Gaps in the windmodel in JV, windspeed down to 2 knts ?
Post by iconPost by yklg | 2019-01-03 | 07:34:20

Hello, router ZEZO seems very strange today, no ? Thanks for your responses

commenticon 21 Comments
Post by iconPost by zezo | 2019-01-03 | 09:39:30
The NOAA server is returning partial data today.

Download is done via http range requests, and the server returns less data than requested.

This had not happened before, so there was not a check for that error.

I added the check and it will avoid the "2kt N wind" problem, but there will still be delays and discrepancies while the NOAA server is acting funny.
Post by iconPost by yklg | 2019-01-03 | 10:25:03
Many thanks Cvetan for your explanation, I thought that it was not a bug of your router.

Post by iconPost by zezo | 2019-01-03 | 11:18:36
It happens 2-3 times/year. I guess some of the load-balanced servers/routers at NOAA goes down and it takes few days before someone notices and fixes it.

The system was giving timeouts before but now is acting weird. A packet dump shows loss of a single packet from the stream, which never gets retransmitted. Could be some firewall/IDS/router along the way doing weird filtering. It can also be in another location along the way from here to Boulder Colorado.
Post by iconPost by yklg | 2019-01-03 | 11:30:18
I hope it's not because of American shutdown due to "Trump wall" ?

Post by iconPost by zezo | 2019-01-03 | 11:36:48
They should not be so stupid to shutdown the national weather forecasting service, but you never know.

BTW the NOAA will be switching the forecast to a new model this year, so there may be some changes in the schedule and other details, but this is unrelated - it's some kind of network issue now.
Post by iconPost by yklg | 2019-01-03 | 12:15:41
I hope, but if you go on website :, it said "see"...
Post by iconPost by zezo | 2019-01-03 | 12:59:14
Found the following piece of information:

Jan 2, 2019: is changing the URLs from http:// to https://

Changed it here too, seems to work.
Post by iconPost by yklg | 2019-01-03 | 13:33:44
Oh yeah it's a good new, you're the best !
Post by iconPost by yklg | 2019-01-04 | 13:50:00
Hello Cvetan, does the problem with NOAA begin again now ?
Post by iconPost by yklg | 2019-01-04 | 13:55:48
I saw it in the predicted times to equator with the file of 4/1- 6h am

Post by iconPost by zezo | 2019-01-04 | 16:21:07
I think it's just the weather pattern this time. Two high pressure systems blocking the trade winds.

With a zero GRIB file there is a 3-hour gap with parallel isochrones and global lack of wind during that time.
Post by iconPost by yklg | 2019-01-04 | 20:06:33
Thanks Cvetan, meteo is actually very difficult to understand, we do with its variability. Excuse for my english, I'm french... You do a good job with the router, it's very helpfull
Post by iconPost by zezo | 2019-01-04 | 20:31:29
A very simplified view of the weather (for virtual routing purposes) is that high-pressure areas are low-wind zones. They also tend to be stationary.

Low-pressure systems (depressions) tend to move around, West to East, the highs affect their movements.

There are two notorious high-pressure systems in the Atlantic - Azores high in the North and St. Helena high in the South (that's where Napoleon was in exile).

Their exact position and size affect the routing big time - the St. Helena decides if you can take a shortcut in the middle of the ocean or have to go all the way down to Rio along the coast to meet a low pressure system.
Post by iconPost by zezo | 2019-01-04 | 20:34:33
Then there are the Doldrums (ITCZ) with low wind, worse on the East side, and the trade winds, which make you pass next to Brazil when sailing from Europe to Cape town.
Post by iconPost by yklg | 2019-01-04 | 20:46:02
So I prefer to wait for a new departure on the TJV, perhaps at the end of next week... Good evening. Thanks for your explanations very usefull
Post by iconPost by zezo | 2019-01-04 | 20:57:29
Historically it's unlikely that there will be a better window after the New Year, but you never know.

There are 6-7 more weeks of opportunity, and even if you don't beat the record you can gain some experience by doing a round-the world race.

Personally, I've done enough of them (virtually) in the past 10 years and won't bother to start again unless there is a sub-6 day time to the Equator, combined with a less than about 15 days to the first gate.

Post by iconPost by yklg | 2019-01-05 | 04:16:51
I began my first attempt at 19th november 8h30 am, for a total time around 41d 19 it's not so bad...
I arrived at 31th december early in the morning and I've immediatly started a new attempt, my time to equator will be around 6d 18h and the first Cap less than 14d.
Përhaps, it will be difficult to make better time in next weeks, so I will go to the finish, for a second experience of complete round over the world, it's pleasant !
Thanks for your experience, mine is 2 years now, I began on the VG 2016
Post by iconPost by zezo | 2019-01-05 | 13:17:48
I started exactly 10 years ago, in the middle of the 2008-2009 VOR.

The router was written a month later and announced in April for the Rio-Boston leg.
Post by iconPost by yklg | 2019-01-05 | 16:29:27
You had a very good idea to create this router, congratulations, it's really good and very easy to use, for a debutant like me a few months ago !
Post by iconPost by etrille17 | 2019-01-05 | 16:34:52
Bonjour yklg.Sur la JV apres fernando, je programme 40S 10W(ile de Gough)et ce n'est pas mauvais;A+
Post by iconPost by yklg | 2019-01-05 | 18:06:22
Merci Etrille17, je passe l'équateur dans 24h, j'ai déjà choisi un point près du Cap Leewin, carrément et ça donne 19j au total, pas mal du tout !

Topics list
Copyright 2009 by ZEZO.ORG. All Rights Reserved.