New User Observation

[George Richardson]

Hello Steve.

Wonderful program. Thank you. Been using only about 3 weeks. Started with 1.8.4 Build 605. Updated yesterday to 1.8.5 Build 690.

At the present, I am using/learning this program on a non-24/7 computer. I would prefer having my recorded day start & stop at midnight. When I first set it up that way I had trouble with "yesterdays extremes" not populating on startup so I changed to 9 to 9 and everything was OK. After upgrading to 1.8.5, I changed my cycle to midnight to midnight. Upon starting this morning, "yesterdays extremes" populated properly however under "Todays extremes" the Min Temp is shown as 0.0 at 00:00. This comes from today.ini, Low=0, Ltime=00:00 and is incorrect.

A request that I haven't seen addressed; could you offer a 12 hour clock option?

A challenge to your smart Davis users: Could someone devise a translation program that would convert WeatherLink historical data to Cumulus format. I really like Cumulus and would like to populate it with all of my history.

Steve, at the risk of stirring the muck, I personally used the "Setting correct annual rainfall so far" thread because I to was having the same problem and hadn't found the solution (was getting ready to ask!). You justifiably mused, "I'd be grateful for any suggestions from anyone who is still here as to what I might call the section of the help file which holds the format of the data log files if not "Data Log File Format"." Just as a suggestion for consideration from a user that has a computer programmer IQ of well under 100, the folder title "Review or Correct Data" as a Help File section would be more descriptive to me.

FWIW, downpayment made.

Thanks for your consideration.

George

This is on a XP Pro machine working off TCP/IP VirtualVP Beta for remote access to Davis VP2 on comm port 8.

[steve]

At the present, I am using/learning this program on a non-24/7 computer. I would prefer having my recorded day start & stop at midnight. When I first set it up that way I had trouble with "yesterdays extremes" not populating on startup so I changed to 9 to 9 and everything was OK.

I don't understand why that should happen - this continued for a number of days? Most people use the 12-12 option, so you would expect that problem to be seen by more people, given that changing to 9-9 apparently fixed it.

After upgrading to 1.8.5, I changed my cycle to midnight to midnight. Upon starting this morning, "yesterdays extremes" populated properly however under "Todays extremes" the Min Temp is shown as 0.0 at 00:00. This comes from today.ini, Low=0, Ltime=00:00 and is incorrect.

Are all the other 'today' values correct? Could you send me the file from the Diags folder for this morning's start up?

A request that I haven't seen addressed; could you offer a 12 hour clock option?

I could, yes. I didn't think to do it as it's a bit unusual for weather recording. I'd have to go through everywhere that I generate or use a time, and put in a change to use a 12-hour format - because they don't all do it the same way. Just a bit tedious and error-prone. And, of course, there would have to be an am/pm indicator, so the times wouldn't all be 5 characters, which complicates things a bit. I'll bear it in mind for Cumulus 2. Remind me when I forget

Steve, at the risk of stirring the muck, I personally used the "Setting correct annual rainfall so far" thread because I to was having the same problem and hadn't found the solution (was getting ready to ask!). You justifiably mused, "I'd be grateful for any suggestions from anyone who is still here as to what I might call the section of the help file which holds the format of the data log files if not "Data Log File Format"." Just as a suggestion for consideration from a user that has a computer programmer IQ of well under 100, the folder title "Review or Correct Data" as a Help File section would be more descriptive to me.

Fair enough, but that isn't specifically what that section is about, so anyone looking for the format of the data log files for some other reason might not think to look there. What I should really provide is a proper guide to correcting the data when things go wrong for whatever reason. But like everything, it's finding the time.

Steve

[George Richardson]

Steve,

"I don't understand why that should happen - this continued for a number of days? Most people use the 12-12 option, so you would expect that problem to be seen by more people, given that changing to 9-9 apparently fixed it." *** Didn't continue for more than a couple of days because I changed to 9 to 9. Something I read caused me to think this was a problem whose fix would defer to Cumulus 2.

Are all the other 'today' values correct? YES Could you send me the file from the Diags folder for this morning's start up? *** Will try to attach.

I could, yes. I didn't think to do it as it's a bit unusual for weather recording. I'd have to go through everywhere that I generate or use a time, and put in a change to use a 12-hour format - because they don't all do it the same way. Just a bit tedious and error-prone. And, of course, there would have to be an am/pm indicator, so the times wouldn't all be 5 characters, which complicates things a bit. I'll bear it in mind for Cumulus 2. Remind me when I forget. *** I'll try to forget to remind you!


Fair enough, but that isn't specifically what that section is about, so anyone looking for the format of the data log files for some other reason might not think to look there. What I should really provide is a proper guide to correcting the data when things go wrong for whatever reason. But like everything, it's finding the time. **** I'm going to invent some extra time to sell. I'd get rich! I expect that most of the problem is language. You speak the proper King's English I expect, while I talk Southern Red-neck American English (and never the twain shall meet!)

Again, thanks for your program and your time.

George

Steve,
The only file in Cumulus\Diags with todays date is Cumulus.009 and the attachment was not allowed due to extension 009. How should I send this?

[steve]

Something I read caused me to think this was a problem whose fix would defer to Cumulus 2.

Ah. I may have given that impression for some specific problem or other. But in general, it should work. Anyway...

I'll try to forget to remind you!

I could make the clock in the top right-hand corner display in 12-hour format fairly easily, but I guess that's not a great deal of use

The only file in Cumulus\Diags with todays date is Cumulus.009 and the attachment was not allowed due to extension 009. How should I send this?

I should probably make the BBS software accept .001 to .009 but I can't remember how just now. If you take a copy, and change the file name to cumulus.txt, then it should attach OK.

Edit: I've changed the settings so .001 to .009 should be accepted now.

Steve

[George Richardson]

Steve

I could make the clock in the top right-hand corner display in 12-hour format fairly easily, but I guess that's not a great deal of use **** That would be great whenever you get a round tuit.

I should probably make the BBS software accept .001 to .009 but I can't remember how just now. If you take a copy, and change the file name to cumulus.txt, then it should attach OK **** Done, will try again.

Getting late over there. I'll hurry.

George

[steve]

I've noticed one problem immediately. Well, not so much a problem, but it makes Cumulus much less efficient at start up. It looks like you have the Cumulus logging interval set to 5 minutes, and your station to 30 minutes. When Cumulus asks for the logger data at start up, it gives the time of the last entry in its log file. The Davis software can't cope with being given a timestamp that it doesn't have an entry for (e.g. 23:25 in this case) so it gives all of the data in the logger - all 2560 entries. Cumulus then has to sift through all of this to find the ones it wants.

It looks like something may have gone wrong during this 'sifting' process which resulted in the zero being recorded as the minimum temperature at the start of day. I have to admit that I'm not completely confident that the sifting code in Cumulus is bug-free, nor that the Davis code behaves correctly anyway in these circumstances. I'd recommend if at all possible that you change so that your station and Cumulus logging intervals are the same. I'll have a look at the 'sifting' code tomorrow anyway, to see if I can see any problems.

Steve

[George Richardson]

Steve,
FWIW I changed the the Cumulus Data logging interval to 30 minutes to match the station which is what I think you are saying is more important, matching vs. speed of frequency. Really, I just would like us to all get along. We'll see what has happened tomorrow AM. around 1 or 2 your time.
George

[steve]

I'd have done it the other way, make them both 5 minutes. That still gives you over 8 days worth of the data in the station logger, should you ever need it.

Steve

[steve]

I could make the clock in the top right-hand corner display in 12-hour format fairly easily, but I guess that's not a great deal of use **** That would be great whenever you get a round tuit.

Latest build (699) has option to make clock display in 12-hour format (click on it to toggle):

http://www.nybbles.co.uk/downloads/beta ... sSetup.exe

Steve

[George Richardson]

12:13: 46 PM