Cumulus Support

Today we have had our first air frost of the Winter but the Cumulus 1.9.2 did not see it for some reason.
At 8am the Davis Pro 2 recorded -0.2C and this was also displayed in weatherlink as the low. But the Cumulus only went down to +0.1C at 7.50AM. I have closed and opened the program again but it will not show this low for some reason.

Any thoughts on this one!

Warren

Also while I've posted. Some people are displaying a box with their post showing the current weather at their sites. How do you get these data boxes?

Warren

wradmore wrote:Today we have had our first air frost of the Winter but the Cumulus 1.9.2 did not see it for some reason.
At 8am the Davis Pro 2 recorded -0.2C and this was also displayed in weatherlink as the low. But the Cumulus only went down to +0.1C at 7.50AM. I have closed and opened the program again but it will not show this low for some reason.

Was Cumulus actually running at the time?

wradmore wrote:Also while I've posted. Some people are displaying a box with their post showing the current weather at their sites. How do you get these data boxes?

See https://cumulus.hosiene.co.uk/viewtopic.php?f=14&t=103

No, Cumulus was off at the time. I turned on PC at 8:10 then started Cumulus.
Thanks for the link.

Warren

wradmore wrote:No, Cumulus was off at the time. I turned on PC at 8:10 then started Cumulus.

Then the minimum reading was effectively 'missed'. Cumulus doesn't use the highs and lows from any station, it uses the raw data and maintains its own highs and lows. For two reasons; firstly it supports an 0900-0900 day, and in those cases the station's highs and lows are not valid. Secondly, in order to support multiple types of weather station, I have to keep as much code 'common' as possible. It would require me to write and maintain a large amount of extra code if I were to offer the option of using the station's highs and lows.

Thanks Steve

Copmpared December low Minimum with Cumulus and Weatherlink and this showed 3 days where there was 0.1 to 0.2C difference between the two.

I will use weatherlink to provide the accurate stats for the month from now on and use Cumulus solely for website live feed.

wradmore wrote:I will use weatherlink to provide the accurate stats for the month from now on and use Cumulus solely for website live feed.

Yes, if you can't run Cumulus all the time, there's always the chance that Weatherlink will be able to give you slightly 'better' highs and lows for some of the stats.

Cumulus has a much better layout than weatherlink. We thank you greatly for doing this Steve.
It's a shame that i have to refer back to weatherlink to get the accurate stats. Weatherlink does not provide straight forward averages for the month unlike your 'this month'.
It means I will have to go through and adjust the errors (data from weatherlink) in Cumulus and then get the averages. Still can't understand why Cumulus can't get the same accurate weather data as weatherlink? Perhaps it might in the future.

Anyway well done on the software

wradmore wrote:Still can't understand why Cumulus can't get the same accurate weather data as weatherlink? Perhaps it might in the future.

It would if you ran it 24 hours a day. But when it reads data from the logger, it only gets the temperature at the time the logger entry was made. If the temperature varied slightly during the period that the logger entry covered, such that a new minimum temperature occurred, Cumulus won't see that.

Weatherlink is specifically designed to work with only Davis stations, it has all of the code it requires to read all of the data from Davis stations and no other station. Cumulus is 'general purpose' software and doesn't have all of the code required to read all of the data from all of the stations that it works with. To stand any chance of making Cumulus work with a number of different stations, I have to make most of the code in Cumulus 'common' and then I try to fit the widely-varying readings available from the various weather stations into that framework. This includes determining my own highs and lows from the basic data readings, rather than reading the highs and lows from each station.

Thanks Steve