The train is using a week counter internally that overflowed. 17.11.2003 is exactly 1024 (2^(10)) weeks ago.
But this would mean the epoch here would be the 2 April 1984 or maybe the 1st, if you respect the time as well, which sounds a tiny bit random lol. But so would be to use a week counter in the first place :D Especially since those trains are from the early 2000s.
EDIT: Other times in 2003 are also possible apparently https://twitter.com/TAltgeld/status/1646386165006716928?ref_src=twsrc^tfw|twcamp^tweetembed|twterm^1646386165006716928|twgr^99a8805bbdf28d116a3a7037dc24347da978e819|twcon^s1_c10&ref_url=https%3A%2F%2Fwww.kreiszeitung.de%2Flokales%2Fbremen%2Fsorgt-fuer-spott-im-netz-geniale-loesung-der-db-der-zug-aus-der-vergangenheit-92212981.html
How did you get to that epoch? I haven't seen any images of this happening earlier than 2003, so the overflow must have happened late last year or early this year, and whatever week 0 is is some time around 2002/2003. That would also fit with your comment that they were introduced in the early 2000s.
Maybe they just didn't expect the trains to last 20 years 😄
Deutsche Bahn making up 20 years of delays in one go. Now that's efficiency
Ah, Deutsche Bahn. Gotta love it.
Old public clocks sometimes use the mains frequency as a clock signal to maintain correct time. At the end of the day, the power authority is supposed to adjust the signal slightly to make sure these clocks maintain the correct time.
So I built a device that analyzes mains power supply phase variance with microsecond resolution. It's accurate enough that I can see the power authority turning on and off supplementary turbines to keeps the mains frequency correct.
That's how I determine my level of trust in public clocks.
Well, that and maintaining sub-relativistic velocities relative to them. Which is pretty easy honestly. If you're having problems with this, you've probably ceased to be biology or chemistry and started being physics. So you probably have more pressing concerns than keeping time!
Fellow vxjunkie, great to see on the other side
This is a train's clock, though. And since these displays also show additional information, I guess there's a bit more sophistication behind it.
Did you construct the analyzer yourself or can you buy that off shelf?
I built it myself. I used an optoisolator to convert the sine wave of AC power to a series of pulses, and also for safety reasons (it communicates data over light pulses, so that the mains power circuit is completely electrically isolated from the stuff I might touch). Then I used a microcontroller to precisely time the deviation from 50Hz, and output a serial port via MAX232. A desktop computer logs the data and gives me a real-time graph of the deviation. I might have some sample files lying around somewhere.
The microcontroller (ATMEGA16) ran at 20MHz using a crystal oscillator, and with some slightly funky tradeoffs, I was able to get a time resolution of 50 nanoseconds with an error of about 20 parts per million. If it deviates too far from 50Hz, I get nonsense measurements due to the tradeoff though (I trade measurement range for precision).
The original idea was to analyze tiny variations in phase at extremely short time scales to use the mains power transmission lines as a sort of large antenna array. Then use that to study space weather (e.g. solar wind, flares, etc.) more cheaply than sending up satellites. As a hardened skeptic, I originally built it as an elaborate practical joke due to all those 'Mayan Calendar Doomsday 2012' folks. You see if the world did end, I would have been in a position of absolutely dreadful embarrassment. So I built it as a sort of 'doomsday detector'.
Sadly, that component of the project never worked. I ran it during some solar activity peaks, and was unable to correlate the measurements taken by satellites to anything I found. It does still allow some weird things though, like I can apparently create a phase variance 'fingerprint' and use that to conclude whether a video was recorded in my city, and at what time. I've heard of some people doing that to youtube videos.
I can also get an idea of the power draw of the city I'm in. As people draw more power, the phase drops, then the power authority turns more turbines on. When the load lightens, the phase goes up, and I can see those turbines being turned off. This is sort of fun to turn on and watch if you get home early, you can see everyone else getting home and turning stuff on.
Learned that the hard way while stationed in Japan where their power is 50 Hz. The clock lost 10 minutes every hour.
I recall some of Japan being 60Hz, and the rest being 50Hz! That always struck me as pretty wild!
I live in SE Asia. You could say we lose 10 minutes every hour regardless of the timekeeping technology :P
Hooray, it's all been a bad drug induced dream! 18 year old me awakens in regional train.
Ah 2003, I remember it well.
I love German regional trains. This is a very common issue. Happens all the time.
wtf that's literally my birthday
Ah, it's about right give or take twenty years.
God I would love it if I was suddenly transported back to 2003 as my current self.
That's a really old picture you got there ;)
walks into train, completely stoned. Sees clock. Panics. Ah fuck, I'm way too high to travel back in time... What if I fuck up the timeline?!
As we have ultimately have set all of our private clocks with public ones as well, we always have to decide which public clock is more trustworthy in those moments. Except if you use a very precise sun dial.
