In 1900 a sponge diver called Elias Stadiatos discovered the wreck of an ancient merchant ship off the tiny island of Antikythera near Crete. The corbita,
dating from the first century B.C., was heavily laden with treasure of
all kinds, original bronze life-size statues, marble reproductions of
older works, jewelry, wine, fine furniture and one immensely
complicated scientific instrument. 
The Antikythera mechanism was originally housed in a wooden box about the size of a shoebox with dials on the outside and a complex clockwork assembly of gears inscribed and configured to produce solar and lunar positions in synchronization with the calendar year. By rotating a handle on its side, its owner could read on its front and back dials the progressions of the lunar and synodic months over four-year cycles. The device has been estimated to be accurate to 1 part in 40,000.
The bronze gearing, remarkable enough on its own right, also contains a further innovation that would not be reinvented until the 19th century, the differential gear. The differential was used to calculate the phases of the moon by subtracting the moon's motion from that of the sun's. This level of sophistication allows us to say without fear of exaggeration that the Antikythera mechanism was an early kind of analog computer.
The device is also thought by some to have been able to model the motion of the five planets using the epicyclical model of planetary movement around a fixed earth devised by Apollonius of Perga and Hipparchus of Rhodes (later superceded by the heliocentric model of Copernicus).
It's been said that the Antikythera mechanism actually dropped and sank twice. The second submersion came after a comprehensive analysis of Antikythera mechanism was done by Derek de Solla Price (see Scientific American June 1959 and Gears from the Greeks: the Antikythera Mechanism: a Calendar Computer from ca. 80 B.C. 1975). Since then surprisingly little scholarly attention has been paid to what is surely the most exciting relic of advanced ancient technology that we have in our possession. After one hundred years, our estimation of the scientific and technology of the ancient Greeks needs to be be seriously revised.
"Suppose a traveller carried into Scythia or Britain the orrery recently constructed by our friend Poseidonius, which at each revolution reproduces the same motions of the sun, the moon, and the five planets that take place in the heavens every day and night, would any single native doubt that this orrery was the work of a rational being?"
? Cicero