Gary - I totally agree with your last sentence - my comments were aimed towards giving Josh some possible options which are available to him for customising his NOAA templates if the 'issue' concerns him.
On Sunday, 18 December 2016 23:30:41 UTC+2, gjr80 wrote: > The fact remains that the driver emits rain as 0.518mm or 0.020393701 > inches per tip. Since it is a tipping rain gauge the database should only > record multiples of this figure. Your initial email mentions 2 numbers; > 0.040551181102 etc and 0.32519685039 etc. The first is clearly meant to be > 2 tips and the second 16 tips. But using the tip value in the driver these > numbers should be 0.040787402 and 0.326299216, about 0.6% and 0.1% under > respectively. Not a huge amount but enough to cause the types of errors you > have seen, particularly when the 3rd digit of the rain figure concerned is > around about 5. Dealing with floats in python adds some inherent inaccuracy > to the resulting values but this is typically only comes into play > somewhere around the 10th or 11th decimal place, way to small to have an > effect here. Sure you can make things look ok by changing the formatting or > the calculations if you dare, but these are only cosmetic fixes that will > mask the issue in this case but it will recur in other cases (what happens > if you use 3 decimal places and then you get a 5 in the fourth decimal > place?). > > I don't believe there is any need to change the underlying weewx > formatting/conversion machinery whilst the above anomalies remain. > > Gary -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
