Welcome to the Cumulus Support forum.

Latest Cumulus MX V3 release 3.28.6 (build 3283) - 21 March 2024

Cumulus MX V4 beta test release 4.0.0 (build 4018) - 28 March 2024

Legacy Cumulus 1 release v1.9.4 (build 1099) - 28 November 2014 (a patch is available for 1.9.4 build 1099 that extends the date range of drop-down menus to 2030)

Download the Software (Cumulus MX / Cumulus 1 and other related items) from the Wiki

Observations of Beta v1.9.4 build 1076

Please discuss beta versions of Cumulus 1 here. Note: There are currently no beta versions of Cumulus 1.
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Observations of Beta v1.9.4 build 1076

Post by Touchtone »

Today I installed the latest Beta version build 1076.
I now notice that Cumulus now gets data from the Davis data logger every second instead of every 2 to 3 seconds as it did on v1.9.3 build 1059.
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by steve »

It depends on the firmware version in your console, whether or not Cumulus has been able to read the firmware version, and whether or not you have set a VP2SleepInterval value in cumulus.ini. Some of the recent versions of the firmware return the data immediately, rather than after a delay. Cumulus tries to compensate for this by adding a delay. If the auto-detection isn't working for your console, then you'll need to manually add VP2SleepInterval=1500 to the [Station] section of cumulus.ini. You may need to adjust the value for your system.
Steve
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Re: Observations of Beta v1.9.4 build 1076

Post by Touchtone »

steve wrote:It depends on the firmware version in your console, whether or not Cumulus has been able to read the firmware version, and whether or not you have set a VP2SleepInterval value in cumulus.ini. Some of the recent versions of the firmware return the data immediately, rather than after a delay. Cumulus tries to compensate for this by adding a delay. If the auto-detection isn't working for your console, then you'll need to manually add VP2SleepInterval=1500 to the [Station] section of cumulus.ini. You may need to adjust the value for your system.
I have now set the sleep interval to 1500 as you suggest & it now updates every 2 seconds approx. I am using firmware v3.12 in my VP2 console as I had trouble with v3.00.
I don't run my computer all the time & I load Cumulus for an hour or so each day & also alternate between that & WeatherLink.

Just as a matter of interest, how much work would be needed to get Cumulus to read the Highs & Lows from the console as WeatherLink does. I realise you have to cater for so many different weather stations & this probably means a huge amount of rewriting code for just one station type. I am not requesting this but it would be a nice advantage & save having to edit the day file in Cumulus each day.
As we are going into the winter here the minimum temps are dropping so each day I often need to edit them.
I have the the log update set at 15 minutes.
n9mfk
Posts: 845
Joined: Sun 10 May 2009 8:52 pm
Weather Station: davis vp2 Serial datalogger
Operating System: Windows 7 64-bit
Location: Springfield, IL

Re: Observations of Beta v1.9.4 build 1076

Post by n9mfk »

Hi Touchtone,
What did you look at to see how fast you were get the data from the Davis vp logger ?
thanks Beau
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by steve »

Touchtone wrote:I have now set the sleep interval to 1500 as you suggest & it now updates every 2 seconds approx. I am using firmware v3.12 in my VP2 console as I had trouble with v3.00.
I don't run my computer all the time & I load Cumulus for an hour or so each day & also alternate between that & WeatherLink.
This is odd; I installed fw 3.12, and mine reads at the 'old' interval without needing any delay in Cumulus. That's why I changed the code so that it only added the delay automatically for fw 3.00.
Just as a matter of interest, how much work would be needed to get Cumulus to read the Highs & Lows from the console as WeatherLink does. I realise you have to cater for so many different weather stations & this probably means a huge amount of rewriting code for just one station type. I am not requesting this but it would be a nice advantage & save having to edit the day file in Cumulus each day.
It would be a lot of work, it's just not the way Cumulus works. It only reads the current temperature from the logger data, it can't handle more than one reading per entry, so it can't use the max and min entries for each entry. It was never really intended to be used the way you use it, just running it for an hour a day and relying on logger data for everything. What logger interval are you using (in the station)? As you rely on the logger data almost entirely, I would expect that you have it set to one minute, in which case the chances of missing a low temperature were very low, I'd have thought.
Steve
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by steve »

n9mfk wrote:Hi Touchtone,
What did you look at to see how fast you were get the data from the Davis vp logger ?
thanks Beau
For an approximate idea, you can look at the update time at the bottom right of the Cumulus man window. For a more accurate figure, if you are using the latest beta version, you can turn on the debug log, and that will show the time each time it loads the data.
Steve
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Re: Observations of Beta v1.9.4 build 1076

Post by Touchtone »

Cumulus has detected the firmware version as 3.12. However the sleep interval had to be edited. I assume as I have copied the Beta v 1.9.4 build 1076 update into the same Cumulus folder as v 1.9.3 build 1059 was in, it just updated the various files & did not put in the sleep interval if it was not already there. I really do not know if it was there or not as I had not checked. However the older version did update every 2 or 3 seconds & when I installed the Beta version it updated every second.
As I have now added the sleep interval to 1500 it updates as before.
I can't say it caused any problems as I did not notice any & the wind speed seemed to adjust very rapidly with the faster update speed.
What problems are likely to occur with the faster update speed?.
BCJKiwi
Posts: 1255
Joined: Mon 09 Jul 2012 8:40 pm
Weather Station: Davis VP2 Cabled
Operating System: Windows 10 Pro
Location: Auckland, New Zealand
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by BCJKiwi »

Mostly High CPU Usage.

@Steve FYI
Have updated my Vue FW from 2.14 to 3.00 and Cumulus to 1076. The sleep interval of 2250 that I had before is still unchanged and the update frequency continues at around 3 secs. CPU utilisation is also similar to what it was before.
I had tested VVP and WL (with both Cumulus and WL talking to VVP by IP) but found in the second week there were a couple of lockups so have gone back to Cumulus alone.
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by steve »

Touchtone wrote:Cumulus has detected the firmware version as 3.12. However the sleep interval had to be edited. I assume as I have copied the Beta v 1.9.4 build 1076 update into the same Cumulus folder as v 1.9.3 build 1059 was in, it just updated the various files & did not put in the sleep interval if it was not already there. I really do not know if it was there or not as I had not checked.
It doesn't write the value to cumulus.ini, it just applies it internally when it knows that it's required (i.e. when it detects fw 3.00) and one hasn't been supplied.
I can't say it caused any problems as I did not notice any & the wind speed seemed to adjust very rapidly with the faster update speed.
If it was actually updating the display with different values, that sounds like it wasn't actually doing excess reads. I noticed some time ago with mine that just after start up, the reads are much more frequent anyway; they seem to get buffered somewhere. The interval then settles down to the expected rate. How long are you running Cumulus after start up before checking the update interval? It does seem odd that everyone else with 3.12 doesn't need the delay setting.
Steve
freddie
Posts: 2434
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2 + Ecowitt
Operating System: GNU/Linux Ubuntu 22.04 LXC
Location: Alcaston, Shropshire, UK
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by freddie »

It does seem odd that everyone else with 3.12 doesn't need the delay setting.
I'm running 3.12 with 1.9.4 (build 1076). I see updates at an interval of less than two seconds - sometimes more than one per second. I have VP2SleepInterval set to 0. I don't know if that is relevant - I think I set it low because of missed wind gusts.
Freddie
Image
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by steve »

The updates won't necessarily be regular, I'd expect some slight variation, but overall, without a sleep interval configured (or zero) then the average update interval is usually around 3 seconds with fw 3.12. I saw this myself when investigating the 'missing high gust' issues, and saw the same with others who supplied logs. Before I changed the code so that the automatic delay was added for fw 3.00 and above, people with fw 3.12 were typically seeing intervals between reads of five seconds or more (i.e. a double delay).

Having said that, if some people with fw 3.12 are finding it updates excessively (even after its been running for a while) for some unexplained reason, then the sleep interval is available to be tweaked if required. The best way of finding the actual update interval is to turn on the debug log in the latest beta and look at the average times between the "Load current data" entries, making sure that Cumulus has been running for several minutes first.

The two potential problems with excessive reads are:

1. High CPU usage
2. The circular buffer holding the recent wind speeds in Cumulus can fill more quickly than expected, i.e. holding a shorter period that expected, so calculations of things like 10-minute wind speed, and recent 10-minute high gust, may be incorrect.
Steve
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Re: Observations of Beta v1.9.4 build 1076

Post by Touchtone »

steve wrote:).Having said that, if some people with fw 3.12 are finding it updates excessively (even after its been running for a while) for some unexplained reason, then the sleep interval is available to be tweaked if required. The best way of finding the actual update interval is to turn on the debug log in the latest beta and look at the average times between the "Load current data" entries, making sure that Cumulus has been running for several minutes first.

The two potential problems with excessive reads are:

1. High CPU usage
2. The circular buffer holding the recent wind speeds in Cumulus can fill more quickly than expected, i.e. holding a shorter period that expected, so calculations of things like 10-minute wind speed, and recent 10-minute high gust, may be incorrect.
I have checked the debug log file with notepad++ while Cumulus is running & all I get is "0000.000 : **** Logging Disabled" repeating itself every second line. What am I doing wrong?.
I have removed the VP2 sleep interval form Cumulus ini & the update interval occurs every one second approx & is still on one second after running for more than 15 minutes.
I have checked CPU usage with Cumulus running & it shows 2.8-3.8% for Cumulus using the current one second update interval. Is this figure in the ball park for normal Cumulus operation.
Thanks & all the best.
Bligh.
BCJKiwi
Posts: 1255
Joined: Mon 09 Jul 2012 8:40 pm
Weather Station: Davis VP2 Cabled
Operating System: Windows 10 Pro
Location: Auckland, New Zealand
Contact:

Re: Observations of Beta v1.9.4 build 1076

Post by BCJKiwi »

Debug logging has to be turned on for say 5 ~ 10 minutes.

Configuration / Debug Log (click the item and it will get a tick).
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Re: Observations of Beta v1.9.4 build 1076

Post by Touchtone »

Thanks BCJKiwi. I have checked the debug log file as you suggested & it is certainly updating at once per second & in some cases it seems even more frequent than that, so I have reset the VP2 Sleep Interval to 1500.
It does seem that some data loggers return information very quickly. As far as I know mine is the latest current one as I bought the station in March this year from the Australian distributor & it was new stock fresh in from the USA. The data logger & WL software are version 6.0.2 which I assume is the latest.
Touchtone
Posts: 93
Joined: Tue 09 Apr 2013 11:36 am
Weather Station: Davis VP2 Plus & FARS
Operating System: Windows 7 Home Premium x64 SP1
Location: Gin Gin, Qld, Australia

Re: Observations of Beta v1.9.4 build 1076

Post by Touchtone »

Just an extra observation that may be of interest.
Since resetting the sleep interval to 1500, Cumulus use of CPU resources has dropped dropped from 3.8% to 2.5%.
Locked