Battery and steps are in the vumeters. The correct value of rotation has a (-2,2) rand variation.
The numbers in red lcd has a (-5,5) rand in the transparency.
Any thoughts and idea are welcome!
If it was more Steam Punk, and you made the (random) variance more cyclic … it could work as is. It starts to look silly after 15-20 seconds, and you might notice that after wearing the watchface a few days.
I would be more inclined to try to match a voltometer on startup, with it starting on left/zero (at wakeup) and slowly moving to correct position (with diminishing over/under movement) … So, it is finally steady after 3-4 seconds.
And the -ve and +ve signs don’t really make sense in current context.
haha first time I see it… thanks!
Yes, that’s the idea, next step on the work…
I had to google the reference, not familiarized with it, but yes again. I will change the bright variation to a sin formula. And still looking for slow down the frequency of the variation in the needle…
Thanks for the feedback!
In my old watchface I tried to simulate the operation of a bar-shaped vu-meter and the intent was to obtain semi-random movements that last longer than a calculation cycle: for this purpose I tried to combine a formula based on thousandths of a second “dirty” with a random component like this:
(Rand (0.1) * 12 + 150 * (1 + sin (#Dsm #)) / 2)
It’s not easy but I guess your intention was to see the hands of your vumeter swing and not “vibrate”.
A road could be to create a trigonometric function based on seconds but with results so to speak unpredictable so as to obtain a forward-back value with amplitude each time different … I do not think it’s impossible but I abandoned the project!
Originally the intention was to vibrate, I was studying the application of rand as variation of a correct tag value. But it was too fast. Now I use sin function to swing the needle and rand to amplitude of the swing and look better (still no apply the changes to this draft). In the numbers bright the sin fuction nailed it! Now Im stuck with the interpaccel issue to start of dead needle to movement.
Thanks fo the feedback…
Advance of the work, new draft:
I´m using this formula now for the steps needle:
((60-((clamp(#ZSC#,0,10000))/83.33))+(rand(1,2)*sin(#DWE#*10))))
where
(clamp(#ZSC#,0,10000) prevents outscale when users walk over 10k steps.
(rand(1,2)*sin(#DWE#10)) is the variation, sin for the frequency and rand for the ampplitude (if you wanna just to swing, replace rand (1,2) for a constant).
Similar without the clamp for the batt needle…
Also, now I´m using this formula in transparency for the pulsating bright in the numbers
(90+(10sin(#DWE#*5)))
Next step of the work, add the interpAccel to the needles formula, or split the animation in two layers: one 0-2 seconds from DIM with interpAccel and the other starting in 2 seconds with the above formula
Finally, I opted for the second option, split the animation in two. Worked ok.
Added interpAccel to the transparency of numbers and the scale of weather icon.
Added text to complete the design. Will change the name to “Slow Start Device” or something like that
Thanks for the feedbacks!
Note: in the web preview everything seems tooooooo slow. In watch and phone app preview work ok.
Very nice improvements, @carlosfilippa. One thing - i would add a Doublepoint at the time display between hours and minutes. It would be better to recognize it…
This looks great and I love the texture. I think if you added a small beveled edge around the number LCDs it would make it look even more realistic. Either way though, love it.