ben@hpcvlx.HP.COM (Benjamin Ellsworth) (12/30/88)
I need an algorithm for predicting the phases of the moon. Can anyone
help me find such a thing?
-----------------------------------------------------------------------
Benjamin Ellsworth | ben%hpcvlx@hplabs.hp.com | INTERNET
Hewlett-Packard Company | {backbone}!hplabs!hpcvlx!ben | UUCP
1000 N.E. Circle | (USA) (503) 750-4980 | FAX
Corvallis, OR 97330 | (USA) (503) 757-2000 | VOICE
-----------------------------------------------------------------------
All relevant disclaimers apply.
-----------------------------------------------------------------------barmar@think.COM (Barry Margolin) (12/31/88)
All you need to know is the time of a recent new moon and the period
of the moon phase cycle. The program we use here has the base new
moon time as Oct 21, 1987 23:47 GMT, and the period as 29 days, 12
hours, 44 minutes, 3 seconds (totaling 2,551,443 seconds). Then
compute (((current_time_in_seconds - base_new_moon_time) % period) /
(float) period).
You can get a more recent base new moon time from an astronomy
magazine, almanac, or ephemeris.
Barry Margolin
Thinking Machines Corp.
barmar@think.com
{uunet,harvard}!think!barmarkibo@pawl10.pawl.rpi.edu (James Parry) (01/01/89)
[]
Here are a couple of little ones. The first is a macro for the TeX
document processor, to teach it the phase of the moon (as well as the day of
the week). The second, much shorter, one is extracted from the game 'nuthack'
(aka nethack):
--
%
% \DayOfWeek expands to the day of the week ("Sunday", etc.)
% \PhaseOfMoon expands to the phase of the moon
%
% Written by Martin Minow of DEC (minow%bolt.dec@decwrl.dec.com).
%
\def\DayOfWeek{%
%
% Calculate day of the week, return "Sunday", etc.
%
\newcount\dow % Gets day of the week
\newcount\leap % Leap year fingaler
\newcount\x % Temp register
\newcount\y % Another temp register
% leap = year + (month - 14)/12;
\leap=\month \advance\leap by -14 \divide\leap by 12
\advance\leap by \year
% dow = (13 * (month + 10 - (month + 10)/13*12) - 1)/5
\dow=\month \advance\dow by 10
\y=\dow \divide\y by 13 \multiply\y by 12
\advance\dow by -\y \multiply\dow by 13 \advance\dow by -1 \divide\dow by 5
% dow += day + 77 + 5 * (leap % 100)/4
\advance\dow by \day \advance\dow by 77
\x=\leap \y=\x \divide\y by 100 \multiply\y by 100 \advance\x by -\y
\multiply\x by 5 \divide\x by 4 \advance\dow by \x
% dow += leap / 400
\x=\leap \divide\x by 400 \advance\dow by \x
% dow -= leap / 100 * 2;
% dow = (dow % 7)
\x=\leap \divide\x by 100 \multiply\x by 2 \advance\dow by -\x
\x=\dow \divide\x by 7 \multiply\x by 7 \advance\dow by -\x
\ifcase\dow Sunday\or Monday\or Tuesday\or Wednesday\or
Thursday\or Friday\or Saturday\fi
}
%%
%%
%%
\def\PhaseOfMoon{% Calculate the phase of the (civil) moon.
%
% The routine calculates the year's epact (the age of the moon on Jan 1.),
% adds this to the number of days in the year, and calculates the phase
% of the moon for this date. It returns the phase as a string, e.g.,
% "new", "full", etc.
%
% In the algorithm:
%
% diy Is the day of the year - 1 (i.e., Jan 1 is day 0).
%
% golden Is the number of the year in the Mentonic cycle, used to
% determine the position of the calender moon.
%
% epact Is the age of the calender moon (in days) at the beginning
% of the year. To calculate epact, two century-based
% corrections are applied:
% Gregorian: (3 * cent)/4 - 12
% is the number of years such as 1700, 1800 when
% leap year was not held.
% Clavian: (((8 * cent) + 5) / 25) - 5
% is a correction to the Mentonic cycle of about
% 8 days every 2500 years. Note that this will
% overflow 16 bits in the year 409600. Beware.
%
% The algorithm is accurate for the Gregorian calender only.
%
% The magic numbers used in the phase calculation are as follows:
% 29.5 The moon's period in days.
% 177 29.5 scaled by 6
% 22 (29.5 / 8) scaled by 6 (this gets the phase)
% 11 ((29.5 / 8) / 2) scaled by 6
%
% Theoretically, this should yield a number in the range 0 .. 7. However,
% two days per year, things don't work out too well.
%
% Epact is calculated by the algorithm given in Knuth vol. 1 (calculation
% of Easter). See also the article on Calenders in the Encyclopaedia
% Britannica and Knuth's algorithm in CACM April 1962, page 209.
%
\newcount\cent % Century number (1979 == 20)
\newcount\epact % Age of the moon on Jan. 1
\newcount\diy % Day in the year
\newcount\golden % Moon's golden number
\newcount\x % Temp
\newcount\m % Temp for modulus
\diy=\day \advance\diy by \ifcase\month % Jan 1 == 0
-1\or -1\or 30\or 58\or 89\or 119\or 150\or % Jan .. Jun
180\or 211\or 241\or 272\or 303\or 333\fi % Jul .. Dec
% if ((month > 2) && ((year % 4 == 0) &&
% ((year % 400 == 0) || (year % 100 != 0))))
% diy++; /* Leapyear fixup */
\ifnum \month>2
\x=\year \m=\x \divide\m by 4 \multiply\m by 4 \advance\x by -\m
\ifnum \x=0 % month > 2 and maybe leapyear
\x=\year \m=\x \divide\m by 400 \multiply\m by 400 \advance\x by -\m
\ifnum \x=0 % 2000 is a leap year
\advance\diy by 1 % so it's one day later
\else % not 2000, check other '00's
\x=\year \m=\x \divide\m by 100 \multiply\m by 100 \advance\x by -\m
\ifnum \x>0 % not some other '00' year
\advance\diy by 1 % it's still one day later
\fi % not odd century
\fi % not 2000-type century
\fi % not leapish year
\fi % not march or later
% cent = (year / 100) + 1; /* Century number */
% golden = (year % 19) + 1; /* Golden number */
\cent=\year \divide\cent by 100 \advance\cent by 1
\golden=\year
\m=\year \divide\m by 19 \multiply\m by 19 \advance\golden by -\m
\advance\golden by 1
% epact = ((11 * golden) + 20 /* Golden number */
% + (((8 * cent) + 5) / 25) - 5 /* 400 year cycle */
% - (((3 * cent) / 4) - 12)) % 30;/* Leap year correction */
\epact=11 \multiply\epact by \golden
\advance\epact by 20
\x=8 \multiply\x by \cent \advance\x by 5
\divide\x by 25 \advance\x by -5
\advance\epact by \x
\x=3 \multiply\x by \cent \divide\x by 4 \advance\x by -12
\advance\epact by -\x
\m=\epact \divide\m by 30 \multiply\m by 30 \advance\epact by -\m
% if (epact <= 0)
% epact += 30; /* Age range is 1 .. 30 */
% if ((epact == 25 && golden > 11) || epact == 24)
% epact++;
\ifnum \epact<0
\advance\epact by 30
\fi
\ifnum \epact=25
\ifnum \golden>11
\advance \epact by 1
\fi
\else
\ifnum \epact=24
\advance \epact by 1
\fi
\fi
%
% Calculate the phase, using the magic numbers defined above.
% Note that phase may be equal to 8 (== 0) on two days of the year
% due to the way the algorithm was implemented.
% phase = (((((diy + epact) * 6) + 11) % 177) / 22) & 7;
%
\x=\diy \advance\x by \epact \multiply\x by 6 \advance\x by 11
\m=\x \divide\m by 177 \multiply\m by 177 \advance\x by -\m
\divide\x by 22
\ifcase\x new\or waxing crescent\or in its first quarter\or
waxing gibbous\or full\or waning gibbous\or
in its last quarter\or waning crescent\or new\fi
}
--
phase_of_the_moon() /* 0-7, with 0: new, 4: full */
{ /* moon period: 29.5306 days */
/* year: 365.2422 days */
register struct tm *lt = getlt();
register int epact, diy, golden;
diy = lt->tm_yday;
golden = (lt->tm_year % 19) + 1;
epact = (11 * golden + 18) % 30;
if ((epact == 25 && golden > 11) || epact == 24)
epact++;
return( (((((diy + epact) * 6) + 11) % 177) / 22) & 7 );
}
--
Have fun. (You'll probably get a dozen more copies of each of these from
other people...) Hopefully, these two produce results that agree with
each other. (They seem to use similar algorithms.)
James "Kibo" Parry
kibo%pawl.rpi.edu@itsgw.rpi.edu
userfe0n@rpitsmts (bitnet)ray3rd@ssc-vax.UUCP (Ray E Saddler III) (01/04/89)
In article <101150002@hpcvlx.HP.COM>, ben@hpcvlx.HP.COM (Benjamin Ellsworth) writes: > > I need an algorithm for predicting the phases of the moon. Can anyone > help me find such a thing? A while back in 1988 I obtained a real nice tool for the Sun, called suntool. It's very exact in representing the phase of the moon and greatly attemts to even provide up-to-the-second distance from the earth. It was written by John Walker, Autodesk, Inc. I have version 2.0. Anybody out there know how to contact the author? -- | Ray E. Saddler III | __ __ __ __ | Path: ..!ssc-vax!ray3rd | | Boeing Aerospace | / / / // //| // | From: ray3rd@ssc-vax.UUCP | | P.O. Box 3999 m.s. 3R-05 | /-< / //- // |// _ |---------------------------| | Seattle, Wa. 98124 USA | /__//_//__ // //__/ | VoiceNet: (206) 657-2824 |
ben@hpcvlx.HP.COM (Benjamin Ellsworth) (01/04/89)
Re: Moon Phase Prediction
After having received a transfinite number of responses to my request,
I now request that further answers be delayed until I can make an
adequate survey of what I have received. If I need more, I will repeat
my request.
Re: Moon Rise/Set Prediction
Not many responses to this problem. If you have an algorithm (esp.
coded in C) that you would be willing to share, please send it to me.
-----------------------------------------------------------------------
Benjamin Ellsworth | ben%hpcvlx@hplabs.hp.com | INTERNET
Hewlett-Packard Company | {backbone}!hplabs!hpcvlx!ben | UUCP
1000 N.E. Circle | (USA) (503) 750-4980 | FAX
Corvallis, OR 97330 | (USA) (503) 757-2000 | VOICE
-----------------------------------------------------------------------
All relevant disclaimers apply.
-----------------------------------------------------------------------ray3rd@ssc-vax.UUCP (Ray E Saddler III) (01/05/89)
In article <2454@ssc-vax.UUCP>, ray3rd@ssc-vax.UUCP (Ray E Saddler III) writes: > In article <101150002@hpcvlx.HP.COM>, ben@hpcvlx.HP.COM (Benjamin Ellsworth) writes: > > > > I need an algorithm for predicting the phases of the moon. Can anyone > > help me find such a thing? > > A while back in 1988 I obtained a real nice tool for the Sun, called > suntool. It's very exact in representing the phase of the moon and ======= SHOULD have been the word moontool. God, how embarassing! -- | Ray E. Saddler III | __ __ __ __ | Path: ..!ssc-vax!ray3rd | | Boeing Aerospace | / / / // //| // | From: ray3rd@ssc-vax.UUCP | | P.O. Box 3999 m.s. 3R-05 | /-< / //- // |// _ |---------------------------| | Seattle, Wa. 98124 USA | /__//_//__ // //__/ | VoiceNet: (206) 657-2824 |
cramer@optilink.UUCP (Clayton Cramer) (01/07/89)
In article <2455@ssc-vax.UUCP., ray3rd@ssc-vax.UUCP (Ray E Saddler III) writes: . In article <2454@ssc-vax.UUCP., ray3rd@ssc-vax.UUCP (Ray E Saddler III) writes: . . In article <101150002@hpcvlx.HP.COM., ben@hpcvlx.HP.COM (Benjamin Ellsworth) writes: . . . . . . I need an algorithm for predicting the phases of the moon. Can anyone . . . help me find such a thing? . . . . A while back in 1988 I obtained a real nice tool for the Sun, called . . suntool. It's very exact in representing the phase of the moon and . ======= . SHOULD have been the word moontool. God, how embarassing! . -- . | Ray E. Saddler III | __ __ __ __ | Path: ..!ssc-vax!ray3rd | That's what we need! A tool for predicting the phase of the Sun. I volunteer to write the math part of the code. :-) (for those of you with no sense of humor, or haven't had your morning stimulants yet). -- Clayton E. Cramer {pyramid,pixar,tekbspa}!optilink!cramer Disclaimer? You must be kidding! No company would hold opinions like mine!
gomez@cisunx.UUCP ( roberto o Gomez) (01/08/89)
The sources for the ``moontool'' program mentioned should be available for anonymous ftp (login as anonymous, any password) in titan.rice.edu, in the directory sun-source. Usenet readers may want to get it from the sun-spots archive server, both maintained by William Lefebvre at Rice. As he says: >For more information about the archive server, send a mail message >containing the word "help" to the address "archive-server@rice.edu". Hope this helps Roberto Gomez Physics Dept., Univ. of Pittsburgh, Pittsburgh PA 15260 Arpa: roberto@bondi.phyast.pittsburgh.edu (128.147.33.13) gomez@unix.cis.pittsburgh.edu UUCP: ...!psuvax1!pitt!cisunx!gomez