Facebook

Login

Support Sailonline

If you haven't already - join the SAILONLINE YACHT CLUB!

Please also consider making a donation - all amounts are greatly appreciated!

Board » Technical Discussion » How to handle wheatherdata differences ?

Posted by
Armchairsailor


Users avatar

Posts
6
Registered at
2012-03-22 09:35:22
2015-09-29 02:43:55 - How to handle wheatherdata differences ?
In my current position in the Timed transat
LAT 49°14.3747'LON W 022°55.8096'
DATE 2015-09-29 TIME 02:30:08
the wind information in the client
differs from the wind information in the gribviewer, which uses the file from
brainaids site:
"weather_186_global_gfs_20150928_2227.grb"

client TWD 122.34° TWS 2.82KT
gribview TWD 120 TWS 5.3KT

Meanwhile I have read that there are differences. The effect is that my routing software sends me in the wrong direction.
How do other SOL users handle this situation ?

--- Last Edited by Armchairsailor at 2015-09-29 08:17:01 ---
Posted by
kroppyer


Users avatar

Posts
347
Registered at
2010-02-05 15:25:55
2015-09-29 17:42:44 - Re: How to handle wheatherdata differences ?
The difference is there because sol uses a different interpolation method than your router/grib viewer does. Some routers/viewers offer multiple interpolation methods. If your viewer/router has that option, you can experiment with the different methods, see which one is closer.

In general, this problem occurs most often in large wind shifts (spacial, or over time). So usually it's in the light winds ("blue goo") that the router is off. I believe most people cope with that by hand steering as much as possible in the blue goo, and even better: avoiding the blue goo.

You could consider it as part of realism: winds are not always as they are in your router, especially not in light, shifty conditions.

If you're really bothered by this, you could try different routers/viewers to see which matches sol the closest, and you could write your own/adapt an existing one.
Posted by
Armchairsailor


Users avatar

Posts
6
Registered at
2012-03-22 09:35:22
2015-09-30 04:31:59 - Re: How to handle wheatherdata differences ? thnks
Tnks kroppyer for your reaction.
It is as you say. This resembles reality.
I was attrackted by the high wind speeds of the depression, but development of it changed, as usual. I had no chance at all to avoid the blue gue afterwards.
I am having thoughts to write my own router. But that won't change reality.
Posted by
Kipper1258


Users avatar

Posts
249
Registered at
2015-06-18 22:08:49
2015-10-12 22:23:32 - Re: How to handle wheatherdata differences ?
kroppyer said @ 2015-09-29 17:42:44:
...you could write your own/adapt an existing one.
Kroppyer, where can we source the SOL server interpolation code to see how it works / replicate it in other tools? It does not seem to be posted anywhere (at least Google can't find it)

Thanks
Posted by
kroppyer


Users avatar

Posts
347
Registered at
2010-02-05 15:25:55
2015-10-13 13:35:49 - Re: How to handle wheatherdata differences ?
Server code is not published, all we (solers) can do is measuring and fitting. I've done this in the past, didn't get perfect results but to me it's close enough for now.

In this thread I explain my findings, and I link to a post from Jakob (who, together with kalle, wrote most of the server-side code) explaining some more about the interpolation.

EDIT: fixed link (you should'be said it didn't work! :)

--- Last Edited by kroppyer at 2015-10-20 14:23:32 ---
Posted by
outlaw


Users avatar

Posts
33
Registered at
2012-06-01 13:37:55
2015-10-13 18:19:01 - Re: How to handle wheatherdata differences ?
Interpolation in space is simple bilinear. In time is a 'shape function', apparently 3*t^2-2*t^3. So you get
v(19:00)=.26*v(18:00)+.74*v(21:00)

For the polar use the 'detailed csv', where TWS interpolation is again linear. Not sure about TWA, but it's 1 degree resolution, *shrug*.
Posted by
kroppyer


Users avatar

Posts
347
Registered at
2010-02-05 15:25:55
2015-10-13 23:04:01 - Re: How to handle wheatherdata differences ?
For interpolation on the polar, I used the same method in SPINNACER as AGage used in his SOL Polar Explorer. See class AGage.PolarExplorer.Interp here.


AGage comments (in the code) "Interpolate as on sailonline.org", so I guess he did exactly the same as sol does.

I'm not sure if this code performs linear interpolation, but I wouldn't know what else. I'm pretty sure it's at least close to linear interpolation. Also "linear" looses meaning when you use polar coords... but now we're talking minor differences. Like outlaw says: *shrug*

Please login to post a reply.

Races

Next Race: 00d 00h 00m

Current Races:

Great Bear Ice Race 2025


You can fly to Great Bear Lake Airport, but if the ice is good, instead hire a pick-up truck, load your disassembled DN into the cargo bed and drive, first following the McKenzie River north, and then straight across the lake to the start of our final Ice Boat series race of roughly 100nm in length. The Airport authorities have agreed to leave all lights on, so you should just about be able to see where you are going as you zing and rumble, PL-free over the ice!
Race #1853
INFOby brainaid.de
DN PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking: DN - SYC
RACE CLOSE: Friday,
December 12 at 2300 UTC.
Race starts: Dec 07th 09:00 Registration Closed
▶ Flash
GO TO RACE

Chile Islands RTI Sprint 2025


Across the Canal Puyuhuapi from Puerto Cines on the Chilean mainland lies the Parque Nacional Isla Magdelena, and further west across the Canal Moraleda from Magdelena lie the islands of the Chonos Archipelago. Returning for the first time since 2017 we find ourselves in vari-charted waters here, with the mapping between physical islands and their names not 1:1 but rather some:some. It’s only 26nm but you could get lost! Good luck.
Race 1980
INFO by brainaid.de
First 40 PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking:
SPQ4 - SPRCH - SUPSOL – SYC
RACE CLOSE: Thursday,
December 11 at 2300 UTC.
Race starts: Dec 06th 21:00 Registration Closed
▶ Flash
GO TO RACE

Christmas to Christmas Island 2025

Two waypoints, start and finish, and between them 5827 NM of two oceans. Not much time if we want to sit at the Christmas Eve table. You will need to maintain an average speed of over 10 knots, maybe quite possible with this Maxi Trimaran, but prepare your vessel well and choose the best of the many possible routes. Hands up everyone who knew there were TWO Christmas Islands?! OK, maybe you have sailed this Sailonline course before... but it's time to get ready for the 2025 challenge of racing between Christmas Island in the Pacific to Christmas Island in the Indian Ocean! It also marks the conclusion of our prestigious Ocean Race Championship 2025. Please have fun! Fair winds!
PRIZE: SMPF
Race #1967
INFO by brainaid.de
Maxi Trimaran PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking: OCQ4 - OCCH - SUPSOL - SYC
Race starts: Dec 01st 11:00 Registration Closed
▶ Flash
GO TO RACE

Sinbad by Balloon 2025 - Carnarvon to Dondra


From here, our home in Bharatavarṣa is now north west of us – said Sinbad to his fellow balloonists. Perhaps we can ride the wind first further north, and then catch the winds that every year bring the rain, perhaps not. It’s 2600nm and we could be aloft awhile, so, Master el-Quarters, victuals only, no sandbags, provisioning the giant hamper. It will be not a picnic!
Race #1886
INFOby brainaid.de
SOL Balloon PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking: SYC - SBB
RACE CLOSE: Tuesday,
December 9 at 2300 UTC.
Race starts: Oct 15th 11:00 Registration Closed
▶ Flash
GO TO RACE

Go to race archive

SYC Ranking

  1. Sailonline Yacht Club Member KaSToR
  2. Sailonline Yacht Club Member CriticalHippo
  3. Sailonline Yacht Club Member WRmirekd
  4. Sailonline Yacht Club Member rafa
  5. Sailonline Yacht Club Member vida
  6. Sailonline Yacht Club Member CollegeFund
  7. Sailonline Yacht Club Member bonknhoot
  8. Sailonline Yacht Club Member Panpyc
  9. Sailonline Yacht Club Member sassy63
  10. Sailonline Yacht Club Member BRENTGRAY

View full list

Series

Mobile Client

SYC members have the benefit of access to our mobile/lightweight web client!

The mobile client