Sean Depagnier's Plugins

[bcn]

I was thinking about route planning (Weather routing) and route checking here.
Same underlying logic.

[Hakan]

Ahaa - didn't now that the base map was used also there. Than I understand it better. Thanks
Håkan

[seandepagnier]

Originally Posted by AedanC

It works much better than the previous beta, but some runs are failing with "Polar: Failed". Any idea where I should start playing to get around this?

Aedan


Sent from my iPhone using Cruisers Sailing Forum

This likely means that the polars used by the specified boat do not define boat speed in the given conditions. This could happen for example, if the wind speed is greater than what the boat's polar defines. You could for example, add an additional polar or augment existing polars to define the needed boat speeds.

[AedanC]

Originally Posted by boat_alexandra

This likely means that the polars used by the specified boat do not define boat speed in the given conditions. This could happen for example, if the wind speed is greater than what the boat's polar defines. You could for example, add an additional polar or augment existing polars to define the needed boat speeds.

While trying out this suggestion I got side tracked with another, possibly related issue. I'm seeing what looks to me like some rather odd isochron propagation, see the attached image. It's like they just don't propagate at all in one area so have to sneak around the bottom..... I suspect it might be to do with periods of very light winds being forecast.

If the lowest wind speed in the polars is (say) 6kts, and the grib file contains a prediction of 3kts will weather_routing interploate or go with zero?

Do these isochrons look like what one would expect to see in light conditions?

Thanks

Aedan

[rgleason]

AedanC, you are trying to get from point A to B on the coast. To do this the fastest route appears to sail directly offshore for quite a way, tack for awhile and then tack back for a long tack.

I am not sure I understand why. Do you? You would have to look at the grib file and weather routing step by step, considering what the best option would be yourself, (just a rough estimation) then you would understand it better. The route is probably the fastest given the grib prediction. The isochrones look normal, but they've been overlayed on top of each other I think, as it calculated each optimal leg, so you get that folded effect.

Also run it again, ( perhaps we need a way to slow the calcs down for watching?) and try watching how it chooses the optimal route based on the next isochrone created (each isochrone represents equal sailing time based on the boat's polars).

I would be interested to hear what others say.

[seandepagnier]

Originally Posted by AedanC

While trying out this suggestion I got side tracked with another, possibly related issue. I'm seeing what looks to me like some rather odd isochron propagation, see the attached image. It's like they just don't propagate at all in one area so have to sneak around the bottom..... I suspect it might be to do with periods of very light winds being forecast.

If the lowest wind speed in the polars is (say) 6kts, and the grib file contains a prediction of 3kts will weather_routing interploate or go with zero?

If the lowest speed in the polar is 6kts then the boat _cannot_ sail in less than 6 knots hence the propagation seen.

If you need it to interpolate down to 0, then add polar speed of 0 for 0 knots of wind, and it will interpolate between 6 knots and 0 knots. This is not a bug, it is a feature.

[AedanC]

Ok, that might explain it. The next time there is a rainy day I'll try again with a column of zeros for zero wind speed and see if that helps.

Thanks,

Aedan


Sent from my iPhone using Cruisers Sailing Forum

[rgleason]

OpenCPN Beta File Thingie rguser rgpass

In directory 4.4 Win PI have uploaded current plugins
watchdog_pi-1.9.050b-ov4.4-win32.exe
weather_routing_pi-1.10c-ov4.4-Sean-Final-win32

which are improvements.

Sean, please note the weather_routing wxForm is not working well on this win tablet which has a high res screen, and the fields are all scrunched. Is there some way to get this fixed? I have the text increased in size 200% to be able to read, and the screen res is 2160 x 1440 and I think that is messing up the format of menus somehow....

[seandepagnier]

rick,

can you make a screenshot?

[rgleason]

Sean here are some screenshots. Maybe I have the text too big. I can reset it to 175% too which works for my eyes, but these shots are taken with 200%. I think I should see if that Win10 Display parameter affects OpenCPN text....?

[rgleason]

Interesting, when I change Win10 Setting Display Size of text to 175% it works better.

[AedanC]

Originally Posted by AedanC

Ok, that might explain it. The next time there is a rainy day I'll try again with a column of zeros for zero wind speed and see if that helps.

I downloaded the newer version (weather_routing_pi-1.10c-ov4.4-Sean-Final-win32), added a column for zero knots wind and things then performed as I was expecting, all looks good now.

One small buglet I bumped into afterwards: After I added a new column in the polar grid for a higher windspeed I started getting Polar Failures for all routes. I cured this by removing the .pol diagram from the boat data and then readding it.

Thanks for all the hard work.

Aedan

[seandepagnier]

Aedan,

Is it easy to reproduce? You add a column for which wind speed?

[AedanC]

I can't reproduce it right now as I'm away from my laptop, but I had a working configuration using a .pol file of my own which was a modified version of Beneteau421.pol. I added a column for 35kts and then all my routings failed with a Polar Failure. When I removed the .pol file from the boat xml file and re-added it then everything worked ok again.

If that's not enough to reproduce it then let me know and I'll do a step-by-step version when I'm back at my laptop.

Aedan.


Sent from my iPhone using Cruisers Sailing Forum

[kiwinz1]

Just want to follow this to see when i can get the Weather Routing working again on my mac