Moon Age ???

[beteljuice]

Whilst looking at some sites I noticed that Moon Age was being reported as 30 days, and no moon image (30) available.
The next appearing data says moonage 1 day.

Surely it is day 0 in the cycle, and the images are numbered 0 > 29

Is this in 'the release', or am I looking at some customised sites ?

[mcrossley]

It's a rounding 'thing' anything over and including 29.5 will show as 30 of you use an integer value. The average lunar cycle is 29.53 days, and the calculation used is an approximation, so using integer values you will see day 30 every so often. You could use a decimal place or two and trunc the value for the image?

[mcrossley]

Ah not what you were saying, 0-29 vs. 1-30. Not sure, the code hasn't changed, I'll take a look at it when I get home next week.

[beteljuice]

.... and also .. as I posted:

I didn't see age = 0. (less than 1 day old)
Went from 30 > 1

[mcrossley]

If it does zero, it will last for less than half a day of course.

[mcrossley]

OK, the current code is very approximate, but I just ran CMX in a VM to see what happened this month.

New Moon was on the 24th at 21:42

The CMX MoonAge rolled over on the 25th at 04:30

It went from 29.52 to zero (using 2 dp), and 30 to 0 (using 0 dp) at that time.

So the code is working as expected, but it isn't very accurate.

EDIT: The actual interval between New Moons is subject to perturbation by the Sun (and other minor bodies), and actually varies between approximately 29.27 and 29.83 days. Using the mean value of 29.53 means that the approximate calculation is likely to have the age rolling over too early or too late compared with the actual phase.

EDIT2: I could add an tag option to truncate the value? Then it would always be in the range 0-29. At the cost of even less "accuracy" of course.
e.g. <#MoonAge tc=y>

[beteljuice]

Excuse my persistence, but I have no kit nor software ...

The figures I see on site(s) are integers.

It went from 29.52 to zero (using 2 dp), and 30 to 0 (using 0 dp) at that time.

From those figures 'day' 30 should have only lasted 0.12 of a day, but it was much longer, and of course 30 days 'old' does not exist ! - presumably why the missing graphic (0 > 29)
I still do not think I saw 0.

The presentation I am observing is a 'dashboard' style, Is that a standard distro or someone's '"value add" ?

Thank you for your patience

Late Edit: Shouldn't the Moon age be rounded down ?

[mcrossley]

Hi beteljuice - I think I may have edited my message above whilst you were replying (bad boy!)

The standard files do not make use of the MoonAge tag.

When I used to have the Cumulus template I believe I duplicated the 29 day file to 30 to cope. As you say it should only happen for 0.12 of a day, plus whatever the logging and FTP intervals are.

The MoonAge is very approximate - and I have people moaning that the phase angle is a few minutes out from the phase name!

[beteljuice]

EDIT2: I could add an tag option to truncate the value? Then it would always be in the range 0-29. At the cost of even less "accuracy" of course.
e.g. <#MoonAge tc=y>

The beteljuice had a senior moment in his edit above
What I really should have said is:
"Shouldn't the Moon age be ....truncate the value"

[mcrossley]

Done, 'twill be in the next release.