Author Topic: solstice  (Read 2668 times)

0 Members and 1 Guest are viewing this topic.

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
solstice
« on: December 21, 2016, 01:16:03 PM »
www.pennlake.us

I do have a post started, in the WD forum under Sunlight Hours,  and some help has been rendered but new it seems to have stalled, so here I am, sorry.

Anyway here is my problem:

My Sun Pie chart still reads 01 second shorter that yesterday. Well the solstice happened in my area, December 21, 2016 at 5:44 am EST, so I believe my Pie Chart should read 00 sec

I have Updated to the latest WD version which I guess corrected the problem for everyone else (but of course me, lol). I have also tried, to the best of my knowledge, recheck all the long and lat settings, time settings on my Davis console, WD, Win 10 to make sure things were in sync.

Is there a switch I am missing a settings some place? In WD or Wx? The only thing I did not do and I will do right after this post is to reboot everything.

Thanks, for any help, suggestion, or ideas.

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #1 on: December 21, 2016, 01:17:46 PM »
Question is if this is calculated in real-time or eg. once a day

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #2 on: December 21, 2016, 01:22:07 PM »
Hi Jachym,

Well that's what I'm kind of thinking, but I don't know where to look on when it gets created and when it uploads.

I was keeping an eye on it hoping that perhaps after (my time) 6am and noon it would update, I can easily live with that, but so far nothing.

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #3 on: December 21, 2016, 01:25:22 PM »
It doesnt really matter when it updates, but it is possible that WD simply calculates this at midnight and does not check every minute if solstice just passed. It is probably showing you the difference between today and yesterday at midnight, which sort of makes sense, so I would wait until tomorrow

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #4 on: December 21, 2016, 01:35:34 PM »
I would agree Jachym except others are showing it correctly ex. Here is a station just down the road from me:
http://www.mikeymsweather.com/index.php

showing 00 secs ??

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #5 on: December 21, 2016, 01:43:38 PM »
Hi,
ok, just make sure that that guy is also using WD, because he could be using the same template but with a different SW, which might calculate this differently - just a thought.

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #6 on: December 21, 2016, 01:46:14 PM »
Yes, At the bottom of his page:

MikeyMs Thornhurst Weather | Weather-Display   (10.37S-(b36))

Oh well, Like you said I can always wait until tomorrow.

Offline Scalphunter

  • Forecaster
  • *****
  • Posts: 2314
Re: solstice
« Reply #7 on: December 21, 2016, 03:18:16 PM »
One second difference. That just an mile in travel. You say he down the road  from  you. Just how far down that road is he.    If you really like to  see an problem with all the WX programs live where your suset comes after midnight. It then  tell you that the  day is only 3 hour and some minutes long.  The authors to lazy to  check the math to differate  if an day carries over  the midnight hour. They take for granted the  sunrise sunset is all in one day time.


John

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #8 on: December 21, 2016, 07:15:43 PM »
As an update: it is now after 7pm and my pie chart now has 02 sec longer   --

So my thought is that it changed after 7pm which would put me 5 hours somewhere or at GMT. I am at the -5 GMT so What setting could I look at to adjust this. It just seems that I am in the next day and I should not be there yet.

Is is a WD or Wx or both anyone any ideas???

Offline Scalphunter

  • Forecaster
  • *****
  • Posts: 2314
Re: solstice
« Reply #9 on: December 21, 2016, 08:55:23 PM »
Hard to say as we don't have an clue to where you are located at. What is your time zone?


John

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #10 on: December 21, 2016, 09:05:31 PM »
To be perfectly honest with you, the way this is calculated in general, and in fact also the way you define what a "sunrise" and "sunset" is, means you should usually aim to about 1 to 2 minute accuracy. Talking about seconds is absolutely irrelevant. If I ask you to write down exactly when the sunrise/sunset was and I was standing next to you doing the same thing, I can guarantee you we will not have the same time.

If the script is accurate within a minute I would say it is very good and I dont even see why I should worry about seconds, in particular in a situation where there really is no exact time. It also depends on elevation etc etc.

The script obviously works ok since it was showing today is the shortest day in the year, or very close to that.

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #11 on: December 21, 2016, 09:08:43 PM »
Just to give you an idea of where the "inaccuracy" can stem from:

- slightly different equation used for calculation (and it is actually quite a complicated one with lots of coefficients and math operations)
- rounding - rounding of the various coefficients, latitude
- definition of sunrise/sunset (ie. the exact angle)
- elevation

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #12 on: December 21, 2016, 10:17:25 PM »
Thanks Jachym -- I understand about the calculations, rounding, position and stuff and believe me I was sort of joking about the 01 sec time --

What I'm looking for - and maybe this is also not realistic - is that my solstice happened this morning at something like 5:55am,  my pie chart showed 01 shorter all day until 7pm tonight the it changed to 2 sec longer than yesterday -- My question is how does the pie chart update?  Is there a place to set when and how often the pie chart updates. I mean why at 7pm, if it was every 24 hours wouldn't midnight be better or twice a day like midnight and noon? Is updating part of the calculation or is there something to set up that I missed or have set wrong? Just curious, trying to be accurate, and perhaps correct a problem of some sort???

Latitude
41.118418
Longitude
-75.745239

Hey Jachym, boy I have to ask, don't you ever sleep? lol

Offline Jáchym

  • Meteotemplate Developer
  • Forecaster
  • *****
  • Posts: 8605
    • Meteotemplate
Re: solstice
« Reply #13 on: December 21, 2016, 10:25:11 PM »
If you can make a script that will calculate it this accurately please let me know, Im more than happy to integrate it into Meteotemplate, replacing the current one :)

Offline hankster

  • Forecaster
  • *****
  • Posts: 482
    • NFM Weather
Re: solstice
« Reply #14 on: December 21, 2016, 10:33:09 PM »
Nevermind :)
« Last Edit: December 21, 2016, 10:40:58 PM by hankster »

Offline gwwilk

  • Southeast Lincoln Weather
  • Forecaster
  • *****
  • Posts: 2578
    • SouthEast Lincoln, NE Weather
Re: solstice
« Reply #15 on: December 22, 2016, 07:51:49 AM »
If you can make a script that will calculate it this accurately please let me know, Im more than happy to integrate it into Meteotemplate, replacing the current one :)
'Accuracy' can be difficult to achieve when moving from point to point, but when you're talking about a fixed point on Earth, e.g my home, it's a different story.  In this case the only thing that's changing is the position of the sun relative to my fixed position.  That being the case the difference CAN be measured in seconds.  That's the reason for displaying seconds in my '$changeinday2' calculations.  After all, the primary user of my web site is me.  Others who live nearby can also use my day-different sunlight times, but those who are located elsewhere should view them as only approximations for their location because as you point out the times will most likely be different.

It's the relative changes at my location that are of interest to me, and that's why the seconds are significant.  That's also why I perform my own calculations rather than relying on another external calculation that hasn't been as reliable near the solstices at my location.  So long as I'm using the same calculation the day-different sunlight times will prove useful TO ME as the year progresses.  THIS is why 'yaimacaveman' is perplexed.
Regards, Jerry Wilkins
gwwilk@gmail.com

Offline hankster

  • Forecaster
  • *****
  • Posts: 482
    • NFM Weather
Re: solstice
« Reply #16 on: December 22, 2016, 08:00:28 AM »
Since you wrote your own script, at what time does it make the calculations for the next day?

Offline gwwilk

  • Southeast Lincoln Weather
  • Forecaster
  • *****
  • Posts: 2578
    • SouthEast Lincoln, NE Weather
Re: solstice
« Reply #17 on: December 22, 2016, 08:23:02 AM »
Since you wrote your own script, at what time does it make the calculations for the next day?
I think you mean 'yesterday' and 'today', although 'tomorrow' is optionally available in the script.  The sunlight times are calculated whenever '$changeinday2' is initially referenced by the calling script.  And as usual I didn't write the basic sunlight calculation but 'borrowed' it from elsewhere and then modified it to suit my purposes.  When scripting i ALWAYS look for solutions others have crafted in order to avoid 'reinventing the wheel'.  Again, here is the code to place it in the Saratoga template's menubar.  The entire thread should be reviewed before attempting to install it.  It's possible to modify the current sunlight pie chart in Scott Webster's 'ajax-dashboard6' to use '$changeinday2', but primarily because of the nested tables it's not a trivial exercise.
Regards, Jerry Wilkins
gwwilk@gmail.com

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #18 on: December 22, 2016, 08:50:29 AM »
Thanks hankster I think you get what I'm asking, lol.

Thanks too Jerry -- I think you answered my question? I apologize, I don't have a clue of what you are talking about, but I am getting the feeling that it is to complicated to have the script update everyday at midnight instead of 7pm every day.

I do understand to complicated, and if that is the case, no problem, I just have to wait until 7pm for the update.

Thanks for taking the time to explain!!!!!!

Offline waiukuweather

  • Forecaster
  • *****
  • Posts: 1072
Re: solstice
« Reply #19 on: December 22, 2016, 12:10:17 PM »
its 7pm for you because the times are UTC based

Offline yamiacaveman

  • Forecaster
  • *****
  • Posts: 439
    • Penn Lake Weather
Re: solstice
« Reply #20 on: December 22, 2016, 01:05:16 PM »
Thanks, Thanks, and Double Thanks waiukuweather!!!!!!!!!!!!!!!!!!!!

That I understand!

The 7pm makes sense now, and also yes, it possibly could be written to update at my desired time, but midnight UTC makes perfect sense for an upload or fetch time or whatever you would call it.

Thanks to everyone for their comments and time!!!!