World Magnetic Model Plugin

[seandepagnier]

Originally Posted by Gilletarom

Hello Pavel,

If we choose a very high accuracy and a very small step, time of calculation becomes very very large.
It is necessary to calculate all points outside the navigation area?

B.R. Gilletarom

This is no surprise I hope. This is why you have the option of using a larger step and lower accuracy.

No, it does not need to calculate all points as it currently does. Instead it could only calculate in the navigation area then update it when you scrolled, but in the case that you moved to view a magnetic pole it would stop for a while to calculate. It would require more logic to implement this correctly, patches welcome.

[rgleason]

Git cloned wmm_pi and cmake, then MSVC debug.
Win7 laptop Dell 15xls, First time tried to render the map, it stopped with a break.
See string of screen shots and text please. Tried to continue, but it did not really want to.

Then tried it again similar.

[seandepagnier]

Thanks for this feedback. The callstack is the only useful data so you could omit the images in the future.

I have put a potential fix in my git, so please try it and we can hopefully close this issue. It is in my fork, not nohal's

[rgleason]

Git cloned WMM_pi from Sean's repository this time.
git clone https://github.com/seandepagnier/wmm_pi.git
did cmake's and found ..build/plugins/wmm_pi/debug had a newly compiled dll (current data and time). Copied to ..build/plugins.
Ran MSVC debug.

Tested WMM version 0.7 (says 0.4 in icon).
Set Step and pole to far right, took a long time (showed 3hrs) finally failed.
Set Step & Pole to middle (about 6 minutes) failed.
Set Step & Pole to left (about 5 minutes) failed.
All failed with similar initial break Bitmap.cpp[1089]
See attached reports please.

It would be helpful to have an indicator on the sliders.
Don't know what but here are some ideas.
Processing speed (-) ====== (+)
Processing (faster)=======(slower)
Steps (big)======(small)
Steps (fewer)=====(smaller)

[seandepagnier]

Originally Posted by rgleason

Git cloned WMM_pi from Sean's repository this time.
git clone https://github.com/seandepagnier/wmm_pi.git
did cmake's and found ..build/plugins/wmm_pi/debug had a newly compiled dll (current data and time). Copied to ..build/plugins.
Ran MSVC debug.

Tested WMM version 0.7 (says 0.4 in icon).
Set Step and pole to far right, took a long time (showed 3hrs) finally failed.
Set Step & Pole to middle (about 6 minutes) failed.
Set Step & Pole to left (about 5 minutes) failed.
All failed with similar initial break Bitmap.cpp[1089]
See attached reports please.

Ok, don't worry about the icon it is old.

The step size is smallest left largest right, this makes it slowest to the left.
Pole accuracy is lowest left highest righht, makes it slowest to the right.
[/quote]
It would be helpful to have an indicator on the sliders.
Don't know what but here are some ideas.
Processing speed (-) ====== (+)
Processing (faster)=======(slower)
Steps (big)======(small)
Steps (fewer)=====(smaller)[/QUOTE]

Yes maybe. I didn't think it was an issue since it only takes 3 seconds to generate the map with the default settings on my netbook. This is without optimizations in the compiler.

That is why step size should be set larger, pole accuracy less.

As for the actual bug.. it has nothing to do with any of that and it occurs in all cases. I pushed another change into git to hopefully correct it, just run a quick test to see if it helps any.

Thanks.

[rgleason]

Thanks Sean, WMM works now.
WMM-pi.dll.doc attached (windows). Take the doc off and test it please.

What you describe for Step size makes perfect sense.
for faster processing:
Big Step size = slider right
Less Polar accuracy = slider left

Took about 3 seconds as you said.

[NAV]

Dutch language files for WMM-Plotting.

Please note that in the plugin now both the terms variation and declination are used to describe the angle between true north and magnetic north.

The correct nautical expression in English is: variation (and not declination).

[cagney]

I'm trying to document the WMM plugin, and comes across an issue.

boat_alexandra makes it possible to plot the total field strength F.
This has no navigational use as far as I know.
What is interesting is the horizontal component H.
This is important as it governs the response of a magnetic compass and the value is important near the magnetic poles to judge if a compass is of any use.

Bowditch 1977 says this about the horizontal intensity in the chapter about Polar Navigation:
< 0.09 oerstedt -> reduced reliability
< 0.06 oerstedt -> erratic
< 0.03 oerstedt -> useless

A quick look at some maps suggest that these limits are very approx 10000, 7000 and 4000 nano tesla.

I suggest that WMM plots H instead of F.

Thomas

[rgleason]

Thomas, a question for clarification. Are you saying that the total field strength F is calculated in the "plot", but WMM should plot H "horizontal" instead because it would advise when the compass readings were unreliable?

Also just as a thought, perhaps an "information" area in the WMM plugin would describe these important factors and considerations for guys like me who don't know that much except about deviation.

Perhaps this stuff should be getting into Flyspray?!!!

[cagney]

Originally Posted by rgleason

Thomas, a question for clarification. Are you saying that the total field strength F is calculated in the "plot", but WMM should plot H "horizontal" instead because it would advise when the compass readings were unreliable?
.....

Yes that's right. There is now an option to plot F. I want it replaced by H, that is useful for planning in certain areas.
It seems easy enough to change.



Thomas

[krafthaus]

I'm probably missing something basic. Open CPN says "get a new WMM.COF from NOAA and place it in the location you can find in the OpenCPN logfile".

I downloaded the WMM.COF file which is 5kb but I don't understand where to copy it to? I would think it should go into Open CPN/plugins with the .dll files for the grib and dashboard files, but when I tried that the WMM didn't show up under Options -> Plugins.

Can someone point me in the right direction where to copy this to please?

[nohal]

krafthaus...
There's no need for a new COF file. The WMM2010 embedded in the plugin is stil valid and there's no newer version available.

Pavel

[krafthaus]

Originally Posted by nohal

There's no need for a new COF file. The WMM2010 embedded in the plugin is stil valid and there's no newer version available.

Which folder in Win7 should I see WMM.COF in then so that Open CPN shows WMM in Options->Plugins?

I don't believe I have it other than the download??

In Options->Plugins I only see the Grib & Dashboard plugins that came with the CPN download.

[nohal]

None, forget about WMM.COF, the table is embedded in the plugin. All you need to do is copy the plugin DLL to your plugin folder (Most likely C:\Program Files\OpenCPN\plugins).

[krafthaus]

Originally Posted by nohal

None, forget about WMM.COF, the table is embedded in the plugin. All you need to do is copy the plugin DLL to your plugin folder (Most likely C:\Program Files\OpenCPN\plugins).

Got it. I understood from the Open CPN info that I needed the .COF file from the NOAA site.

I went to the site:

opencpnplugins | Free Science & Engineering software downloads at SourceForge.net

and clicked the "Files" tab where I found a selection of plugins including WMM. When I clicked wmm_pi I was able to download the .dll file.

Once this was in the C:\Program Files\OpenCPN\plugins folder and I restarted OpenCPN it worked.

Appreciate your work on these Pavel.