The Last POWER1 CPU on Mars Is Dead
talospace.com
talospace.com
If you take a reference with you on a journey through the solar system, that reference will stay accurate for an observer travelling with it and drift as seen from an observer that stayed behind. Both views are accurate. Locally, the amount of time that has passed for each observer was different. If you would correct for any observed drift, you would mess with your fundamental unit of time in your local frame of reference and get incorrect measures of time passed in your own frame.
An example we already deal with is GPS satellites orbiting Earth. Their clocks tick a little bit slower on account of how fast they're orbiting, and tick a bit faster on account of being further up Earth's gravity well. The gravitational effect is stronger, and the net effect is that a GPS clock advances an extra 38 microseconds over the course of a day (as measured by a clock on the ground).
http://physicscentral.com/explore/writers/will.cfm
So when you try to standardize on "seconds since the epoch," the inevitable question is "Seconds according to who?"
Because of relativity, clocks on other planets would very slowly drift relative to UTC on Earth. But the drift is on the order of a few parts per billion (see e.g. [1]) which is comparable to what you'd expect anyway, even from a very high-quality temperature-controlled crystal oscillator. So it doesn't add any new clock synchronization difficulties that you wouldn't have anyway.
For the most demanding applications -- the ones that require atomic clocks -- you would still need to take relativity into account. You can either measure time passing at the local rate (in cases where you need to know locally elapsed time to high precision) or you can measure UTC, which allows you to assign a consistent ordering to events on different planets. But for most ordinary purposes, the distinction is irrelevant.
[1]: https://space.stackexchange.com/questions/33590/time-dilatio...
I suppose the more noticeable oddness for most people would be that the speed-of-light delay in communications from Earth to Mars varies so much. Around 3 light minutes up to 22 depending on where they are in their orbits.
It may need to be compensated for when receiving transmissions between the planets, but I don't know enough about RF to judge how meaningful the difference is. There was a spacecraft (I think Huygens?) that actually had problems with Doppler shift. I think in that case they forgot to take it into account entirely, so it wasn't that they just forgot the relativistic component.
According to a very accurate pulsar, of course.
You would simply also correct for relativistic effects caused by the orbits of different planets, and boom, you have a pretty consistently defined time that is valid across the entire solar system.
Relativity compensation in GPS satellites does not increase spatial accuracy.
https://www.w3.org/TR/html52/textlevel-semantics.html#the-ti...
This fundamentally cannot exist. Any "universal" system would drift for observers on different planets. Having different times in New York and New Delhi isn't the worst right now; doing a calendar look-up for the time on Mars isn't as neat. But given we won't have real-time communication, it isn't as big of a problem either.
Fantasy use case: Asteroid mining and traffic control. If you want to boost prepared asteroids to an orbit closer to Mars and you had a space station / spaceships to watch out for, different mining companies might want a clock protocol, a request buffer, and a map instead of synchronously planning, timing, and verifying each and every asteroid orbit change in a central location ("What do you mean you were using Earth time and not correcting for relativity??").
https://en.wikipedia.org/wiki/Royal_Observatory,_Greenwich#O...
For example we are not just in the sun’s and earth’s gravity field but also the other planets as well. So, with enough precision you would need to account for their locations.
Edit - I looked it up - it's "a clock that performs exactly the same movements as the Solar system but is outside the system's gravity well."
The second, symbol s, is the SI unit of time. It is defined by taking the fixed numerical value of the caesium frequency {\displaystyle \Delta \nu _{\text{Cs}}} {\displaystyle \Delta \nu _{\text{Cs}}}, the unperturbed ground-state hyperfine transition frequency of the caesium-133 atom, to be 9192631770 when expressed in the unit Hz, which is equal to s−1.
Interplanetary internet designs and peering are interesting problems to think about though.
https://en.wikipedia.org/wiki/Intergalactic_Computer_Network
[0] https://en.wikipedia.org/wiki/CCSDS_File_Delivery_Protocol
Mostly from idle curiosity. Was wondering a few days ago and guessing "probably yes", but no idea who to ask.
So, taking the opportunity (heh) now. :)
Wikipedia say MSL used VxWorks as well.
I’d have to ask a buddy still doing Mission Ops Support at JPL but I’d think the answer is no SQLite on MER or MSL missions. The OSes were fairly custom, non-POSIX builds. The ESA folks might have tried though on their missions.
I hear this argument a lot for why we cannot have a universal time standard but it's a weird cyclic argument because it suggests the measurement of time has to be constant while acknowledging that time is relative. So why can't we have a measurement that is also relative?
For example it could be n "ticks" where mass and speed is at a predefined value and allow for decimal points to account for people travelling faster (for example). This means people's life expectancy (for example) could not be measured in "ticks" because "ticks" is used for time synchronisation rather than measuring the passage of time and those who are only interested in the passage of time can use local time zones just like we do now with UTC for keeping equipment synchronised but local time zones for scheduling our human lives.
To be honest I'm really not qualified to be making these sorts of comments so I'm going to back away from the keyboard and if there's any merit to the idea at all (which there probably isn't) then someone else who understands this stuff better can chime in.
This is actually not that a big deal, in many countries, like Afghanistan and Iran, people observe 3 different calendars for different purposes and they all drift from each other! That is, 1st Jan (Georgian) is not always 24th Jaddi/Dey (Afghan/Iranian - Solar) nor 7th of 7th Jumada i-ula (Lunar/Islamic)..
So people use calendars that look like this for each day: https://en.wikipedia.org/wiki/Rumi_calendar#/media/File:1911...
Martian days being around 37 min longer, you might need more leap days though.
The weird thing with Mars would be that if you contact someone at 5pm local today it might not be okay to call them at the same time tomorrow, because they would clock in 37 minutes later.
Right, a whole 24 hours ;)
> if you contact someone at 5pm local today it might not be okay to call them at the same time tomorrow, because they would clock in 37 minutes later.
That is assuming that people on mars will have 24hr and 37 min clocks instead of 24hr.
The only thing that will be different on Mars would be that the sunrise will change 37 minutes everyday. Weird indeed.
Everything about it, including the great distance, or perhaps especially the great distance says that Earthers will have to make the affordances. I expect anyone on Mars who didn't come through the ranks of the Armed Forces (and half of the ones who did), and any medical, psychological, or political scientist brought in as an advisor, are all going to give a big ol' "Fuck You" to anyone who suggests otherwise.
Setup a transmitter at earth which emits a signal at uniform intervals containing current time (seconds since kanye west's marriage). An observer can then deduce current time if it knows where it is. I don't think relativistic dilation would effect this method.
With multiple such emitters synchronized with each other, one might even be able to 'triangulate' the current time. (I haven't worked out any details)
We've known since the Viking landers of the 70s that clocks on Mars don't run the same as on Earth, and that's not even accounting for the speed-effects as the Earth keeps "lapping" Mars on their respective race around the sun.
Universal time is, of course, impossible in a relativistic universe, but (no sarcasm) that doesn't prevent us from trying, and getting to "close enough to work with for reasonable amounts of effort". Just as only a small portion of the world works on UTC directly right now, this time standard would be something computers would translate for people. Traveling at 80% the speed of light away from Earth in Earth's frame means you wouldn't want to directly use the resulting definition of a Standard Second for human time, but computers would be able to work with it.
I don't know what the orbiters use, presumably some variant of Julian day.
[0] https://en.wikipedia.org/wiki/Barycentric_Coordinate_Time
The fact that this phrase of yours is literally true, and it's already 2019, makes me very sad.
this made me think of Swatch Internet Time, and whether or not it could have worked.
That's a complete amateur physicist taking a stab at it!! I'm probably horribly wrong :)
Even if you did this, your universal time would just be a synchronizing tool. Each human consumer would expect it to be translated into the most relevant time to them. If you think it's hard getting people to adopt a standardized date/time on one planet, every planet you add to the system makes that an exponentially more difficult task.
Vernor Vinge got there way ahead of you, If you haven't read his books and given the way this community slants you are missing out.
> Take the Traders' method of timekeeping. The frame corrections were incredibly complex - and down at the very bottom of it was a little program that ran a counter. Second by second, the Qeng Ho counted from the instant that a human had first set foot on Old Earth's moon. But if you looked at it still more closely ... the starting instant was actually about fifteen million seconds later, the 0-second of one of Humankind's first computer operating systems.
https://en.wikipedia.org/wiki/International_System_of_Units#...
"The SI unit of time is the second (s): The second is the duration of 9 192 631 770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium 133 atom."
I'm not a physicist, but I think that's a universal definition - the rate of the periods should be constant everywhere in the universe.
So universal seconds are actually very easy to nail down. Dates are always a little more troublesome. Perhaps seconds since the start of the big bang would be a possibility.
Observer A says that event X happened before Y.
Observer B says that event Y happened before X.
And they're both right from their POV.
It's even useful as a consistency model for some distributed systems.
There's no preferred frame of reference, meaning that any frame of reference could work, but you can derive one from the cosmic microwave background if you don't like picking something too arbitrary.
Im not sure pulsars are actually that precise and predictable to not require regular adjustments to equate local times, but it would provide a standard anybody could utilize as long as they could keep an eye on it.
Of course it would break down if someone broke light speed, either with wormholes or some unknown technology, but as long as you are below lightspeed then there shouldn't be any problems.
Due to their small size, pulsars are relatively weak radio sources. Therefore, the largest radio telescopes in the world are usually needed to observe them. As we have seen, pulsars emit their largest intensity at low radio frequencies around 400 MHz. In particular at such frequencies, however, the pulses suffer from propagation effects when they travel to Earth through the interstellar medium. [0]
[0http://www.jb.man.ac.uk/distance/frontiers/pulsars/section4....
Milliseconds since UNIX epoch!
But one thing I am sure of -- we would make good friends ;)
Though I do see the irony with at the same time shaming imperial units.
We'll highly likely stick with second as the base unit but create base 10 multiples on that .
Also I think the confidence interval for instrumentation looking that far ahead would be so big comparative to current/possible local solutions, we're probably better off doing something in the Solar System instead.
One advantage of synchronizing to some reference in the Solar System, and maybe setting clock boosters as floating checkpoints in space, is we will always be able to find our way home :)
Astronomy generally uses "Julian days" or "Julian years" which are measured in seconds since a certain calendar date. Since they're based on SI seconds intervals are independent of the Earth's movement and leap seconds and all that nonsense.
Most popular epoch is J2000: measured since noon on 1 January 2000.
Mars landers each track local solar time separately and a colony is likely to invent its own timezone with 24 hours and 39 minute days. You need this so that notions like "sunrise" and "sunset" make sense.
We already do that with UTC, and even with something that simple, it gets hard: politics gets in play, and we have to reajust our time slices with leap periods regularly.
UTC (and international atomic time - TAI) is an average measure of elapsed time that already needs adjusting for relativistic time dilation. None of the accurate clocks used for TAI are at sea level but TAI is reported as though you had all your caesium atoms ringing away at sea level [0].
Everyone else adjusts their clocks with reference to this, making allowance for propagation delays where needed. This includes things like navigation satellites (GPS, Galileo, Glonass, BeiDou etc) which have accurate clocks and significant (for their operation) relativistic dilation.
Since we already have to deal with relativity and propagation delay here on earth I suppose that, if your mission can still ping mission control here on Earth, then you can still set the system clock to UTC or TAI [1]. You will need a look up table for local sunrise times and the like but that seems quite simple compared to the mess that tzdata has to allow for back home.
[0] As I write this I realise that I don't know if TAI will be adjusted for sea level changes or not. Anyone know?
[1] Using TAI might save you from leap second problems while landing on an alien planet.
If that's the case Unix time then becomes a decent option, because it's equally human unreadable to everyone. Each planet can then have their own calendar based on that, and computers can deal with the nitty gritty
> That's not the end of Power ISA chips on Mars, though: Curiosity, which is running a pair of RAD750s (one main and one backup, plus two SPARC accessory CPUs), is still in operation at 2,319 Mars solar days and ticking.
Is it certain that the chip is not still getting some power and running something?
(Disclosure: I work for IBM.)
Well, at least they should be able to browse the web with the G3 version of TenFourFox. ;)
For the former, different contractors for the features, with some contractors preferring RHPPC and other RHSPARC?
For the latter, one SPARCs controlled the thrusters and descent stage, the other controls the movement (it's the main chip of the motor controller).
Source: NASA mission update for sols 5340-5346 (Jan. 31-Feb. 6, 2019) https://mars.nasa.gov/mer/mission/rover-status/opportunity/r...
Newer rovers like curiosity are using radioisotope generators which should last a much longer time.