Physicists Find That As Clocks Get More Precise, Time Gets More Fuzzy (sciencealert.com)

Physicists "have combined two grand theories of physics to conclude not only is time not universally consistent, any clock we use to measure it will blur the flow of time in its surrounding space." An anonymous reader quotes ScienceAlert: A team of physicists from the University of Vienna and the Austrian Academy of Sciences have applied quantum mechanics and general relativity to argue that increasing the precision of measurements on clocks in the same space also increases their warping of time... [W]hile the theories are both supported by experiments, they usually don't play well together, forcing physicists to consider a new theory that will allow them both to be correct at the same time...

In this case, the physicists hypothesized the act of measuring time in greater detail requires the possibility of increasing amounts of energy, in turn making measurements in the immediate neighborhood of any time-keeping devices less precise. "Our findings suggest that we need to re-examine our ideas about the nature of time when both quantum mechanics and general relativity are taken into account," says researcher Esteban Castro.
The article opens with the statement that "time is weird," noting that despite our own human-centric expectations, "the Universe doesn't have a master clock to run by."

Comments

Re:It's all a simulation

[goombah99]

It's interesting that all the funny bits of quantum theory and relativity and light are infact identical to what you would expect to be the rules of any simulation.

For example, if you aren't looking at something in a video game it doesn't get rendered, ergo schrodingers cat like phenomena. The moon in fact is not there if you don't look at it.

Bells theorem rules out local hidden variables (that is variables that are in the game but are not coupled to you the observer) but it allows global hidden variables to explain all spooky action at a distance by means other that quantum entanglement. that is to say it's what should happen in any simulation in which you are part of the simulation too.

diffraction and the heisenberg uncertainy relationships come from discrete binning. For example, in a pixelated universe you can'e actually resolve angles of far away objects since they are pixelated. hence there's a direction-position uncertainty.

Likewise the more finely you allow a simulation to measure time the more finely you have to bin or divide the external clock requiring more energy.

Hmm...

[Whooty McWhooface]

So if their clock turns out not to be accurate. it's the universe's fault?

It's a poor scientist who blames the universe for their shortcomings.

Re:Hmm...

[Anonymous Coward]

Or it's Heisenberg up to his usual antics. Time and energy appear as conjugate variables in the quantum wave function solution to the Schroedinger equation for an oscillator (like a ticking clock), so the precision of your clock (delta-t) is inversely proportional to the precision of your energy measurement (delta-E), in the same way that the precision of position and momentum measurements are limited by the uncertainty principle.

Energy curves its surrounding space under General Relativity. This would imply energy of whatever system does the ticking in your clock is constantly being "measured" by, at a minimum, the fabric of space-time, independent of how well you isolate it from the rest of the clock. So that puts a limit on the uncertainty in the energy measurement of whatever does the ticking. If delta-E is limited to be below a certain size, then delta-t is forced to be above some size, so you necessarily get some small variation in the time between ticks of the clock.

This results in a tradeoff between precision and accuracy. Precision requires many small ticks, so delta-t makes up a larger fraction of the duration of each clock's tick. A clock which ticks less often becomes more accurate (delta-t is a smaller fraction of the total time between ticks), but fewer ticks limits the precision of your measurement.

At least, as a physics grad student, that's how I've interpreted the result that TFA is utterly failing to convey properly.

Forget time

[little1973]

https://arxiv.org/abs/0903.383...

Time is just the sequence of events. And events on the smallest scale are particle interactions.

What if there is an unknown quantum field which creates a barrier between particles? And particles have to "tunnel" through it in order to interact?

When this quantum field is more disturbed (warped, etc) this barrier will be greater and it would be harder to particles to interact with each other. The end result is "time" slows down since the number of interactions drops.

Note that an observer (in its own reference frame) will not notice anything (in the same way as in general relativity) since the observer just counts the number of interactions. To that observer the same number of interactions means the same amount of "time" passes.

And yes, this means this quantum field would be a distinguished reference frame.

Re:Heisenberg uncertainty?

[See Attached]

Are you saying the Planck Constant is not constant, or that the observed frequency is not constant? Perhaps the closer we look at something, the more we are likely to observe variability? Maybe we are sampling over a smaller number of events/atomic-interactions or too short of a time slice? There are few things that are absolute, so we use the observed average as a constant, but in reality, its a curve-distribution.

Re:Heisenberg uncertainty?

[goombah99]

All the uncertainty relationships in QM come from fourier conjugate variables. So for example, if you measure a low frequency for a short time you will be uncertain about the exact frequency. If you restrict a wave to a narrow slit then it take more direction forier terms to represent the truncated plane wave.

time and frequency are fourier conjugates. and plank's constant, which is constant, has the units that convert frequency to energy. This is why we say that time and energy are conjugates.

Makes sense

[Solandri]

Einstein's theory of relativity tells us that time and space are the same thing (your perception of the two skews with your relative velocity, which causes all of relativity's time dilation effects). So I would expect there to be a time-corollary of Heisenberg's Uncertainty Principle. Just as extremely precise measurements of position lead to poor measurements of momentum, extremely precise measurements of time should result in poor measurements of... something else.

Re:Makes sense

[Baloroth]

So I would expect there to be a time-corollary of Heisenberg's Uncertainty Principle [wikipedia.org].

There is, but it's probably not what you're thinking of. Technically speaking, the Heisenberg uncertainty pair applies to any two pairs of non-commutating quantum variables (or, depending on how you look at it, any two Fourier partners). Position and momentum happen to be one such pair. Another is time and energy. What that means, however, is that the energy of a particle in an unstable state (i.e. a state that can spontaneously decay into a lower energy state) is not perfectly well-defined, and the variance in energy is inversely proportional to the average decay time. In other words, the faster a particle (or state) decays, the wider the range of energies that particle/state is allowed to have, so that only long-lived states of physical systems have well-defined energies (by "long lived" I mean something like microseconds or even nanoseconds, which is long by quantum standards).

In the case of time measurements, this would generally mean the energy of our clock becomes less well defined as we make more and more precise measurements of the time. That's not really a problem, though: we just have to be greater that 1/2 h_bar, which is ~3e-16 eV*s. That means if the uncertainty in our time is 1 part in 10,000,000,000,000,000 (modern atomic clocks are very roughly in that range), we have an uncertainty of about 1 eV in the energy of our state. That's decently large (in terms of atomic scale physics), but pretty negligible in terms of everything else (nuclear physics involves energies a million times greater than that).

Re:It's all a simulation

[Greyfox]

Yes! And time is so problematic because the simulation is being run on a massively parallel system. While each processor is able to handle the physics and timing of a small area reasonably well, keeping time synchronized for the entire universe would slow the entire thing down far too much. Moreover, since the project was designed as a simple demonstration of how to convert hydrogen to plutonium over time, making an effort to do so was deemed unnecessary. We also had a problem with some particles being uninitialized upon creation and going off at a very high velocity, so the top speed in this particular universe simulation was capped to prevent anything too untoward from happening.

The simulation has been running reasonably well for the amount of effort put into it, although there are still some issues of localized processors crashing when mass values in specific locations go too high, and some number of processors have been having to synchronize their timing signals across boundaries for reasons we do not currently understand. There is also the minor issue that eventually the plutonium degrades back to hydrogen, along with everything else, but we had no intention of ever allowing the simulation to run that long anyway.

Semantics matter

[93 Escort Wagon]

"Physicists Find That..."

Given that this wasn't a presentation of new research data, but rather an argument attempting to reconcile two theories - it is incorrect to claim that they "found" anything. Replacing that word with "argue" would fix that.

Although perhaps there's a Slashdot corollary to all this stating the more accurate a headline is, the more fuzzy the linked article will be...