Every GPS-enabled phone is running a live relativity experiment. The satellites broadcasting your location are moving fast enough, and sitting far enough from Earth’s gravity, that their onboard clocks genuinely tick at a different rate than a clock on the ground. Ignore that difference and your position on the map would drift by roughly 10 kilometers a day. Engineers correct for it by literally detuning the satellite clocks before launch. Understanding why reveals something most explanations of relativity skip over: there isn’t one effect at work here, there are two, and they pull in opposite directions.
Time Isn’t One Speed for Everyone
Einstein’s relativity describes two separate reasons a clock can run at a different rate than another clock: how fast it’s moving, and how deep it sits in a gravitational field. Both apply to GPS satellites simultaneously, which is what makes the GPS case such a clean real-world demonstration: you get to see both effects, measure them separately, and watch them partially cancel out.
Special relativity: motion slows clocks down
Special relativity says that a moving clock runs slower relative to a stationary observer. GPS satellites orbit at about 14,000 km/h, roughly 3.87 kilometers per second. That’s fast enough for the effect to be measurable, even though it’s nowhere near the speed of light. At that velocity, satellite clocks lose about 7 microseconds per day compared to a clock sitting still on Earth’s surface.
General relativity: weaker gravity speeds clocks up
General relativity adds a second, larger effect. Gravity itself slows down time: a clock closer to a massive object ticks slower than one farther away, because gravity curves spacetime itself. GPS satellites orbit at roughly 20,200 kilometers altitude, far outside the bulk of Earth’s gravitational pull compared to a clock at sea level. Being in a weaker gravitational field means their clocks run faster than ground clocks, by about 45 microseconds per day.
| Effect | Cause | Result on satellite clock |
|---|---|---|
| Special relativity | Orbital velocity (~3.87 km/s) | Runs slower by ~7 microseconds/day |
| General relativity | Weaker gravity at ~20,200 km altitude | Runs faster by ~45 microseconds/day |
| Net effect | Combined | Runs faster by ~38 microseconds/day |
The two relativistic effects on a GPS satellite don’t cancel each other out, they nearly triple in the opposite direction, and that surviving 38-microsecond gap is what engineers actually have to fix. worldtimedata
Why 38 Microseconds a Day Actually Matters
Thirty-eight microseconds sounds negligible, but GPS positioning depends on timing precision at the nanosecond level, synchronized against the same UTC time standard that keeps clocks aligned worldwide. Your receiver calculates location by measuring exactly how long a signal took to arrive from each satellite, then multiplying by the speed of light. A timing error of one microsecond translates to a position error of about 300 meters, since light travels roughly 300 meters in that time. Left uncorrected, the 38-microsecond daily drift would compound into a positioning error that grows by about 10 kilometers every single day, enough to make GPS functionally useless within hours, let alone days.
This is why the correction isn’t a software patch applied after the fact. It’s built into the hardware before the satellite ever leaves the ground.
How Engineers Actually Fix It
GPS satellite clocks are deliberately manufactured to run slightly slow, by exactly the amount relativity predicts they’ll speed up once in orbit. Before launch, the atomic clocks are tuned to tick at 10.22999999543 MHz instead of the standard 10.23 MHz. Once the satellite reaches orbital velocity and altitude, the combined effects of special and general relativity bring the clock rate back up to exactly 10.23 MHz as observed from the ground. The deliberate slowdown and the relativistic speedup cancel out almost perfectly.
This detail matters because it shows relativity isn’t a theoretical correction applied to already-working technology. It’s a design requirement, calculated in advance, without which the system wouldn’t function as intended from day one.
What Happens If the Correction Is Turned Off
This isn’t hypothetical. GPS’s own engineering documentation, published by physicist Neil Ashby in his review of relativity in the system, confirms that early satellites included a switchable frequency offset specifically so engineers could compare performance with and without the relativistic correction applied. When it was tested without the correction, the clock drift matched Einstein’s prediction almost exactly, and positioning accuracy degraded exactly as the physics predicted. It remains one of the most direct, continuously running confirmations of general relativity in existence, not in a laboratory, but in a consumer technology used billions of times a day.
Beyond GPS: Where Else This Shows Up
GPS is the most visible example, but time dilation applies to every satellite system that depends on precise timing, including GLONASS, Galileo, and BeiDou, each built by a different country, each independently required to solve the same relativistic correction. Atomic clocks on Earth’s surface show the same principle at a smaller scale: in a 2010 NIST experiment, two optical clocks placed roughly 33 centimeters apart in height, one about a foot higher than the other, showed a measurable difference in tick rate, confirming that even a small change in elevation changes how fast time passes.
The core idea connects directly back to a more basic question worth understanding on its own: what time actually is, once you stop treating it as a fixed background and start treating it as something gravity and velocity can measurably change.
The Bigger Pattern
What makes the GPS case compelling isn’t just that relativity is real; it’s that the two effects push in opposite directions and don’t fully cancel. If Earth’s gravity well were slightly deeper, or the satellites orbited slightly faster, the correction would need to run the other way. The 38-microsecond figure isn’t a universal constant; it’s the specific outcome of GPS’s specific orbital design, which means every satellite navigation system has had to run its own version of this calculation independently. The same underlying math that governs how atomic clocks measure time down to the second is what makes it possible to measure a drift this small in the first place.
Time, in other words, isn’t the fixed backdrop most people assume. It’s a variable that depends on how fast you’re moving and how deep in a gravity well you sit, and GPS is the proof running in your pocket right now.









