if a computer is in a frame of reference rotating at a high rate of speed, do relativistic issues affect keeping the cpu clock in sync etc?

(just curious. don't have enough math/physics background to fully appreciate an answer but would find it pleasing to know someone who knew it better could tell what would happen :D)

· · Web · 4 · 2 · 4

(my gut check is that material stress from the acceleration would destroy the chip before reaching a rotational velocity where the difference in acceleration/velocity in different parts of the chip matters though :D)

@brion Wasn't this part of the issues that had to be calculated when setting up the GPS system initially?

@brion i suspect that if your computer is big enough that it doesn't get torn apart, then you would have to… but also that signal propagation would take so long that you have multiple problems to deal with.


Calculation: 1/2

r_comp = 0.1 # computer radius in m
dtheta_dt = 1e9 # radians/s versus inertial reference frame
v = dtheta_dt*r_comp # lin. speed on boundary
beta = v/299792458 # v in natural units
phi = atanh(beta) # rapidity on boundary
L_m = 2*pi*r_comp # circumf. in m
L_s = L/299792458 # same in s
gamma = cosh(phi)
tprime_offset = beta*phi*L_s


Description: 2/2

A 10cm radius circular computer has a circumference of L_m = 63 cm in its own reference frame, which is 2.1 ns writing in units of time. Let's spin the computer at 1 Gigaradian/s and assume that it doesn't break apart.

Then the constant t' line (in the computer's frame (x', t')) around the computer's circular boundary will mismatch itself by about beta gamma L_s = 0.24 ns. So a synchronisation signal sent one loop around the circumference will mismatch by 0.24 ns.

Sign in to participate in the conversation
Mastodon for Tech Folks

This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either!