OpenCPN Beta Version 3.1.1206 Released

[Gilletarom]

Hello,

Maybe someone has already said what I reported here. If this is the case, I apologize.

I installed with a permanent layer zone boundaries weather. I loaded the data here:
Layers Library | Official OpenCPN Homepage

If I click on the border areas, OpenCPN displays the same menu for traces

See screen copy.

B.R.

Gilletarom.

[RonSouthworth]

Hi Development team,

Thanks for a great collection of work.

I am hoping to develop some sail performance tool(s) (plugin most likely), once I get back up to coding speed - not done regular programming work for more years than I care to admit. I am presently working with the math and trying various algorhythms for accuracy vs performance

I have been testing Beta 3.1.1105 for about a month now on linux UBUNTU 12.04 LTS using the CM93 V 2 Charts

In setting things up (e.g. finding suitable NMEA instrumentation simulation software) to be able to do justice in testing this version is stable in what testing I have been performing.

I am hoping to continue to use the program's NEMA connection feature that uses UDP and network traffic. which has proved to be a real good way to get to know the program and some of the various OS quirks along the way. After going thru the various FAQ etc I discovered the "final problem" with this feature "not working for me" on UBUNTU required the iptables to be flushed clear.

The standard tables are overly complicated, when I figure out a fix to the iptables "problem" I will let you know the necessary mods to make it work under that platform with that configuration.

So in working out where the issues were I have found that the packaged up version recently released 3.1.1206 under windows or Linux UBUNTU 12.04 LTS does not see the NMEA UDP sentences. The USB Garmin GPS support works well with windows. UBUNTU - I am still sorting outthe garmin USB issues.

I have downloaded the latest beta from git in an attempt to look at compile options / compare the two versions however what I have downloaded via git comes up with compile errors sadly.

A fresh snapshot un -updated 3.1.1105 downloads and compiles successfully.

The latest beta screen perfomance appears to be much improved however without the UDP feature enabled I cannot do any real soak testing with my setup at least.

I hope this feedback is helpful and once again thanks for what looks like a great product from where I have been sitting so far. Eagerly awaiting the time in the not so distant future when I can try her out in the water....

[nohal]

Ron...
Which compile errors exactly are you seeing with the latest code? I'm on Ubuntu 12.04 based 64bit Mint and see none and believe Dave's primary machine is running stock 12.04 so I can imagine problems anywhere but on this config...
Be sure not to forget to run "cmake .." after you pull the latest code from git

Pavel

[RonSouthworth]

Hi Pavel,

Thanks for the quick reply!

Between when I last posted and you replied I got 5 minutes and checked the repository for updates using git per the documentation. To my surprise it found quite a few files this morning

I re-ran cmake which I had done before in case there was some changing structures to the new version.

good news is I can now compile the latest vanila beta code .

The only problem I was seeing with it was that the compiler was stumbling on a declaration error not able to find version.h all over the place so somethign had to be missing somewhere.

Maybe I grabbed a very early hit on the version from repository and it was not complete as I have had that problem with git before where it flaggs as up to date but isn't really....

I saw the updates were available before the announcement and have been compiling on that lot of documents the previous version has bee quite stable so I have been reluctant to change.....

I have just tried the later version after compiling and sadly the tcpip feature appears to be non functional out of the barrier.

I will delve into why that is the case. I roll back to the prevous version re-write the config info and the interface works again.

I am setting it all up into Eclipse at the moment to make my life easier with coding. As I say just getting back into doing some stuff (been playing with a TCL gps simulator for a week or so)

Loving these fancy tools that are available now for helping with debugging things. It should be fairly easy to find what tha difference is in that area that is stopping it from working on my config. I am running the intel version of Linux on a older dell box for what that is worth at this stage.

Once again thanks for the reply.

Ron

[nohal]

Ron...
version.h is generated by cmake (you have to run "cmake .." even after each make clean) so it looks
Good that you are now able to do the build. If you find some more problems, let us know.

Pavel

[MarkJ]

Hi

I have one more problem that seems to be from a change in the AIS setup.

When there is an aboration in the GPS position the boat, either mine or the AIS target gets shifted to 00000 0000 L&L

I cant send an image at the moment as i am at sea.

The other one is the inability to delete waypoints from a route. Either delete them or remove them from the route.

This is highly annoying. It has been in every other version except this one.




Mark

[bcn]

Mark,

something like this?

http://www.cruisersforum.com/forums/...ost-94469.html

bcn

[bdbcat]

MarkJ....

We will need to know what NMEA source produced the move to lat/lon 0,0.
Do you have any way to reproduce the effect?
NMEA sentences are protected by checksums, so it seems unlikely that OpenCPN would respond to a garbled message with incorrect checksum. So, we may start be assuming that it was a legitimate message reporting lat/lon of 0,0.
Tell us more about your hardware setup, please.

Regarding delete of routepoint: We heard you the first time. Patience, please....

Thanks
Dave

[muttnik]

You seem to have a problem in the UDP code in 3.1.1206, on ubuntu 12.04 64-bit at least. I was wondering why I was seeing nothing in the debugging window, despite being able to see the broadcasts with tcpdump or with kplex and why opencpn didn't appear to have a socket listening on the port I'd specified. lsof showed a UDP socket on a high port I hadn't specified, so I tried broadcasting to that and Bingo! nmea input, tagged in the debugging window with the port I'd asked for, not the one I ended up using. The port used changes each time I hit "ok" in the connections tab even if I don't alter the connections settings. Ports I tried requesting were 10110 and later (because it was used in docs) 4661. ports actually bound to are all >2^15: 56632,48212,57184,56244 and 56331 were the last 5.

Also...documentation doesn't say whether "Address" for UDP should be the address of the server, the broadcast address or the unicast address of a local interface to be used. I note from post 1 in this thread that blank means INADDR_ANY. Tried all of the above and wildcard is used regardless of what I specified, although perhaps that is part of the other bug?

Hopefully I'm not just doing something incredibly stupid here...

[bcn]

Hi Dave,

I have the impression that rmc.cpp is not checking the content of the

Mode Field (#12) of the RMC message

so we might have a sender who puts 0.0N, 0.0E for position and an "N" in field 12 indicating a non-valid GPS fix.
A = Autonomous mode
D = Differential Mode
E = Estimated (dead-reckoning) mode
M = Manual Input Mode
S = Simulated Mode
N = Data Not Valid

I will try on the next trip to log the NMEA stream from the AIS unit and if there is another glitch in the GPS reception we can have a look.
On the other hand I'm waiting for info from em-trak how they (think they) treat this situation.

bcn

[cagney]

Originally Posted by Gilletarom

Hello,

Maybe someone has already said what I reported here. If this is the case, I apologize.

I installed with a permanent layer zone boundaries weather. I loaded the data here:
Layers Library | Official OpenCPN Homepage

If I click on the border areas, OpenCPN displays the same menu for traces

See screen copy.

B.R.

Gilletarom.

The layer is made up of tracks. There is currently no other way to draw lines. So of course, if you right click on a line, you will see a menu.
There is one problem, that needs correcting. Originally the wpts were invisible. Not so anymore, due to changes. Easy enough to fix. Just change the wpt icon to "Empty".

Thomas

[Hakan]

Originally Posted by bcn

Hi Dave,

I have the impression that rmc.cpp is not checking the content of the

Mode Field (#12) of the RMC message

so we might have a sender who puts 0.0N, 0.0E for position and an "N" in field 12 indicating a non-valid GPS fix.
A = Autonomous mode
D = Differential Mode
E = Estimated (dead-reckoning) mode
M = Manual Input Mode
S = Simulated Mode
N = Data Not Valid

I will try on the next trip to log the NMEA stream from the AIS unit and if there is another glitch in the GPS reception we can have a look.
On the other hand I'm waiting for info from em-trak how they (think they) treat this situation.

bcn

Hi. Just a hint. Maybe unnecessary but anyhow: Keep in mind that some old items don't transmit the 12th field of the RMC sentence.

[P_Dub]

Good Day
Is it possible to have a plugin for implementing the new CHS current and tides features?
Here is the address:- http://ogsl.ca/ocean/index.jsp?lg=fr
The 3 type of informations are:- current, tides and ice coverage.
Take a look and let me know what you think of it.
Best Regards.

Nota:- I have no means of building a plugin

[bcn]

Originally Posted by Hakan

Hi. Just a hint. Maybe unnecessary but anyhow: Keep in mind that some old items don't transmit the 12th field of the RMC sentence.

Hakan,

yes, and this is correctly contemplated in the NMEA library.

bcn

[MarkJ]

Originally Posted by bdbcat

MarkJ....

We will need to know what NMEA source produced the move to lat/lon 0,0.
Do you have any way to reproduce the effect?
NMEA sentences are protected by checksums, so it seems unlikely that OpenCPN would respond to a garbled message with incorrect checksum. So, we may start be assuming that it was a legitimate message reporting lat/lon of 0,0.
Tell us more about your hardware setup, please.

Regarding delete of routepoint: We heard you the first time. Patience, please....

Thanks
Dave

I have a Raymarine AIS650 Class B Transponder via USB into a Windows 7 laptop as AIS GPS Shared.

It is definitly this update thats done it cos i can show the tracks off the previous 6 months.

I know sometimes the AIS on its own software, ProAis will show the 000 postion sometimes for a moment. And of course, when i start open cpn before the AIS has a gps location it sets at 000


Hope this helps.

Originally Posted by bcn

Mark,

something like this?

http://www.cruisersforum.com/forums/...ost-94469.html

bcn

Yes! Exactly!!!

Took a photo of the laptop on the iPad... Lol i hope this works