# Astro sunset sunrise: 1 minute later

Hi.

Using the astro binding 1.x, the values displayed today are 7h31 and 20h21.
When I check on the net, for example here http://www.timeanddate.com/sun/france/paris , it looks like the values should be 7h30 and 20h20. The calculated values are always 1 minute later.
Ok, that is only a minor problem and of course I can live with it, but is there a bug in the calculation (rounding problem) ?

Astro takes into account your actual lat/lon location. So if you are a little east or west of the center of Paris your times will be slightly different from those at that link.

You have an idea how kilometers correspond to 1 minute ? I would not have expected such a difference.

I could put Paris lat/lon too to check if you’re right.

Now that I think about it where you are north or south also impacts sunrise and sunset since the day length changes depending on your latitude as well. In short it isn’t a simple calculation that I don’t know how to do. I think your best bet is to do what you suggest and put the Paris lat/lon in and see if it is a minute different.

Alternatively you can see what time Astro is projecting and look to see if the sun actually sets at the Astro predicted time or that website’s predicted time.

I just tried with Paris latitude/longitude and I got the same values, that is with a shift of 1 minute compared to what WEB sites report.

By the way, the values are not always the same on different WEB sites.
So I have no idea what are the correct values.

I’m reminded of a saying. “If you have one thermometer you know the temperature. If you have two thermometers you have confusion.” One minute isn’t all that far off for practical purposes and after reading some articles on calculating sunset I’ve learned it is actually a pretty challenging problem influenced by not only lat/lon but also altitude and time of year. I wouldn’t stress over it.

AFAIK, the calculations used in the astro binding (for the sun at least) are in some way based on information from the following web site:

SunCalc

It would be interesting to see if the values reported by this web site are the same as the values reported by the astro binding.

If they are, I think we can conclude - as indicated by @rlkoshak - that the difference that you see is simply caused by sligthly different algottihms used for calculations (or precision of the input and/or output).

Interesting, from this site I got 7h29 and 20h22 while the binding calculated 7h29 and 20h23. So sunrise is correct but sunset is wrong from 1 minute. As seconds are not displayed, maybe it is just a different rounding to the minute ?

PS: I can see from this site that the center of Paris results in 7h28 and 20h22. That makes sense as I am a little more at West.