I thought it would be a lot smoother than that.
still jerky!
Thank you anyways
Hi its proven to be 10hz smoothnessâŚ
As if u use the floor fomula to move second hand it be same as if you set 10hz on the second hand
Not much difference. Sorry.
I use ur formula too.
Thank you knughtwing.
See this threadâŚthere is a difference just u cant see it
If without floor you will see strobing as white and black hands are moving differently.
With floor both moved at exactly 10hz so the white second hand is covered and u dont see it.
If @Mayalona notice something different then its you.
1 Like
Hi, I checked the two formulas, inserting mine on the top plane and the one with âfloorâ on the bottom plane. Seen together, there is no difference, probably the 10Hz is not calculated.
I attach a copy of the watchface to replace the .zip extension with .wfs.
lebanon Copia.zip (1.1 MB)
1 Like
Hi @Mayalona , cannot be its the milliseconds updates at every 100millis, if you want to test u should u do 2 text colors align togethetâŚu will notice instead of getting one color text u get 2(blinking). Instead of one.
This is because one is chasing the other.
Or your multiplier by 60 already fixed itâŚto prove it
Please u 2 text alignment one with old tag, another with my.
See if values are kept at one decimal place.
This was a long discussion so i cant be wrong.
You are in that thread
1 Like
@russellcresser if you can examine that wfs see if my tag got brackets errors. Or the improvements is to difficult to see
Without test mentioned
I hope I have interpreted your suggestions well, in the attachment I have put the two planes with the same characteristics with the exception of PLANE 1 in red and PLANE 2 in white and respectively PLANE 1 without âfloorâ and PLANE 2 with âfloorâ. To me there doesnât seem to be any substantial difference, they go in spurts at the same time and neither of them chases the other. Of course Iâm also in the discussion you attached but you were referring to the second hands, which perhaps the GW4 manages differently.
Possibly if you have a copy to send me of your test to show me that there are differences in the use of the two formulas, I would be grateful (just to understand how to apply it)
Attached is a copy of the test to replace the .zip extension with .wfs.
lebanon Copia.zip (1.1 MB)
Nope your conclusion is wrong.
Like i said one is actual second hand.
The other
is a tag control image so should work the same.
So what we are comparing is actual 10hz vs the formula
As the hang is setable to 10hz.
So in this example if the formula is not 10hz the hangs will not match speed
sorry you did correct but do you see the yellow line and the back the red plane?
thats the effect of âchasingâ you shouldnât be seeing the yellow line if really no different in the two formula soâŚ
Hi, do look at preview in wfs to see the difference.
As the preview refresh rate also is a mystery.
Compare on the watch like how we did in the thread
Maybe now I understand what you mean to tell me. I have seen both in motion and when stationary there is a small difference. And checking the differences under the âmicroscopeâ I only see the numbers on the âPLACEMENTâ. On Plane 1 without floor the data on X is all decimal, while on Plane 2 with floor the data on X is all integers. Beyond this small deviation on the starting X axis then to the naked eye they both go the same way. I donât see any âlengthening - shorteningâ effect of the distance between the planesâ fins. The distance remains constant. I also looked at the other discussion and in fact the difference is noticeable on the hands.
I really canât explain these differences.
Ya youâre right Mayalona, as I said earlier, there is no difference. still choppy.
Thank you.
Hi, all sent it to your watchesâŚlike i said the preview are not that accurate.
And the actual values in x axis is up to 0.1 only. You see the 0.1 number of old formula being more jumpy then whole number
Or probably your formula already make it 10hz, and that what it should be.
As x60 will remove 1/1000 digits
And pecisiion becomes 1/100 (sometimes constant increaments sometimes random) (this cause the number of pixel to move to be jumpy)
Add the floor u get precision up to 1/10
Which is what is needed
It will not be as smooth as 15hz.
See this thread
Hi, if you dont believe my formula, then suits yourself. I am trying to help. But if seem no help then bye!
@Knightwing @Mayalona I appreciate both your efforts! thanks for helping me out!
Hi @Mayalona why your formula is x60 %1200?
Maybe to make it smoother is every 100ms move less pixel alsoâŚ
Now every 100ms is 6pixelsâŚ
I saying this because if each frame is closer to each other will it be not such a big jump for frame by frame?
Hi, x60 is basically the speed I assign to the plane at the top, x120 is the speed I assign to the plane at the bottom. Having two different speeds makes it possible to differentiate the reading better, especially in the news programs which show the scrolling text of the news and that of the trend of Wall Street. While %1200 repeats the step every 20 seconds and %1800 repeats the step every 15 seconds. in both cases it is used to have more or less reading length of the texts. In my case, I assigned the top plane a next event complication, while for the bottom plane I assigned the INFORMER complication to read notifications (especially long WhatsApp and Outlook messages).
Now Iâm trying a âmergeâ between the various formulas, I found another one but it doesnât work on images. I am attaching both my wfs with the two planes, in order to highlight what I have (even if it goes in spurts I am of little interest because I can still read) and the other with the 3 planes with the relative formulas combined, as well as the detail of the movements in X.
Rightly, as you highlighted, they move at 6 pixels, but unfortunately I need these speeds and above all I noticed that all three formulas in the text boxes move almost synchronized. it could, in my opinion, be a non-modifiable management by the clock CPU.
asd.zip (4.5 MB)
Copia new.zip (1.1 MB)
Hi, its nothing to do with that really its to do with refresh rate of cpu is at 100ms per updateâŚhas been provenâŚ
So if mathically your final values comes out to be in
E.g 1.0, 1.1, 1.3 (actual value before being rounded of by x position field)
It be exactly 10hzâŚso if u dont see a different is its because somehow both formulas give u 10hz.
And jumpyness is something else not the 10hz issue
As the video in my thread explainsâŚ
For example if we use x 1 instead of x 60
1.190
1.245
1.356
1.456
If you look closerly on my video u know what i mean
1.4xx (xx are seem random not constant increasing by 1)
Only the 1.x is constant refresh every 100ms
@russellcresser has tools to go frame by frame and he notice that too.
For example so pixel moved will be
Could be
1.2
1.2
1.4
1.5
See where the jump may happen different?
So when i floor i get
1.1
1.2
1.3
1.4
1.5
Maybe u dont know what floor does?
It round down to not decimal number.
That means my formula makes
floor([SEC_MSEC]*10) will make
1.124 = 11
Then divide by 10
= 1.1
Making the frame stay at a pos for 100ms before moving to the next 6pixel and not random pixel move.
which makes the reading consistent 
Maths dont lie
And when u say it sync on the watch or wfs preview
My second hand experiment in preview was strobing (so i thought floor didnt work also)
But when i run on watch it became a single second hand. But i was using x6 per second
So i will repeat dont base your conclusions on wfs preview but on the watch
And if u compare with something that already changes the precision to close to what my formula does it be same effect. And maybe both is correct in the first place. And u aiming for higher hz which is not possible. I am trying to get a cleaner 10hz. And if still jumpy to u. It means 10hz is as is.
Because by doing a x60 your values will be
Without floor should be
E.g 1.xy and y is not very random compare to the last digit (1.xyz) (z)
So maybe it more closer to 1.x values.
Which make it more constant movement then if u didnt x60 but x6 instead. (This could be why u seem to see it sync)
Anyways samsung already says for element beside second hand max is 10hz due to batteryj
1 Like
Now I understand how your formula works. Now weâre here⌠Thank you 
