Lost in progress

[tggzzz]

I have seen tickless clocks, where the seconds hand advances smoothly and silently like the others.

Except for Vetinari clocks. But you can have a digital Vetinari clock too

[nctnico]

1. I'm not sure if I perceive these benefits because I was born before wrist digital clocks were a thing, or if it has something to do with the analog representation being better than a numerical one?

If you want to see trends or need speedy comprehension and don't need much precision, then analogue displays are often better than digital displays.

Canonical examples:

• control panels where all needles should be pointing roughly at the same angle, and you need to quickly spot any that are "out of line".
• artifical horizons, where the changing angle is much more useful than a succession of numbers

Conversely, I prefer the digital speedometer that many cars offer now.  Especially because a speeding offence is defined as an absolute limit (while there is leeway in the UK, this is not uniform and not legally stipulated - 0.1 mph over *is* an offence).  But I'd agree analog readouts are better for most other things.

Check the permissible errors in the values a speedometer displays Precision != accuracy

afair -0/+10% and that has too account for the size of the wheels so it always show on the high side

Yes. A speedometer may never indicate too low. In my car the speed is consistently 8% too low up to 140km/h (on the dial which is about 130km/h for real). Above the error seems to get a little bit smaller.

[magic]

For me it is digital clock all the way..
I intuitively know where I am in time based on numbers.

I wonder if you really do or if it's just Stockholm Syndrome?
How much time is it from 11:47 to 12:09? Surely you can do the math, but on analog you simply visualize the angle and it can be as meaningful as the number 22.

Of course for any sort of high precision you need the numbers, and digital provides them directly, so it wins there.

[RoGeorge]

I have seen tickless clocks, where the seconds hand advances smoothly and silently like the others.

This one ticks, in fact it's quite clever, it has 2 stepper motors, one for the seconds indicator and one for the minutes indicator.  the minutes steps precisely 6 times to advance one minute, but the seconds and the minutes indicators have no common gears between them.  The hours indicator is driven by gears with the minutes indicator.

The clock has sensors to independently detect 12:00 for the minutes arm and for the seconds arm.  Therefore the alignment is always perfect for this particular design.  It's funny to observe them moving independently at power up. 

There is a routine at power up, each arm seeks the 12:00 position and align themselves (the clock has internal optical sensors for 12:00), then the clock seeks for the signal of a DCF (from Germany) or a MSF (from UK) time radio-beacon to auto-position themselves at the current hour. 

Here's a 3 minutes video showing the PCB, the gears and the steppers dismantled:

[themadhippy]

The digital clock haters would have a heart attack if they saw my kitchen clock

[Siwastaja]

The digital clock haters would have a heart attack if they saw my kitchen clock

You are still a prison of that ridiculous 365-24-60-60 system. How about unix timestamp instead? Maybe in hexadecimal if decimal is too normal.

Unix timestamp like second count which resets every midnight would also be interesting.

Or replacing that weird 60 thing with 100 (us mere mortals) or maybe 256 (binary thinkers). Or redefine a second so that instead of 86400 metric seconds per day, you'd have 65536 of New Seconds. 256 New Seconds would form a New Larger Time Unit, 256 of which would be a full day. Time of day could be represented as a simple 4-character hexadecimal, e.g. 7FFF would be just before noon.

[2N3055]

For me it is digital clock all the way..
I intuitively know where I am in time based on numbers.

I wonder if you really do or if it's just Stockholm Syndrome?
How much time is it from 11:47 to 12:09? Surely you can do the math, but on analog you simply visualize the angle and it can be as meaningful as the number 22.

Of course for any sort of high precision you need the numbers, and digital provides them directly, so it wins there.

Intuition is learned.
In your example I don't do real calculation. It is more like 10 min before and 10 min later so 20 min something (2min error is less than what you get with analog clock and parallax) .... It's guesstimating.

Most of time with estimating time you don't need exact numbers but a time budget. If I need 10 minutes to get somewhere, and need to be there at 17:00, and now is 16:3x or 16:4x something I'm good. If it is 16:5x then I'm going to be late.

With analog clock I need to remember how many degrees or rotation I need to travel and compare it visually to current angle of clock hands and target time. Or you are converting angular notation to numbers, calculate and convert back into angular.. ?

[RoGeorge]

The digital clock haters would have a heart attack if they saw my kitchen clock

That's a nice clock! 


Here's one to give a heart attack to any haters, analog or digital alike! 
3D printed sundial displays the time digitally

[shapirus]

Unix timestamp like second count which resets every midnight would also be interesting.

I think you just gave me an idea for my next electronics DIY project.

[tggzzz]

For me it is digital clock all the way..
I intuitively know where I am in time based on numbers.

How often do you need to know the time to the minute?

Very rarely do I benefit from knowing the time is 10:24; usually "coming up to half past 10 is sufficient".

For measuring intervals, digits can be of help. But you don't need a clock for that.

[RoGeorge]

Unix timestamp like second count which resets every midnight would also be interesting.

I think you just gave me an idea for my next electronics DIY project.

Tinkering around with Unix timestamp as we speak (because the DCF77 signal is too small here, and I need to have an easy way to set the clock, so I'm trying to make a local DCF77 Tx).

This is another complaint about features lost in progress:  manual override (if there is any) usually sucks.

For example this wall-clock has a single press button for manual set.  You have to keep it pressed for 3 seconds, and the minute+hours start to move. Only that the move is slow, and takes 1-2 minutes to sweep all the 12 hours.  It moves only one direction, so if you pass over the right hour, you have to wait 2 more minutes.  And during all this time you have to keep pressed on the back of the clock a minuscule button, with the tip of the pencil, while looking at the front of the clock.  You can't tell if you press too hard or too slow.  And once you get to the desired hour, the seconds start ticking at a few seconds after releasing the rubber button.  It doesn't wait for a manual start.  It's impossible to match the seconds properly! 

We've lost proper knobs and buttons, the ones with good tactile feedback.
We've lost the potentiometers, which used to have an absolute position, for never ending rotary encoders.
We've lost forward control and replaced them with tedious visual control loops, where you have to dial a little more, read a number, then dial more maybe, and so on.

Sure, the new controls have their advantages, and can do new things, but we've lost some very good/basic features that were always there as a side effect of the physical objects.

[themadhippy]

That's a nice clock

cheers,its part of a long term project that was ignited  by a throw away comment by my school electronics/physics teacher 40+ years ago about binary time.Theirs  also  its circular brotherbut thats suffering from loosing about a minute a day so needs some attention.And an early build, complete with  genuine 80's  parts
The half baked plan is a fully binary calendar,only using good ole cmos logic

[RoGeorge]

 

[rstofer]

I spent a lot of time watching analog clocks in elementary school!  When will this class EVER BE OVER>

Many analog clocks use a universal mechanism and there are silent models.  I have changed out a couple of mechannisms strictly to reduce noise.  Some come with replacement (more stylish) hands.  Scroll down on the page:

https://www.amazon.com/XYBHRC-Mechanism-Replacement-Operated-Movement/dp/B0B7NS5MBW

There are many others, Google is your friend.

[RoGeorge]

Against the ticking noise of a plastic gears clock, a few drops of oil can reduce the noise close to nothing.
I don't know why they don't grease them from the factory, or at least put some holes in the case, for oiling.

[james_s]

I also prefer analog dial clocks generally, with a digital clock I have to read the numbers and process what they mean, with a dial clock I just glance at it and instantly know the time, I don't have to read anything, it's visual.

Same reason a lot of race cars have a row of colored LEDs instead of a tachometer, the precise RPM doesn't matter, the important thing is knowing when you're in the power band and when you're approaching the redline. It's faster to interpret the bargraph than a tachometer.

[mwb1100]

The recent complaints about the second hand disregard what I think seconds hands are usually used for.  It's not so that people can know the absolute time down to the second - it's so you can tell when some number of seconds have passed.

Timeless questions like:

  - who can hold their breath longest?
  - how long can I hold my hand over this candle?
  - how fast can I chug this beer?

PS: looks like the Svalbard runs about $150.  Might have to get a new watch.

[SiliconWizard]

I also prefer analog dial clocks generally, with a digital clock I have to read the numbers and process what they mean, with a dial clock I just glance at it and instantly know the time, I don't have to read anything, it's visual.

It's pretty much a matter of what we've been exposed to as kids I think.
Now it's possible that "processing the numbers" may take a tiny bit more effort than processing the dial view for all of us. Or maybe it depends on each individual. Not sure.

One thing that works much better with "analog" displays in general is evaluating variations at a glance.

[james_s]

I grew up in the 80s, digital clocks were common by then and we had several in our house, along with several analog clocks. I don't feel all that strongly about it but I do think analog clocks are attractive and when I want to know how long it's been or how long until some event they are definitely faster for me since I don't have to do math. Some of my favorite clocks are nixie tube based which are of course digital. I also have a liking of weird clocks, things like binary clocks and other quirky stuff. I like clocks in general, they can be a combination of kinetic artwork and functional tool.