Page 3 of 3

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 2:07 pm
by mcrossley
sfws wrote: Tue 21 Jul 2020 3:05 pm
mcrossley wrote: Tue 21 Jul 2020 11:28 am add more options to the Cumulus.ini file
Actually this would be using options previously in the file; you have forgotten that in Cumulus 1 settings both determined which graphs were uploaded as images for its trends.htm, and also what was plotted on the charts (like temperature) that had multiple lines. If you look at https://cumuluswiki.org/a/Cumulus.ini#Section:_Graphs you will see that C1 let you decide not to plot the Heat index (HIVisible=0 ) that I have been suggesting is not useful.
Good idea, I'll implement the "xxxVisible" (and add some missing ones).
For now they will be probably be ini file read-only settings - no user interface configuration.

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 2:15 pm
by HansR
I assume that, when false, it means the data will not be included in the JSON file?
But when true, the order of the data in the file is always the same (as it is now)?

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 3:04 pm
by mcrossley
HansR wrote: Wed 22 Jul 2020 2:15 pm I assume that, when false, it means the data will not be included in the JSON file?
But when true, the order of the data in the file is always the same (as it is now)?
Correct, setting to false (=0) would exclude the the data from the JSON.

The JSON order will be changing as well, just because it makes the chart code much simpler. But item order in JSON should not normally matter, that is how JSON works - the normal rule for APIs is not to assume anything about JSON data order, use the names.

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 3:38 pm
by HansR
Well, I think it does matter to me. I'll look into it. Check the current graph procedure!
At least it has to be predictable ;)

Could you specify how this will work?
If I see an issue I'll start another thread.

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 3:57 pm
by sfws
mcrossley wrote: Wed 22 Jul 2020 2:07 pm Good idea, I'll implement the "xxxVisible" (and add some missing ones).
For now they will be probably be ini file read-only settings - no user interface configuration.
Glad you saw my post, given the muddling discussion of my answer to your question by another contributor.

I can't speak for all users, but read-only implementation should be fine, these are settings that are only likely to be considered and set once.

Is the default going to be all enabled, or are you making a guess at which lines people will want/not want for setting defaults?

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 4:52 pm
by mcrossley
sfws wrote: Wed 22 Jul 2020 3:57 pm I can't speak for all users, but read-only implementation should be fine, these are settings that are only likely to be considered and set once.

Is the default going to be all enabled, or are you making a guess at which lines people will want/not want for setting defaults?
Leaving ini file only is fine by me too :)

The default will be enabled for all values.
HansR wrote: Wed 22 Jul 2020 3:38 pm Well, I think it does matter to me. I'll look into it. Check the current graph procedure!
At least it has to be predictable ;)

Could you specify how this will work?
If I see an issue I'll start another thread.
I took my own advice and made the chart code work with whatever order the JSON is supplied in, so I can leave the current order. And actually the code turned out a lot cleaner when I had to think about it!

Re: Humidex - Discuss!

Posted: Wed 22 Jul 2020 6:56 pm
by HansR
mcrossley wrote: Wed 22 Jul 2020 4:52 pm I took my own advice and made the chart code work with whatever order the JSON is supplied in, so I can leave the current order. And actually the code turned out a lot cleaner when I had to think about it!
OK, I think I get it and how to handle it in my case.
Does this mean that the inside yes/no will also be handled by CMX and thus will not be in the datafile?

Re: Humidex - Discuss!

Posted: Fri 31 Jul 2020 1:59 pm
by sfws
freddie wrote: Sat 18 Jul 2020 5:53 pm IThe UK Met Office doesn't "do" humidex, and that speaks volumes to me. It is of limited use in the UK climate, so I guess that is not entirely surprising.
Today is very muggy, at least here (it is fairly cloudy here this afternoon), but I think it is a perfect day to illustrate that the Canadian Humidity Index is useful in UK Climate.

The dew point (and absolute humidity - currently 13.21 g m-3, which is also today's highest) are a good indication of how uncomfortable it feels and I'm glad I am reporting them.

But I believe Humidex= 35.3 is giving me a much better reflection of the Muggy nature of today than the 31.0 oC from Feels Like or the 31.5 oC from outside temperature.
Those were values showing as I type (I am not promising my Fine Offset is giving perfect Celsius air temperature)
The maximum so far today for Humidex is 36.4, of Feels like is 33.3 oC and of outside temperature 33.2 oC, all of these peaked in last hour.

To requote Freddie: That current and peak Humidex speaks volumes to me!

Re: Humidex - Discuss!

Posted: Fri 31 Jul 2020 10:46 pm
by BCJKiwi
An interesting set of graphs clearly showing the different "temperatures".
Temperature.PNG
Wind.PNG
Pressure.PNG
Humidity.PNG

Re: Humidex - Discuss!

Posted: Sat 01 Aug 2020 7:30 am
by sfws
BCJKiwi wrote: Fri 31 Jul 2020 10:46 pm An interesting set of graphs clearly showing the different "temperatures".
Yes!
Makes my numbers look dull, but my charts don't show such clear separation, too much crossing over. I will only include the main one:

Re: Humidex - Discuss!

Posted: Sat 01 Aug 2020 9:00 pm
by BCJKiwi
I only posted that graph set as it was unusual, they normally have all sorts of overlaid/crossover traces.
We are currently in a quite stable weather pattern.