or the Sun instead. Those are different systems rather than different implementations, and a site using one will not resemble a site using another.
The nine rulers and their terms
For reference, since the numbers are the system:
Ketu 7 years, Venus 20, Sun 6, Moon 10, Mars 7, Rahu 18, Jupiter 16, Saturn 19, Mercury 17.
The order is fixed and cyclic, so a Ketu mahadasha is always followed by Venus, and a Mercury mahadasha always rolls back round to Ketu. The same nine-step cycle runs through the twenty-seven nakshatras, three times over, which is why Ashwini, Magha and Mula all belong to Ketu.
Two consequences follow from the lengths alone, and neither requires any interpretive claim. The long periods, Venus at twenty and Saturn at nineteen, cover more than a third of the cycle between them, so most people spend a large share of their life under one of two rulers. That has a bearing on how much any single dasha reading can be claiming. If thirty-two of every hundred and twenty years belong to two planets, then a large fraction of any population is under Venus or Saturn at any moment, and a description that fits a Venus period is a description that fits a great many people at once. And whichever ruler you begin under, you will not see the full cycle: 120 years is a notional total, not a life expectancy, so the later mahadashas in any table are there for completeness rather than for reading.