Blow (a reprise article from 2008)

by Vin Mariani

[Note: This article was originally posted January 25, 2008
by Vin Mariani on Although it is five years old,
the information in the article (and in the subsequent “Part II“)
still applies to many SL boats that have wind algorithms
based on Kanker Greenacre’s original Tako scripts. /JFos]

Several weeks ago a group of us were sitting around the table on the terrace of Nantucket Yacht Club, knocking back a few and talking about the wind. Jane Fossett asked, “do all boats in a given race feel the same variation of wind at the same time?” A simple question, yet a critical one, as the answer tells us whether winning races is a matter of randomness or of skill. We all want to believe it is skill, don’t we? But if I experience a different wind variation in a race from my competitors, I can always say, “you won because you happened to get better wind than me, this time.” But if we all experienced identical wind fluctuations during the race, then I must admit, “you made more skillful use of that wind than I did.” So Jane’s question is central to the meaning of sailboat racing in SL.  We had better hope that all racers experience the same wind variations, or winning a race would be pretty much just a roll of the dice.

We already know that wind variations affect hot lap times. The variation adds a random factor that can only be overcome by persistent repetition, or by setting the wind variations to zero. But, is the order of race results also affected by wind variation? The more we talked about the subject, the more questions we uncovered. Turns out, we don’t really know much about the wind at all. So I decided to study the subject intensively and see how much I could learn. This article presents the results of an investigation into how race wind works in detail. It’s surprising how much we CAN know about the wind. I’ll present the answers to all of the following questions. If you think you understand the wind already, you can consider this a quiz.


  1. Once a boat gets its wind parameters from a windsetter, how does it still know the “current wind” when it is far, far away from the windsetter?
  2. Jane’s question: in a race, do all boats of a type feel the exact same wind variation at the same time? (Are they lifted and headed together?)
  3. Even if they are in different sims?
  4. Do different types of sailboats in the same race feel the very same wind variation as each other? Or does the wind depend on the type of boat?
  5. Does the wind vary from place to place at a given time, the way that it does in RL?
  6. Does the wind change when you cross a sim boundary?
  7. Does the wind have any day/night variation, the way it does in RL?
  8. Wind speed and direction are seen to occasionally deviate by more than the +- parameter values; how much does it really vary?
  9. Can we know the entire wind variation in advance?
  10. Would knowing it affect racing or hot laps?


  1. Each boat that can use race wind carries two internal functions inside its scripts that can tell that boat the exact wind speed and direction whenever the boat “asks”.  These two functions trace back in time at least to Kanker Greenacre’s Tako version 2.1 scripts, (which can be found on the LSL Wiki, a repository for scripting knowledge). Here is Kanker’s original code fragment:

In case you find the scripting opaque, here are the functions themselves, pretty-printed for clarity: 

The specific outputs of these two functions depend on the five wind parameters, which are shown in red. Once those five parameters have been received by the boat, that boat’s wind functions will then proceed to vary only as a function of SL time, wherever the boat goes. At the default rate of 1.0, the variation repeats itself every 24 minutes, all day long. Below are plots of the variations over one cycle, using the particular wind parameters of the NYC race line (spd = 11, dir = 5, spd+- = 3, dir+- = 15, rate = 1.0). The basic curve shapes would always be the same at other race locations, although the axes scales would depend on the five parameters of the local windsetter.

Although these two plots “tell the whole story” I find it hard to relate speed and direction to one another. What I would like to see is both speed and direction on the same plot. Rather than try a 3-dimensional plot, which can be very hard to visualize, I made a polar plot of just direction and strength, where time is only implied. In the plot below, the instantaneous wind wanders slowly along the curve over time, leaving a blue dot every 15 seconds. The average wind of 11 m/s coming from 5° is marked as a big green dot in the center. Note that in SL, 0° is due East, not due North, as we might expect.

Here we can see that the wind wiggles around the average with many small deviations, and it has a few rare larger deviations, like real wind does. And the variations are complicated enough that you probably would never notice a pattern.  But shown like this, my mind sees a butterfly in the red Rorschach blot. Nice work, Kanker.

This is a good place to stop for a rest. We still have nine more questions to answer and, although the answers will be brief, this article is plenty long already.

To be continued. Questions are welcomed, but first check to see if I’ve already promised to answer that question in Part 2.  If your brain is full, go sailing.


14 Responses to “Blow”

  1.   # Harmony Bloch on 25 Jan 2008 at 9:05 pm

    Interesting information. Having the wind as a time function against time of day means all the boats should be changing at the same time becasue the clocks global (as I understand it).
    As for graphing…maybe take each maximum from both graphs, run a time plot against the time of each maximum against wind speed and comment each datapoint with the direction? (just a thought to make it a little more intuitive)


  2.   # Jane Fossett on 26 Jan 2008 at 9:14 am

    Harmony said: “… Having the wind as a time function against time of day means all the boats should be changing at the same time becasue the clocks global…”
    Yes! That’s the idea… but have you noticed there’s often a small ‘jerky’ shift in the wind when you cross a sim line and get a handoff from one server to the next? I’ve been peeking over Vin’s shoulder at this stuff, and it turns out not all clocks are exactly the same… but I’ll let him tell you the details!


  3.   # Jane Fossett on 26 Jan 2008 at 9:36 am

    I have to admit it was a lot of fun testing out Vin’s predictions on different boats.
    Here’s a picture where we hit the rocks in Santa Catalina on Vin’s BeachCat.
    Vin is explaining how he knew the exact wind velocity and directional deviance at the moment of impact.
    Apparently, he just forgot to turn…
    Vin contacting his Havok insurance agent


  4.   # MarkTwain White on 26 Jan 2008 at 7:46 pm

    In college one of my professors had great stories about Albert Einstein. My professor was his student assistant at Princeton.

    He had many many stories about how forgetful Einstein was. It was not at all unusual for my professor to get a phone call from Mrs. Einstein asking him to go to Albert’s office and tell him he is two hours overdue for dinner. Vin might be operationing from the same end of that particular bell curve.

  5.   # Vin Marianion 26 Jan 2008 at 8:05 pm

    In solidarity with my fellow columnist, I swear that Jane’s narrative of that crash is absolutely correct, except for two very tiny baldfaced inaccuracies. Honi soit qui mal y pense.
    (Sent from my iPhone, in the Florida Keys)

  6.   # Vin Marianion 26 Jan 2008 at 9:43 pm

    Oh, for those of you who don’t speak French, that translates as, “My lady, hit that guy who’s abusing himself.”

  7.   # jane fossetton 27 Jan 2008 at 8:30 am

    Vin said: “Jane’s narrative of that crash is absolutely correct, except for two very tiny baldfaced inaccuracies”
    OK, so maybe it wasn’t Santa Catalina…

  8.   # jane fossetton 28 Jan 2008 at 7:07 am

    Even a quick look at Vin’s graphs reveals a number of fascinating details. For example, at NYC the wind speed is 11m/s and the variance is +/-3m/s. However, I often end up with HUD reading of 15m/s while sailing. Vin’s graphs seem to explain that.
    Vin shows the wind speed has a complex oscillation that repeats every 24 minutes. The graph shows the wind speed actually overshoots 14m/s for about thirty seconds, twice in each cycle.
    Vin’s “Wind Direction” graph is even more interesting. The wind direction is 5 degrees at NYC, and the variance is +/- 15. Vin? It looks to me as though the Dir modulation actually repeats every 12 seconds (“A” below). Is that correct?
    It also looks like nearly all the variation occurs withing 5 degrees of baseline, with the exception of larger, one minute ‘spikes’ that swing between 5 and 10 degrees away from baseline (“B” below). The wind direction actually never varies over ten degrees, even though the windsetter is set for +/- 15.

    NYC's deviant direction

    This past week at Bea Woodget’s Sailing Skills Class, M1sha Dallin commented that while sailing the Olympic Course she could on rare occasion hit the top mark from the Santa Rosa Islands on a single close haul tack. I assume that happens because of the “directional deviance spikes” nicely documented by Vin.

  9.   # Liv Leigh on 28 Jan 2008 at 10:35 am

    Can M1sha maybe tell at what precise time of day that happened? :P Some of us will be very much interested by this information…

  10.   # Liv Leigh on 28 Jan 2008 at 11:04 am

    Knowing that wind is fixed for all eternity by the windsetter we will now soon be able to fabricate the next miracle in sl sailing engineering: the ‘autopilot-HUD’:

    By knowledge of current wind settings from the windsetter and exact location, rotation and speed of the boat, it will calculate the fastest way over the water for finishing the course. After this it will produce the needed chat commands to trim sails accordingly and navigate the boat in the right direction.

    An added advantage of this device will be that it will lag the area to such an extend that any competitors will think twice before entering a race with the HUD’s owner.

    Coming up soon!

  11.   # jane fossett on 28 Jan 2008 at 12:35 pm

    Liv said: “Can M1sha maybe tell at what precise time of day that happened? Some of us will be very much interested by this information…”
    I doubt M1sha remembers, but I’m sure Vin can tell you. Just wait till his next article… !

  12.   # Vin Marianion 29 Jan 2008 at 10:48 am

    Jane, you are correct that the direction variation repeats exactly every 12 minutes. Since the speed variance does not quite repeat until 24 minutes have passed, the whole wind behavior has a 24 minute cycle.

  13.   # Bea Woodgeton 29 Jan 2008 at 1:55 pm

    Don’t worry Liv, the Autopilot-HUD already exists…as a course HUD with buoys on minimap. Both working together, while racing with a fleet, you can even avoid other boats by steering sometimes, and have a drink, everything is computed again… Great…
    Is next step to publish “secret” part of sailing scripts?
    Now, having this information, have “hotlaps” windsetting a sense yet?
    But, I know i am going in the wrong direction, even the only boat without gesture will support them soon, majority is always right.
    I think i am going to focuse more on cruising than racing now, or something else than sailing. At least I’ll feel i can do something by myself, without calculating.

  14.   # Delinda Dyrssenon 14 Feb 2008 at 3:30 am

    Now I know….. why I have never won a race.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s