In India, there is a temple named Konark Sun Temple which is around 750 years old and made up of 100% stones and rocks, and has a chariot which is headed by 7 horses and includes the Hindu god Surya(Sun) in the chariot. This chariot has 24 wheels out of which, only 2 were understood by humans, the rest 22 are still a mystery. Those 2 wheels work like a perfect sundial.

You can refer this video:

Question: How were ancient Indians able to perform such mind-boggling calculations without any kind of technology? Is there any other way to make such a huge sundial without modern technology?

  • 15
    $\begingroup$ Not sure if this might be more suited to History of Science and Mathematics. See what more experienced members say. $\endgroup$ Commented Jan 15, 2023 at 2:10
  • 64
    $\begingroup$ What technology is needed beyond placing a stick in a wall and seeing where the shadow is every hour? Then reproduce that pattern in any design of interest. $\endgroup$
    – JohnHoltz
    Commented Jan 15, 2023 at 4:16
  • 28
    $\begingroup$ Sundials date back to at least 3500 years ago, and some say over 5500 years ago. There are objects from multiple cultures that date back to 1500 BCE that without a doubt were used as sundials. Whether even older obelisks were used to tell time of day is debated; some obelisks go back to 3500 BCE. $\endgroup$ Commented Jan 15, 2023 at 12:13
  • 16
    $\begingroup$ "only 2 were understood by humans" The ancient Indians weren't human? $\endgroup$ Commented Jan 16, 2023 at 4:04
  • 9
    $\begingroup$ You should be aware that random YouTube videos are not a reliable source of factual information, no matter how authoritative they sound. Of course some YouTube producers are reliable and authoritative, but you have to research them and their reputation to figure that out. $\endgroup$ Commented Jan 16, 2023 at 18:44

3 Answers 3


As @JohnHoltz points out in a comment, planting a stick in the ground or in a wall and watching where the shadow falls is something very easy; sundials have been known since prehistoric times.

I’m not sure where you got the idea that this implies “mind-boggling calculations,” because it’s very easy and has been known since the earliest times how to divide numbers. Much more complex calculations already appear in the Rhind Papyrus from ancient Egypt or on cuneiform clay tablets from Mesopotamia.

In India, the work of Aryabhata in astronomy was much more complex than telling the time with the Sun, already in the fifth century CE.

I’m no expert on ancient monuments or civilizations, and even less about the Konark temple, but a quick Google search reveals that “Twelve wheels represent 12 months of the year. According to the Indian calendar, each month has a Shukla paksha and a Krishna paksha, so the other 12 wheels stand for them.” Considering the maximum altitude of the Sun in the sky changes from one month to another, it would be reasonable to assume that the position of each wheel is calculated to correspond to the Sun’s position during the corresponding month. Maybe the hub of each wheel sticks out by a different amount? This is not mentioned in the video.

I wholeheartedly disagree with the video’s conclusion that “If ancient people spent a lot of time creating something, there’s a very good chance that it was done for a valuable, scientific purpose.” Case in point: the pyramids in Egypt certainly took a lot of time to build, yet serve only as tombs for pharaohs and their entourage. Other examples are the numerous temples in any region of the world, or the Coliseum in Rome, which served only for housing games and such.

Just because someone did something a long time ago that we can’t seem to understand now, doesn’t mean that it was done using “advanced” knowledge or techniques. Touristic sites may also like to keep an aura of mystery for the visiting public, so that not everything might be revealed about them.

  • 45
    $\begingroup$ Case in point about the last sentence about touristic sites: There are multiple touristic sites on the equator where tour guides will show that water draining from a smallish bowl rotates in one direction north of the equator and in the other south of the equator due to the Coriolis effect. It's a swindle. It's rather hard to show the Coriolis effect in a very large pan of water 40° north or south of the equator, let along a few steps north or south of the equator. $\endgroup$ Commented Jan 15, 2023 at 12:25
  • 3
    $\begingroup$ "The pyramids in Egypt [...] serve only as tombs for pharaohs and their entourage." Pharaohs were not buried in the pyramids. That's a common misconception. $\endgroup$ Commented Jan 17, 2023 at 18:14
  • 3
    $\begingroup$ @Carl-FredrikNybergBrodda: So, then, why does the first paragraph about Egyptian pyramids on Wikipedia end with “Of those located in modern Egypt, most were built as tombs for the country's pharaohs and their consorts during the Old and Middle Kingdom periods.”? en.wikipedia.org/wiki/Egyptian_pyramids (with three references) $\endgroup$ Commented Jan 18, 2023 at 1:44

Those 2 wheels work like a perfect sundial.

No they don't. They're flat. The timespan from one hour to the next on a flat sundial varies from season to season. At the same time that that temple was being constructed, people elsewhere had developed curved sundials that overcame that key problem with flat sundials.

How were ancient Indians able to perform such mind boggling calculations whithout any kind of technology, is there any other way to make such huge sundial without modern technology?

What mind boggling calculations?

If one defines time as that quantity which is measured by those sundials, then yes, they accurately measure time to the minute. But that's just a tautological definition. That isn't how we define time anymore. There's a point in the video where someone points to his wristwatch, implying accuracy to the minute. This is a blatant falsehood. There are two key issues with this, the equation of time and timezones.

The equation of time represents the difference between apparent solar time (time measured by a perfect sundial; keep in mind that a flat sundial cannot be a perfect sundial) and mean solar time. Mechanical, electronic, and atomic clocks are designed to measure mean solar time. Apparent solar time can differ from mean solar time by up to 16 minutes. The equation of time was known to ancient Greeks, and possibly ancient Babylonians before them.

For a long time, local apparent solar time was deemed to be the true measure of local time, while time on a mechanical clock was deemed to deviate from this. This point of view switched with the development of increasingly more accurate mechanical timekeeping devices, coupled with the inability to tell time by a sundial at night or when the sky was overcast. Now we tell time by our mechanical or electronic devices, preferably synchronized with atomic clocks.

The other problem with that part of the video is timezones. India has one timezone. If someone in the far eastern part of India called someone in the far western part of India and compared the time on their perfect sundials, they would agree to disagree by ninety minutes. They would however agree on the time displayed on their cellphones, and that time would disagree with the time displayed on their perfect sundials.

  • 1
    $\begingroup$ "Mechanical, electronic, and atomic clocks are designed to measure mean solar time." While the second was once defined by mean solar time, it is no longer. Precise timekeeping has become disconnected from the Sun. Thus, we have the troublesome leap seconds. $\endgroup$
    – John Doty
    Commented Jan 16, 2023 at 15:09
  • 1
    $\begingroup$ @JohnDoty I intentionally avoided the issue of leap seconds in my answer. Leap seconds are designed to keep civilian time (UTC) in sync with mean solar time (UT1), to within 0.8 seconds. International Atomic Time (TAI; while the acronym is based on French, the acronym is TAI regardless of language) does not have leap seconds. $\endgroup$ Commented Jan 16, 2023 at 15:32
  • 1
    $\begingroup$ Timezones were, of course, not a problem 800 years ago when this was built. Not that the apparent time wouldn't vary by longitude, of course, but no person or information could travel fast enough for that difference to be all that significant to them. $\endgroup$ Commented Jan 17, 2023 at 16:36
  • 1
    $\begingroup$ @DarrelHoffman The problem with timezones is the people in the video who show their watches, which presumably are set to the Indian time zone, supposedly matching the time read from the sundial. $\endgroup$ Commented Jan 17, 2023 at 16:53

They were possible millenia ago as all that was needed to set them out was observation and marking of hour angles.

Equatorial sundials, i.e. a sundial whose plane is oriented parallel to the equator's plane and whose gnomon (pointer) is set at 90 degrees to the dial at its center, are easy to construct as the hour angles are equal.

And because hour angles are equal we can make an equation of time correction by rotating the dial by the correct amount for the day in question relative to the meridian at the equinox.

There need not be elaborate calculations for the equation of time because it can be determined empirically through the year and recorded. Its variation with each passing year due to precession, nutation, slowing of the earth, variation of planetary separations, etc are small relative to the life of the dial. In any case one can remeasure the equation of time some years after construction and re-record the daily hour angle offsets.

I looked at the Konark dial and it seems vertical even though for an equatorial dial with equal hour angles it should really be inclined ~ 20 degrees to the vertical for a city on latitude 19.89 degrees N.

And neither is it accurate to just place a finger roughly square to the dial as the pointer on an equatorial dial at Konark should be inclined ~ 20 degrees below the horizontal.

The basic dial is therefore simple.

Dials which are not equatorial and/or are not south facing may have their hour angles calculated from geometry that has been around for at least 1500 years.

And don't forget that guys like Al-Battani were no mean mathematicians.

  • 1
    $\begingroup$ Indeed! You can project an equatorial hour angle dial to any plane using trigonometry, although projection to horizontal or south-facing (in the northern hemisphere) planes is easiest. Indian mathematicians 750 years ago had considerable knowledge of trigonometry, and it took Western mathematicians a couple of centuries to catch up to them. en.wikipedia.org/wiki/… $\endgroup$
    – PM 2Ring
    Commented Jan 15, 2023 at 20:47
  • $\begingroup$ @PM 2Ring But be careful that the projection is from the new plane hour-angle back to the equatorial plane hour-angle and not vice-versa ! But the equations in the Sundial wiki page are precise - I derived them myself via 3D Euclidean geometry and checked them against rotation matrix methods used by others - same result. And a 24 hour dial is possible using a glass plate. $\endgroup$
    – Trunk
    Commented Jan 15, 2023 at 21:52
  • 1
    $\begingroup$ Here's a sundial I designed a couple of decades ago. And here's a more complete version. $\endgroup$
    – PM 2Ring
    Commented Jan 16, 2023 at 4:59
  • $\begingroup$ @ PM 2Ring Nice idea. Note that the gnomons on the east and west-facing sides must be parallel to that on the principal north/south facing sides. This is usually done by raising the gnomon (by about quarter of the diameter of the dial plate) above the plate plane. qph.cf2.quoracdn.net/… It seems your craftsman just inserted a gnomon square to the east and west facing plates. $\endgroup$
    – Trunk
    Commented Jan 16, 2023 at 12:45
  • $\begingroup$ Using a pin or rod for the gnomon on the east & west dials has the advantage that it can show the declination. OTOH, such a long protruding gnomon can be hazardous (and a target for vandals) unless the dial is very large. I don't know what you mean by "your craftsman". As I said, I created that design, and "built" and rendered it using POV-Ray. The dial and its shadows are mathematically accurate. $\endgroup$
    – PM 2Ring
    Commented Jan 16, 2023 at 13:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .