Zezo forum:

Post by Moderation | 2011-12-07 | 04:03:18

Here is how I am assuming the winds in the VORG work. At 0700 CEST (that's 0200 AST - my time) a new batch of wind data is loaded into the game. This consists of winds we will be working/sailing in for the next 12 hours. Also loaded into the game is the predicted winds that will become the real winds we will work with when 1900 CEST arrives. Predicted winds for up to the next 3 or 4 days are also loaded by purely for reference and probably not completely accurate since they will be replaced at the next 0700 CEST wind update. Is this correct or do the games folks load a whole new set of winds at the 1900 CEST as well?

12 Comments

Post by zezo | 2011-12-07 | 08:29:05

The entire set of latest available forecasts is loaded every 12 hours.

Post by Moderation | 2011-12-07 | 12:48:59

Thanks Cvetan, I guess I will have to tell the boss to hold on a bit, it's 2 in the afternoon and I have to check on my boat! :-)

Oh! Another question arises from reading the "forum" here. How long before the change do you have those new winds available in the SailingSimulator?

Post by zezo | 2011-12-07 | 19:35:12

About hour and a half with partial update starting 2 hours before the game.

Wind barbs are rendered 1 hour before the wind change, but the optimizer uses the latest available NOAA data. Which may be misleading on a long leg - if you have 2 days from the new forecast and the other 5 from the old one.

Post by Moderation | 2011-12-08 | 02:19:43

Of most importance to me would be the winds my boat would be directly affected by when the update is applied to the game at wind change time. If I knew I could reasonably rely on "the plan" as developed in the SS then I would not fuss over not being able to get to the boat right at the wind change time. Something I'm sure my customers would appreciate :-) I can imagine their reaction should I tell them their equipment is going to have to wait a few minutes while I attend to some spurious data manipulations on my computer :-}

I don't think I have been caught with unusual winds during the afternoon/evening update as much as I have with the 0200 (0700cest) update. I now make a point of waking up at least 10 to 15 minutes before that one since I had been caught literally asleep at the wheel when something dramatically different showed up. The boat slogged along for a few hours at less than optimal trim (during the previous VORG).

Post by zezo | 2011-12-08 | 08:16:28

The forecast is 95% reliable 1 hour before the in-game update. The 5% uncertainty comes from the non-zero probability of some technical problems along the way - NOAA delaying the forecast data (happened once this leg) to zezo.org having internet/power problems or the game having some problem, but they are doing very well this time, especially compared to the last VORG where there was not a leg without serious issues.

But that 5% become a nightmare when you're fighting for position in the TopX. I was thinking about some kind of notification in such case - send a email (which could become email-to-sms at user side) or something. And pop a big red warning at the site if the forecasts don't match so you will notice it at 4AM or whatever time the update happens in your TZ. But then the low probability of the event shifts the feature down on the TODO list ;)

Post by pgfiore | 2011-12-13 | 00:37:03

This is really a innocent question: is the present wind (the one working) always replaced with the 12h-forecast one?

Or there is a risk the forecast "is wrong" (as in real life) and therefore the present wind differs from its previous forecast?

((p.s. sounds clear in English like in my mind... before translation? :P))

Post by zezo | 2011-12-13 | 15:14:02

I don't think that I can explain it better than in the previous post. The risk is NOAA delaying the forecast. Then the game could use something different as the next wind and I would have to change it to conform. This was happening relatively often with the french game, because they received the forecasts from a 3th party for a long time and it created additional delays and errors. Not sure about UG.

Post by pgfiore | 2011-12-13 | 16:53:59

Nope, sorry my English, I did talk about risk of errors...
The normal way, when everything works correctly.

Example: it's noon (whatever) at due coordinates
- I check forecast (the next, 19:00) and see that the future wind will be NE 10kts
- later, after 19, lets say 20:00, I control that wind

In real life forecast is a matter of luck. And nobody can bet how it will be in real.

Is in VORG the wind at 20:00 (when I check) always NE 10kts (the example) or could be SW 15kts (just cooked up) and a there is a meteoroligist deeply ashamed somewhere?

Post by pgfiore | 2011-12-13 | 18:45:54

errata corrige: "I did *NOT* talk about risk of errors"

Post by zezo | 2011-12-13 | 19:25:35

It won't be exactly NE 10 knots.

What it will be becomes known at 17h when NOAA publishes the GFS model used for the 19h change.

Post by pgfiore | 2011-12-13 | 23:25:20

ok. great. I misunderstood the above thread, sorry.

Post by pgfiore | 2011-12-28 | 21:57:59

sorry Cvetan/folks, I'm a bit slow...

- after winds are downloaded at 6am, SS shows the new forecast.

- But the "real" wind does not change; it only changes at 7am at the same time and with the same values of VORG.

- during 6am to 7am, the difference between SS and VORG are only the forecast

Am I right?