Researchers home in on possible "day zero" for Antikythera mechanism:
The mysterious Antikythera mechanism—an ancient device believed to have been used for tracking the heavens—has fascinated scientists and the public alike since it was first recovered from a shipwreck over a century ago. Much progress has been made in recent years to reconstruct the surviving fragments and learn more about how the mechanism might have been used. And now, members of a team of Greek researchers believe they have pinpointed the start date for the Antikythera mechanism, according to a preprint posted to the physics arXiv repository. Knowing that "day zero" is critical to ensuring the accuracy of the device.
"Any measuring system, from a thermometer to the Antikythera mechanism, needs a calibration in order to [perform] its calculations correctly," co-author Aristeidis Voulgaris of the Thessaloniki Directorate of Culture and Tourism in Greece told New Scientist. "Of course it wouldn't have been perfect—it's not a digital computer, it's gears—but it would have been very good at predicting solar and lunar eclipses."
[...] Voulgaris and his co-authors based their new analysis on a 223-month cycle called a Saros, represented by a spiral inset on the back of the device. The cycle covers the time it takes for the Sun, Moon, and Earth to return to their same positions and includes associated solar and lunar eclipses. Given our current knowledge about how the device likely functioned, as well as the inscriptions, the team believed the start date would coincide with an annular solar eclipse.
In such an event, the Sun and Moon are precisely aligned with Earth, such that the Moon appears smaller and only covers the Sun's center, leaving the Sun's visible outer edges to form a "ring of fire." An annular eclipse in which the Moon was at the furthest point from Earth in its orbit (the apogee) would have been of particularly long duration. So Voulgaris and his cohorts searched NASA's database to find all the examples of such events falling within the time period the Antikythera mechanism was likely built.
Only the Saros series 58 included long annular eclipses. The longest occurred on December 23, 178 BCE. "Usually, in order to perform time calculations, it is more common to select a date from the recent past than one in the future, especially during ancient times, when time calculations and predictions for a large time span were more uncertain and doubtful than today," the authors wrote. "This fact could also be the most probable reason for the construction of the Antikythera mechanism in that era."
As further evidence, Voulgaris et al. cite several other culturally significant astronomical events that would have occurred around the same time. One is the annual winter solstice, helpfully engraved on the front top left of the mechanism. Voulgaris et al. believe that's a strong indication that the solstice was involved in the calibration. Another is the religious festival Isia, marking the assassination of Osiris and tied to lunar and solar eclipses. There would have been a visible solar eclipse at sunrise on December 22, 178 BCE, per the authors, a rare occurrence and hence likely to hold significance for priests of that period.
"This is a very specific and unique date," Voulgaris said. "In one day, there occurred too many astronomical events for it to be coincidence. This date was a new moon, the new moon was at apogee, there was a solar eclipse, the Sun entered into the constellation Capricorn, it was the winter solstice."
(Score: 5, Interesting) by bzipitidoo on Wednesday April 13 2022, @02:59AM (2 children)
Seems a bit awkwardly put, their claim that calibration is necessary. Yes, but the way they said it, makes it sound like the calibration point was designed into it, before construction, and I'm not at all sure that would have been so. Taking their example of a thermometer, first you build it, then you figure out where the markings go simply by looking at the level with another, known measurement device handy for comparison. You don't put markings on it in the middle or at the start of construction, you do that at the end.
I presume that it being cyclic, the Antikythera doesn't have a start or end point, though a spiral grove, yeah, that wouldn't be something easily made repeatable. But I am thinking it could be like an odometer that just rolls over back to 0 after passing 999,999 (or however many miles/km they design into odometers today). It could have simply been cranked to whichever of the 223 months it had been since the last astronomical combo of eclipses and solstices or whatever-- which could have been Dec 22, 178 BCE, but need not have been.
The ancients were real big on eternal cycles, and we, today, are still laboring under that assumption. We now know that the Earth's day is very slowly lengthening, thanks to tides transferring Earth's rotational momentum to the Moon, boosting it into a higher orbit. Yet our time keeping systems still implicitly assume that a day is exactly 86,400 seconds, always was, and always will be. It's built into our computers. Yeah, we dealt with the Y2K prob, but our current method for dealing with the lengthening day is to either ignore it, or add "leap seconds" as a kind of special, one off adjustment, as if it need not ever be done again, until it has to be done again.
(Score: 1, Interesting) by Anonymous Coward on Wednesday April 13 2022, @03:52AM
The problem is that the 223 month cycle is the common multiple for the Sun, Moon, and Earth. The mechanism predicts a number of other events that are not synced with the 223 month cycle. To get a common multiple for all the different events would be a *much* longer cycle. Thus the spiral triggers other events, and there is a "starting day" to line everything up. There are also elliptical approximations in the mechanism (speed up and slow down within one cycle), one gear drives another on an offset axis, through a pin-slot arrangement.
If you are interested, look for the videos by "clickspring" who builds a copy based on the x-ray tomography.
(Score: 2, Interesting) by khallow on Wednesday April 13 2022, @04:42AM
"But". Our current method isn't going to require a lot of such adjustments for thousands of years. Plenty of time to come up with something better. My bet is that eventually we'll settle on the second as the unit of measure for time, both because it's already used universally in scientific circles - far more widespread than any language. And second, the second isn't Earth-centric. That is, you don't need to know an Earth day to this level of precision, if you're somewhere else in the Solar System.
I think a better measure of the true frivolousness of measuring the accuracy of the Earth day is Daylight Savings Time, with the standard moving clocks forwards and backwards with such poor justification or little impact on civilization.
(Score: 2, Offtopic) by kazzie on Wednesday April 13 2022, @03:47AM
There's far more significance in that date than the Unix Epoch.
(Score: 5, Funny) by WeekendMonkey on Wednesday April 13 2022, @08:58AM (1 child)
I misread the title and thought it was a zero day exploit announcement. Admittedly the number of affected machines is very small in this case.
(Score: 2) by Freeman on Wednesday April 13 2022, @01:55PM
Nah, more like a Y2K bug, but way earlier.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"