Tactics Plugin

[hasse_A]

I am trying out OpenCPN 5.0 with tactics pi but find it terribly sluggish.
So sluggish that you get the sign coming up saying the program has stopped
working. Do you want to end it or wait for it to answer.
This happens as soon as you open the tactics window.

Running on win7
OpenCPN 5.0.0+9065270
tactics_pi-1.0008-ov50-win32

All works fine on 4.8.8

/Hans

[hasse_A]

Additional info:

CPU usage = 25 % when tactics window present. otherwise only 1 %.

/Hans

[rgleason]

Hans, I tried it a little with VDR play and the second test file called Sample2.txt


While the control of the menus is less than ideal, the task manager was just about 3% CPU for O.


I don't know what to suggest. Perhaps try downloading a new file?

[RonSouthworth]

Sorry to read.

Only thing I can suggest Hans is maybe uninstalling and reinstalling the apps and plugins.

[hasse_A]

Originally Posted by rgleason

Hans, I tried it a little with VDR play and the second test file called Sample2.txt


While the control of the menus is less than ideal, the task manager was just about 3% CPU for O.


I don't know what to suggest. Perhaps try downloading a new file?

Hi Rick!

I have desiccated this. Of course I did a clean install and used my old opencpn.ini file. The program more or less freezes.
I then removed the .ini-file and started from scratch. I have found that as soon as you add wind history you are in trouble.
It's enough with just that instrument. With 4.8.8 no problem.

I am not really in the mode to do more debugging at the moment.
Please try and reproduce this behavior and maybe Tom can have a look at it.

/Hans

[transmitterdan]

Originally Posted by hasse_A

Hi Rick!



I have desiccated this. Of course I did a clean install and used my old opencpn.ini file. The program more or less freezes.

I then removed the .ini-file and started from scratch. I have found that as soon as you add wind history you are in trouble.

It's enough with just that instrument. With 4.8.8 no problem.



I am not really in the mode to do more debugging at the moment.

Please try and reproduce this behavior and maybe Tom can have a look at it.



/Hans

I think I know why wind history is a problem. I thought I fixed that long time ago. Let me look at it again.

P.S. the issue was the wind history used lots of calls to the system timezone function which is CPU expensive for some reason.

[rgleason]

TDan, I think I know what might have happened. You might have fixed it in sweepplot or dashboard (...but Dashboard does not have a history instrument does it?).


Anyway that would be great if you could fix it.

[transmitterdan]

Yes, I think it was Dashboard that was fixed. It has a wind history graph.

Where in GitHub is this plugin maintained?

[transmitterdan]

Ok, I found the github repository for tactics_pi. And it does not have the speed fixes for the history plots. I will submit a PR in a couple days that should speed up the history calculations by 1000:1.

[transmitterdan]

I have submitted the PR.

But there are many issues still remaining with tactics_pi. The components copied from dashboard_pi have changed a lot since they were copied. But no one has copied the changes over to tactics_pi.

Maybe it doesn’t matter because the dashboard components are always available to a user so they don’t really need the same ones from tactics_pi.

[rgleason]

Originally Posted by transmitterdan

I have submitted the PR.

But there are many issues still remaining with tactics_pi. The components copied from dashboard_pi have changed a lot since they were copied. But no one has copied the changes over to tactics_pi.

Maybe it doesn’t matter because the dashboard components are always available to a user so they don’t really need the same ones from tactics_pi.

I think it does matter, how hard is it to get them in alignment?

Or perhaps merge them together? --however Power boats don't need all that tactics has.



Are there particular areas that need alignment that you know of?


I think your PR History Optimization also includes this fix

"Fix crash when lat or long are empty in RMB message"?
Is that correct?

I have pulled your PR and was going to update Tactics for ov42 and ov50, but perhaps we should look at updating it more?


How would you suggest approaching this?

[Canne]

rgleason, I am interesting in testing the PR, I've pulled it into a local. But where I can get opencpn.lib to link with VS2013 ? Wiki-links are for <5.0.

[transmitterdan]

Originally Posted by Canne

rgleason, I am interesting in testing the PR, I've pulled it into a local. But where I can get opencpn.lib to link with VS2013 ? Wiki-links are for <5.0.

I think you need to move up to Visual Studio 2017 to work effectively with V5.0. Or you can wait until someone builds the plugin for OV5.0.

None of the O developers use VS2013 so far as I know.

[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?

[rgleason]

Canne, I haven't updated appveyor to run the ov50 build, I was going to do that but wanted to get the other issues resolved first.


To answer your first question. I have a copy of opencpn-4.2.lib -->opencpn.lib which should work, if it doesnt I have a later version
See below take the .doc off and rename it. Please.



However you can use the ov42 version that I tested, here
https://ci.appveyor.com/project/rgle...3867/artifacts
Use it with OpenCPN v4.8.8

while we get things fixed hopefully. Dan, thank you so much for the big fix. Can you advise please?