Tactics Plugin

[rgleason]

OK, the windows version for ov50 is here
https://ci.appveyor.com/project/rgle...3141/artifacts
tactics_pi-1.0009-win32.exe It is not named with an "-ov50" so please rename it!!



Also more importantly, I have not tested it!! Good Nit.

[transmitterdan]

Originally Posted by Canne

Thanks, transmitterdan, for your efforts. I'll wait, no problem! Don't forget the polar performance history graph, it is also sluggish, can't live without it. For the rest, why not integrate tactics functions in the O dashboard, like a race-mode?

I added the polar performance history to the pull request this morning. Someone will have to pull it or you can build from here:

https://github.com/transmitterdan/ta...ryOptimization

[rgleason]

Thanks Tdan, I haven't tested these but they both compiled


for Testers only:


tactics for Opencpn v4.2 - v4.8.8 please rename to "-ov42"

https://ci.appveyor.com/project/rgle...3981/artifacts
This is from my tactics_pi "master" branch

Testers please report results here! Thank you.


tactics for Opencpn v50 please rename to "-ov50"
https://ci.appveyor.com/project/rgle...uilds/24204042
Then pick the "Artifacts" Tab.

This is from my repository tactics_pi "ov50" branch
https://github.com/rgleason/tactics_pi/commits/ov50 You should see a little green check in a few minutes. Click on that and select Appveyor.


Testers please report results here.

[RonSouthworth]

Hello Team

Thanks for uploading the changes.

I have been waiting for someone to make a plugin like this for a while so I am very appreciative of the efforts.

I've been testing OpenCPN5 and the changes to the tactics plugin generally since V5 release. Compiling wxWigets 3.1.2 Has caused me some challenges for a while and recently I am confident I have sorted it for MacOS Mojave and Ubuntu Studio 19 (Development version of Ubuntu) I have found OpenCPN 5 is very happy with the low latency kernel as an aside.

Performance wise on macOS Mojave The new code for the data trending doesn't adversely impact on the CPU here, I don't use it a lot only intermittently.

The polar performance and true wind angle indication on the chart is really useful.

I have not yet commenced any code debugging of the tactic's plugin code base, so below are bugs I have observed. I have OpenCPN configured to retransmit all messages to aid in diagnosing what is happening on the communications layer. No other filtering or message rewriting is enabled. If any is required I have missed this info in the documentation.

Based on above,


1. If the tactic's temp WP is not removed when you go to restart OpenCPN5 on macOS it crashes OpenCPN forcing a work around (delete the WP someway and then restart).

I haven't been able to do anything at all to the plugin thus far to cause a crash in any other way.

Some of the value calculations being incorrect have been reported previously. I know the team is aware of these issues and is planning fixes from reports. The next item might be a clue.

2. SoG is being used to calculate SoW, even when an instrumented value is presented to OpenCPN5.

I seemed to recall there are SOLAS requirements* for - AIS vessel speed to be "SoW (* yes there are additional requirements). The ShipDriver Plugin is an AIS message source !....

In trying to gain some clarity which I could not gain from using the GPS source simulator data I have available, I started using the Ship Driver Plugin.(Tweaked a little on the U/I, AIS messaging un altered, now, here under test is another potentially working on MacOS plugin).

The water only got a little muddier instead of clearer, maybe..

The Open CPN message Box "decode" shows the AIS data speed as "SoG". (See attached log image)

OpenCPN does rebroadcast NMEA VHW messages using "SoW data", and from my testing I believe it is getting that data from the AIS message provided by Ship Driver.

This being the case, it seems to suggest to me the possibility that the data message frames are actually valid, and it is just a typo on the message data type label for AIS message decode display ?

Using that assumption, I have observed the VHW data isn't being used to display or used to create the calculations. Is it really a prerequisite for plugin operation?

From testing here it is only using SoG to calculate SoW. (Screen Shots provided)

Using my GPS simulator software which does not generate SoW data, the plugin happily calculates SoW data.

In addition to this I have found if on restarting the app if I send a few data messages from the GPS source in particular, valid true wind and applied wind values, the Applied angle and True wind angle indication instruments start to display values when I source the AIS data alone. (Screen shot provided)

This is the point I have reached so far in testing the tactics plugin in the last week including the recent changes.

I thought I would provide some early feedback and see if I can gain some confirmation the above instrument value info and performance is correct?

All the best


/Ron

[bcn]

Originally Posted by RonSouthworth

2. SoG is being used to calculate SoW, even when an instrumented value is presented to OpenCPN5.

I seemed to recall there are SOLAS requirements* for - AIS vessel speed to be "SoW (* yes there are additional requirements). The ShipDriver Plugin is an AIS message source !....


/Ron

Ron,

all references I have for AIS speed tell that it is SOG - what make sense to me.
http://catb.org/gpsd/AIVDM.html as public reference for example
Your source?

Thanks

Hubert

[RonSouthworth]

Hi Hubert

I sent you a Private Reply to Off Topic In response to your question

/Ron

[rgleason]

Thank you for testing this plugin which was started by Thomas a number of years ago. All found errors have been corrected so far as I know, with the exception of the bug you found.


So please advise in more detail if you have found other problems.


Tactics_pi is an extension of code from Dashboard_pi so this may be helpful:

https://opencpn.org/wiki/dokuwiki/do...shboard_plugin

See bcn's link above - Types 1, 2 and 3: Position Report Class A

50-59
10
Speed Over Ground (SOG)
speed
U1
See below

Speed over ground is in 0.1-knot resolution from 0 to 102 knots. Value 1023 indicates speed is not available, value 1022 indicates 102.2 knots or higher.

[rgleason]

You do know about the preferences menu?
There is a choice for "Use SOG instead of STW for true wind" for example.

[rgleason]

One thought, you could use VDR or ShipDriver and create nmea sentences with a text editor to test each instrument sequentially.

[rgleason]

Ron,
I would like to make a CI that compiles plugins for MacOS, but I don't have that environment. You might be able to help me. It appears you are compiling with MacOS, if this is correct are you using this process? -Also which method?
https://opencpn.org/wiki/dokuwiki/do...piling_mac_osx


Also then what are the specific steps that you must take to compile a plugin so that it can be packaged?

[RonSouthworth]

Originally Posted by rgleason

One thought, you could use VDR or ShipDriver and create nmea sentences with a text editor to test each instrument sequentially.

Thanks Rick

Using VDR and scripts is one way, for certain,

some folks have suggested that to me.

VDR doesn’t compile on macos Mojave one of a few plugins I want to use that has some errors to debug. The offical install works a good example of the challenge with reverse compatibility

Im keen on using simulating tools that i can set the variables to try across the valid range of values for a given message type
Doing this in unison with testing as installed instruments on the work bench and at the dock before a voyage is my preferred approach

As approaches go it is tried and tru.1

:Ron

[RonSouthworth]

Originally Posted by rgleason

Ron,
I would like to make a CI that compiles plugins for MacOS, but I don't have that environment. You might be able to help me. It appears you are compiling with MacOS, if this is correct are you using this process? -Also which method?
https://opencpn.org/wiki/dokuwiki/do...piling_mac_osx


Also then what are the specific steps that you must take to compile a plugin so that it can be packaged?

Hi Rick

The process to package is as per the document wilki.

It is very similar to how other open source projects are packaged.. e.g. MuseScore3

I found it is more successful to issue the command line sudo make create-dmg rather than letting mesa prompt for the root password, if the cpu is loaded down processes can time out issuing the elevated user up front seems to reduce this happening it is a very cpu intensive activity. bst to not be doing other tasks on a low end machine such as i have

The same issues with 32 bit OS vs 64 Bit apply. So i can compile for Mojave but backwards can have issues. Same problems and more for Qt 5.12 MuseScore3 has to be compiled with a few versions to save the end users they have included code for release versions to check updates and download either on startup or manually invoked.

with macOS people also may have a preference for the dark look over the vanilla look dark look isn’t fully supported by wxWidgets 3.1.2 either.

I prefer that look it does impact on how things are compiled colour defaults file menu and the like. Night mode on the chart screen is a lot easier on the eye for example.

/Ron

[RonSouthworth]

Originally Posted by rgleason

Thank you for testing this plugin which was started by Thomas a number of years ago. All found errors have been corrected so far as I know, with the exception of the bug you found.


So please advise in more detail if you have found other problems.


Tactics_pi is an extension of code from Dashboard_pi so this may be helpful:

https://opencpn.org/wiki/dokuwiki/do...shboard_plugin

See bcn's link above - Types 1, 2 and 3: Position Report Class A

50-59
10
Speed Over Ground (SOG)
speed
U1
See below

Speed over ground is in 0.1-knot resolution from 0 to 102 knots. Value 1023 indicates speed is not available, value 1022 indicates 102.2 knots or higher.

Thanks For the helpful reply.

/ Ron

[rgleason]

Ron,

"The process to package is as per the document wilki.
It is very similar to how other open source projects are packaged.. e.g. MuseScore3"


Musescore has a lot of plugins I find. Are you familiar with how they produce these by any chance? Looks like an interesting opensource program!

https://musescore.org/en/plugins

[rgleason]

Ron, everyone has their own way of doing things, but TransmitterDan has made a python script that will play nmea files just like VDR_pi. In many ways it is better because it is a separate program. I've used it for testing quite a lot. Mac may be different for setup, but it is here:

VDR Server
https://opencpn.org/wiki/dokuwiki/do...rver_for_tests


Thanks for the build info.

What are the critical differences building for Mojav? In terms of setup?