

26012013, 15:03

#451

Moderator
Join Date: Mar 2009
Location: The boat: Cowes (Winter), Above 60N (Summer); me: somewhere in the air!
Boat: CutterRigged Moody 54
Posts: 21,292

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by Seaworthy Lass
Example 5
CURRENT
1st hour: 3.6 knots 270 T (pushing you west)
2nd hour: 1.6 knots 270 T
3rd hour: slack
Boat speed 4 knots
Destination is 8.4 nm due south
All perfectly reasonable realistic figures
What is the CTS and time taken?
Try the RYA method and SWL method and work it out mathematically and see how results compare

Another real easy one which doesn't need to be plotted.
It's also cleverly ( ) designed to put RYA in a bad light  the last hour (slack) is way off the passage average (2.6) and you get a good chunk of it in this passage. The RYA doesn't analyze the last hour at all, and there are only two full hours, so the proportion between the unanalyzed partial hour and the whole passage is large. SWL, you are mean!!!
So the best solution will be done with math, and without any RYA or SWL plotting. We have more than two hours, closer to 2.5. But since the last hour is slack, we can relax  no need for any recursive passage time calculations. In any case we will have exactly 5.2 miles of displacement W, no more and no less even with varying passage time.
So course correction is exactly 31.7595 degrees, which by the way is already into the territory where our 1 degree per mile at 60 miles rule of thumb starts to go out of kilter. Water path is 9.8793 miles. Passage time is 2.469825 hours. This solution is mathematically perfect based on the assumptions, based on assuming that not only is the average rate of the last hour 0, but the average rate of the first half hour of the last hour is also 0.
Now, SWL. She will get the same results (although with a protractor she can't calculate it with such precision).
Now RYA. You will solve for the two full hours and then assume the passage average for the last partial hour. This will lead to a big error  about 1.3 miles of additional displacement W which will not actually happen, a big proportion of the real 5.2 miles.
So by inflating the whole hour vector triangle we get a course correction of 37.7330 and water track line of 10.6212 miles, for passage time of 2.6552, whoops, there will be a bit more displacement than 1.3 miles, but no need to screw around with this more . . . I think everyone gets the picture that this problem looks very bad indeed for RYA. Of course, an intelligent navigator would see the problem ahead and fudge for it, as Dave correctly said.
Now let's do it all again, with a slightly more sophisticated approach. Let's suppose that the rate at 0 minute of the 3rd hour is the average between 1.6 and 0, or 0.8, and varies linearly to 0 in the middle of the hour (very artificial, but better than assuming the tide suddenly drops from 1.6 to 0 in one second at the turn of the hour). It means the real average rate for the first half hour of the 3rd hour is really going to be 0.4, and not 0.
So with math we get 33.6901 as our course correction, and 10.0955 as a water track, so passage time of 2.523875.
This refinement won't change either SWL or RYA. SWL uses the average of the whole last hour, and RYA doesn't even analyze the last hour.
SWL gives 31.7595, about two degrees off, and RYA gives 37.7330, about four degrees off, in the other direction.
__________________



26012013, 15:31

#452

Registered User
Join Date: Mar 2012
Location: Nova Scotia
Boat: Wauquiez Centurion 42
Posts: 274

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by Dockhead
Another real easy one which doesn't need to be plotted.
It's also cleverly ( ) designed to put RYA in a bad light  the last hour (slack) is way off the passage average (2.6) and you get a good chunk of it in this passage. The RYA doesn't analyze the last hour at all, and there are only two full hours, so the proportion between the unanalyzed partial hour and the whole passage is large. SWL, you are mean!!!
So the best solution will be done with math, and without any RYA or SWL plotting. We have more than two hours, closer to 2.5. But since the last hour is slack, we can relax  no need for any recursive passage time calculations. In any case we will have exactly 5.2 miles of displacement W, no more and no less even with varying passage time.
So course correction is exactly 31.7595 degrees, which by the way is already into the territory where our 1 degree per mile at 60 miles rule of thumb starts to go out of kilter. Water path is 9.8793 miles. Passage time is 2.469825 hours. This solution is mathematically perfect based on the assumptions, based on assuming that not only is the average rate of the last hour 0, but the average rate of the first half hour of the last hour is also 0.
Now, SWL. She will get the same results (although with a protractor she can't calculate it with such precision).
Now RYA. You will solve for the two full hours and then assume the passage average for the last partial hour. This will lead to a big error  about 1.3 miles of additional displacement W which will not actually happen, a big proportion of the real 5.2 miles.
So by inflating the whole hour vector triangle we get a course correction of 37.7330 and water track line of 10.6212 miles, for passage time of 2.6552, whoops, there will be a bit more displacement than 1.3 miles, but no need to screw around with this more . . . I think everyone gets the picture that this problem looks very bad indeed for RYA. Of course, an intelligent navigator would see the problem ahead and fudge for it, as Dave correctly said.
Now let's do it all again, with a slightly more sophisticated approach. Let's suppose that the rate at 0 minute of the 3rd hour is the average between 1.6 and 0, or 0.8, and varies linearly to 0 in the middle of the hour (very artificial, but better than assuming the tide suddenly drops from 1.6 to 0 in one second at the turn of the hour). It means the real average rate for the first half hour of the 3rd hour is really going to be 0.4, and not 0.
So with math we get 33.6901 as our course correction, and 10.0955 as a water track, so passage time of 2.523875.
This refinement won't change either SWL or RYA. SWL uses the average of the whole last hour, and RYA doesn't even analyze the last hour.
SWL gives 31.7595, about two degrees off, and RYA gives 37.7330, about four degrees off, in the other direction.

She did put it almost exactly between the two RYA vectors.
Doing it "graphically" in CPN I came up with 148 degrees at 2 h 28 mins.
Doing my math quickly, SWL gives me a heading of 148.1535 degrees for 2.4709 hrs. Working backwards that would be a course correction of 31.8465 with my math. Effectively the same, especially at the helm without my glasses!
__________________



26012013, 15:47

#453

Moderator
Join Date: Oct 2008
Boat: Aluminium cutter rigged sloop
Posts: 12,937

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by Dockhead
Another real easy one which doesn't need to be plotted.
It's also cleverly ( ) designed to put RYA in a bad light  the last hour (slack) is way off the passage average (2.6) and you get a good chunk of it in this passage. The RYA doesn't analyze the last hour at all, and there are only two full hours, so the proportion between the unanalyzed partial hour and the whole passage is large. SWL, you are mean!!!

Yes . No angel here .
Quote:
Originally Posted by Dockhead
So the best solution will be done with math, and without any RYA or SWL plotting. We have more than two hours, closer to 2.5. But since the last hour is slack, we can relax  no need for any recursive passage time calculations. In any case we will have exactly 5.2 miles of displacement W, no more and no less even with varying passage time.
So course correction is exactly 31.7595 degrees, which by the way is already into the territory where our 1 degree per mile at 60 miles rule of thumb starts to go out of kilter. Water path is 9.8793 miles. Passage time is 2.469825 hours. This solution is mathematically perfect based on the assumptions, based on assuming that not only is the average rate of the last hour 0, but the average rate of the first half hour of the last hour is also 0.

Figures are correct.
CTS = 148.24 mathematically
I set a very easy example so that it can be quickly calculated mathematically. If the tide directions are oblique and variable, it is far quicker computing the CTS and time taken using the SWL method and if your pencil is super sharp and you are careful, you will be within a degree of the right answer
Quote:
Originally Posted by Dockhead
Now, SWL. She will get the same results (although with a protractor she can't calculate it with such precision).

Yes, Precision is not perfect, but I get:
SWL CTS = 148 degrees
Time taken = 2.5 hours
. IMHO this is a perfectly acceptable error . I will post the diagram tomorrow.
Quote:
Originally Posted by Dockhead
Now RYA. You will solve for the two full hours and then assume the passage average for the last partial hour. This will lead to a big error  about 1.3 miles of additional displacement W which will not actually happen, a big proportion of the real 5.2 mile.
So by inflating the whole hour vector triangle we get a course correction of 37.7330 and water track line of 10.6212 miles, for passage time of 2.6552, whoops, there will be a bit more displacement than 1.3 miles, but no need to screw around with this more . . . I think everyone gets the picture that this problem looks very bad indeed for RYA. Of course, an intelligent navigator would see the problem ahead and fudge for it, as Dave correctly said.

No.
The RYA would use a triangle with one side 5.2 nm and the hypotenuse 8 nm
which gives a CTS correction of 40.54 degrees
RYA CTS = 139 degrees (139.46 degrees rounded off)
SMG = AD / 2 = 6.08 / 2 = 3.04 knots
Time taken = AB / SMG = 8.4 / 3.04 = 2.76 hours
I can see nothing about fudging an example like this in the RYA notes .
Has their been an updated edition in the last few months to include this?
Quote:
Originally Posted by Dockhead
Now let's do it all again, with a slightly more sophisticated approach. Let's suppose that the rate at 0 minute of the 3rd hour is the average between 1.6 and 0, or 0.8, and varies linearly to 0 in the middle of the hour (very artificial, but better than assuming the tide suddenly drops from 1.6 to 0 in one second at the turn of the hour). It means the real average rate for the first half hour of the 3rd hour is really going to be 0.4, and not 0.
So with math we get 33.6901 as our course correction, and 10.0955 as a water track, so passage time of 2.523875.
This refinement won't change either SWL or RYA. SWL uses the average of the whole last hour, and RYA doesn't even analyze the last hour.
SWL gives 31.7595, about two degrees off, and RYA gives 37.7330, about four degrees off, in the other direction.

NO, RYA gave 41 degrees correction, so they would be 7 degrees off!
I think as you became more familiar with the SWL method you could look to see if the current was actually increasing or decreasing in the last hour and tweak the value a little according to the variation during the hour. That is adding a level of sophistication to the method that would come with frequent use and skill. I mentioned this in one of my earlier posts
__________________
"The cure for anything is salt water: sweat, tears or the sea." Isak Dinesen
"To me the simple act of tying a knot is an adventure in unlimited space." Clifford Ashley



26012013, 15:55

#454

Senior Cruiser
Join Date: Mar 2006
Location: Winnipeg
Boat: None at this time
Posts: 7,949

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Anyway, thanks all for teaching me that the rhumb line is not always the quickest way to get from point A to B. All the rest of the arguments are making my head hurt...



26012013, 16:03

#455

Moderator
Join Date: Oct 2008
Boat: Aluminium cutter rigged sloop
Posts: 12,937

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by DeepFrz
Anyway, thanks all for teaching me that the rhumb line is not always the quickest way to get from point A to B. All the rest of the arguments are making my head hurt...

DeepFrz, so glad the penny has dropped .
Makes discussions like this really worthwhile!
__________________
"The cure for anything is salt water: sweat, tears or the sea." Isak Dinesen
"To me the simple act of tying a knot is an adventure in unlimited space." Clifford Ashley



26012013, 16:09

#456

Senior Cruiser
Join Date: Mar 2006
Location: Winnipeg
Boat: None at this time
Posts: 7,949

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Oh, it dropped a few days ago. I have been following the rest of the thread with bemused interest.



26012013, 16:23

#457

Nearly an old salt
Join Date: Jun 2009
Posts: 13,649

Quote:
Originally Posted by Seaworthy Lass
Yes . No angel here .
Figures are correct.
CTS = 148.24 mathematically
I set a very easy example so that it can be quickly calculated mathematically. If the tide directions are oblique and variable, it is far quicker computing the CTS and time taken using the SWL method and if your pencil is super sharp and you are careful, you will be within a degree of the right answer
Yes, Precision is not perfect, but I get:
SWL CTS = 148 degrees
Time taken = 2.5 hours
. IMHO this is a perfectly acceptable error . I will post the diagram tomorrow.
No.
The RYA would use a triangle with one side 5.2 nm and the hypotenuse 8 nm
which gives a CTS correction of 40.54 degrees
RYA CTS = 139 degrees (139.46 degrees rounded off)
SMG = AD / 2 = 6.08 / 2 = 3.04 knots
Time taken = AB / SMG = 8.4 / 3.04 = 2.76 hours
I can see nothing about fudging an example like this in the RYA notes .
Has their been an updated edition in the last few months to include this?
NO, RYA gave 41 degrees correction, so they would be 7 degrees off!
I think as you became more familiar with the SWL method you could look to see if the current was actually increasing or decreasing in the last hour and tweak the value a little according to the variation during the hour. That is adding a level of sophistication to the method that would come with frequent use and skill. I mentioned this in one of my earlier posts

What notes would that be. The RYA YM theoretical course is 16 nights of approx 4 hours each night. Chart work in all its guises takes up about a quater of that . In the class room , good instructors analyse the various method including interpolation , tide accuracy. Situations where the model needs examination.
Seaworthy justify you methods on their own , stop trying to present half the RYA method. , as ive said before it ill becomes you to engage such a shrill campaign . The RYA method assumes certain conditions. It only looks at 30 minutes scenarios. In real life , unlike sleight of hand maths , the differences are minimal
Et etc , I'm a bit fed up repeating myself even Dockhead is chiding you.
Take some real tidal diamond data and then give us an example.
Dave
__________________
Check out my new blog on smart boat technology, networking and gadgets for the connected sailor!  http://smartboats.tumblr.com



26012013, 16:30

#458

Nearly an old salt
Join Date: Jun 2009
Posts: 13,649

Quote:
Originally Posted by Dockhead
Oddball numbers are really useful. There is a powerful logical principle called reductio ad absurdum which uses oddball cases to test principles which should be universal. How really universal they are can be very efficiently tested sometimes using extreme ranges of parameters.

I am well aware of what be colloquially called boundary data.
The issue here is comparing the RYA method which assumes tidal time boundaries are not abrupt and tides change slowly in respect of boat progress and SWLs method which takes a literal quantum approach to the tidal data
In reality its comparison with more real life data that's the issue not comparing a model with one assumption to a model with another.
as I said unless a method improves the total passage error circle to the extend that it exposes the other. It's of no additional value in real life. Both models merely provide answers that lie within the same circle of passage error.
Dave
__________________
Check out my new blog on smart boat technology, networking and gadgets for the connected sailor!  http://smartboats.tumblr.com



26012013, 17:59

#459

Registered User
Join Date: Dec 2010
Posts: 264

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Ok  painted a dagram
Hor Axis : Distance made on a hourly basis, marked as B(hours)
Vert Axis : Diff hourly tidal vectors
A>B(1.h) : standard RYA example
A>B(2.h) : dito but there comes issue SWL is adressing, but here I have to say that RYA YM tutorial explicitely says, the tidal vector can be less then "hourly" based if circumstances require ! So to play safe I just would add another hour to see what happens.
A>B(3.h) : oh dear, boat got into a heavy tidal stream and I realize I am not moving ahead. So here I would assume the great SWL method would also fail. Point is in such situation you need to look ahead.
A>B(4.h) : due to weaker tidal stream it's moving ahead, but watch for the huge offset  B(4.h) vs arcedoff point. We are 1.5h off B(4.h) so we have to add 1h to look head
A>B(5.h) : the tidal stream got much weaker, but still 1h off from B(5.h), so we need to add 1h.
Bottom line  you don't know how long it take unless the arced off point gets close to B. But this might be only reached by iteratively arcingoff to the destination point B(t)
May be you get my point  there is just no "perfect" method  SWL would fail also. Unless you have a passage routing software or a very good navigator, which does more then just apply a "method".
Btw I still think you do the CTS calculation on a hourly basis, especially with slow boats, to get at least "straigt" line. It's like with EP calculation, with adding up vectors, sometimes you sail "through" an island without noticing it ... :) ... or too late !



26012013, 18:29

#460

Registered User
Join Date: Dec 2010
Posts: 264

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Sry missed half of what I wanted to write :
Ok  if your target is B(3.h) then it would be indeed a bad idea just to extrapolate from the CTS(3.h). For a good navigator it is pretty obvious that 4.h and 5.h have to taken in account and to calc. the CTSes.
Then you would realize that to get to B(3.h) a CTS halway between CTS(4.h) and CTS(5.h) would be required.



26012013, 19:38

#461

Registered User
Join Date: Sep 2009
Posts: 888

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by bewitched
Consider the following: A to B is 90 deg and 5nm.
What is the fastest course to steer in each of these scenarios? (ignoring all variables other than tide)
1. Zero tide
2. A constant north south tide at 1 kt
3. A constant north south tide at 2 kts
The answer to 1. is obviously a CTS of 90 deg
The answer to 2. will be somewhere less than 90deg
And the answer to 3. will be somewhere less than 90deg and also less than answer 2.
So what does this very simple exercise tell us? Well it tells me that if the tidal flow rate varies, the CTS will have to change to remain optimal. Maintaining a single CTS, derived from an average tide state over the course, really is not the way to get there quickly.

Quote:
Originally Posted by Seaworthy Lass
Winning many races?

Yes thank you. 6 wins from 6 starts in our current race series
Quote:
Originally Posted by Seaworthy Lass
Suggest you review your strategy .

Oh, I don't do strategy....simple main trimmer me. My strategy doesn't go beyond the next few seconds. No, we have a paid professional tactician who does that for the more serious races, and he looks like he's doing fine.
But I can report that he has never used a single CTS to get anywhere quickly. Quite the opposite in fact. The tablet computer that is strapped to his left hand for most of the race runs very sophisticated mathematical models and these require very accurate measured and predicted data. The software we use is Expedition.
For an example of its output, take a look at the 'positions' section of the Vende Globe site to get an idea of how their ground tracks look. They to will be using routing software and it's obvious they are not utilising a single CTS.
I don't have an opinion either way as to whether the RYA or your method is more accurate, but I think a reality check is necessary. They both may be useful outline passage planning tools and they may be useful as a technique to get you home if for some reason you loose the ability to accurately determine your position regularly, but presenting them as a route optimisation technique is way off the mark. They don't even scratch at the edge of the subject.



26012013, 22:15

#462

Moderator
Join Date: Mar 2009
Location: The boat: Cowes (Winter), Above 60N (Summer); me: somewhere in the air!
Boat: CutterRigged Moody 54
Posts: 21,292

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by Seaworthy Lass
I think as you became more familiar with the SWL method you could look to see if the current was actually increasing or decreasing in the last hour and tweak the value a little according to the variation during the hour. That is adding a level of sophistication to the method that would come with frequent use and skill. I mentioned this in one of my earlier posts

Why "become more familiar"? I have completely absorbed it; can do it blindfolded in my head.
Of course you can fudge it  every decent navigator fudges everything which is approximated. It's an essential tool.
But I think it's hardly needed  your method is close enough in any but really extreme cases, I think (need to check it with high resolution model). That's a key advantage.



26012013, 22:29

#463

Moderator
Join Date: Mar 2009
Location: The boat: Cowes (Winter), Above 60N (Summer); me: somewhere in the air!
Boat: CutterRigged Moody 54
Posts: 21,292

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by goboatingnow
I am well aware of what be colloquially called boundary data.
The issue here is comparing the RYA method which assumes tidal time boundaries are not abrupt and tides change slowly in respect of boat progress and SWLs method which takes a literal quantum approach to the tidal data
In reality its comparison with more real life data that's the issue not comparing a model with one assumption to a model with another.
as I said unless a method improves the total passage error circle to the extend that it exposes the other. It's of no additional value in real life. Both models merely provide answers that lie within the same circle of passage error.
Dave

Cutting through the emotion, I think Dave has formulated the question correctly here.
SWL has given us a method with higher level of precision at no extra cost of effort. That has to be good; but to the extent the additional precision is actually usable is, as we have discussed, an empirical question.



26012013, 23:55

#464

Moderator
Join Date: Oct 2008
Boat: Aluminium cutter rigged sloop
Posts: 12,937

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by bewitched
I do a lot of racing too and at quite a high level too. If it was simply a case of calculating a single course to steer to get the fastest passage, racing would be really boring.
Consider the following: A to B is 90 deg and 5nm.
What is the fastest course to steer in each of these scenarios? (ignoring all variables other than tide)
1. Zero tide
2. A constant north south tide at 1 kt
3. A constant north south tide at 2 kts
The answer to 1. is obviously a CTS of 90 deg
The answer to 2. will be somewhere less than 90deg
And the answer to 3. will be somewhere less than 90deg and also less than answer 2.
So what does this very simple exercise tell us? Well it tells me that if the tidal flow rate varies, the CTS will have to change to remain optimal. Maintaining a single CTS, derived from an average tide state over the course, really is not the way to get there quickly.

Bewitched, are you convinced yet that the quickest way to sail a leg with variable cross current is by starting with a single computed CTS (if water depth is constant, so advantage can't be taken of less or more current in certain areas and you don't need to tack or gybe) and then thinking about leeway, wind in certain areas, etc to determine the best route?
This initial CTS determination will be the quickest way of getting from A to B if you are motoring in calm water with variable cross current. The quickest way is NOT following a straight track on the GPS!
Quote:
Originally Posted by bewitched
Yes thank you. 6 wins from 6 starts in our current race series

Well done! It is such a high winning isn't it?
Quote:
Originally Posted by bewitched
Oh, I don't do strategy....simple main trimmer me. My strategy doesn't go beyond the next few seconds. No, we have a paid professional tactician who does that for the more serious races, and he looks like he's doing fine.

I was always foredeck hand when racing (apart from helming half the night on one Bass Strait crossing when all but one other crew member was severely, severely sea sick LOL) because I happened to sail with all male crews and I was by far the lightest on board. You don't want weight going forward in a race!
I even got sent up the mast once during one of the 'Ocean Racing Club of Victoria' races! I was a valued crew member for my weight, lack of fear of heights, and sense of humour if nothing else .
As crews I sailed with were small (max 6) and no computers were used and I sailed in tidal areas, I picked up a lot as the tactician would always explain his strategy. I was lucky enough to sail with one good one (the passionate windsurfer whose holiday home on Phillip Island was in an area subject to large amounts of current in some spots), who was very aware of how to handle current and I learned a lot from him.
Quote:
Originally Posted by bewitched
But I can report that he has never used a single CTS to get anywhere quickly. Quite the opposite in fact. The tablet computer that is strapped to his left hand for most of the race runs very sophisticated mathematical models and these require very accurate measured and predicted data.

For longer races where one leg may be more than a couple of hours and you have variable cross current, the CTS definitely should be one input in route determination.
Quote:
Originally Posted by bewitched
The software we use is Expedition.

I had a look at your link to Expedition. I noted this tactical/navigation software was first offered in 1995. In 1996 they seem to have added tide and current data. Oops! Did they forget about it initially?
I bet the program inputs the mathematically calculated CTS if there is variable cross current during one leg of a long passage and then works on that in determining what route to follow .
Next time you are racing, could you please ask your tactician (when he has some free time) to input several currents for a long leg of a race in deep water (eg greater than two hours) and see what the program comes up with? Also ask him if the program tells you what tack to be on if the wind is on the nose? One of the two tacks will be advantageous. I would be very interested in the results, as I am sure you would be
If the Expedition software doesn't tell you what tack to be on with a headwind and variable cross current over the leg, then this would be one way the software could be improved. I wonder if Expedition would be grateful enough to me for suggesting this improvement, that they may send me a case of fine single malt Scotch? I can't source any easily on the Greek islands although I am still hopeful of getting one from the RYA. I'm nothing but an optimist here LOL.
Note: Computers don't tell you what wind is ahead, but your eyeballs can. Watching the pattern of wind on the water ahead is absolutely critical in a race and software is useless in this instance to tell you what to do .
Quote:
Originally Posted by bewitched
For an example of its output, take a look at the 'positions' section of the Vende Globe site to get an idea of how their ground tracks look. They to will be using routing software and it's obvious they are not utilising a single CTS.

They will certainly be utilising a CTS computation somewhere in the software and this becomes part of the route planning
Quote:
Originally Posted by bewitched
I don't have an opinion either way as to whether the RYA or your method is more accurate, but I think a reality check is necessary. They both may be useful outline passage planning tools and they may be useful as a technique to get you home if for some reason you loose the ability to accurately determine your position regularly, but presenting them as a route optimisation technique is way off the mark. They don't even scratch at the edge of the subject.

Route optimisation is a complex determination. Only one part of it is computing a CTS if you have variable cross current during one leg of a journey. I am attempting to do nothing more with my method than to find the optimum CTS if you don't have a computer program to do this. The rest if the route planning must of course proceed as usual.
CTS is just one if the inputs, but surely you would want a reasonably accurate means of determining it? In the last example the RYA method was seven degrees out if you had data for the current every five minutes of the last hour of a perfectly realistic journey (see post # 453 for a correction to Dockhead's calculation) That is an error of over 1nm for a destination that is only 8.4 nm away!!! This is an unacceptable error for me and I would not use the RYA method. I have a more accurate one.
It is always satisfying learning something new (or teaching someone who is enthusiastic and willing to learn). I get a kick from either .
__________________
"The cure for anything is salt water: sweat, tears or the sea." Isak Dinesen
"To me the simple act of tying a knot is an adventure in unlimited space." Clifford Ashley



26012013, 23:55

#465

Moderator
Join Date: Sep 2006
Posts: 4,220

Re: Inaccurate RYA Teaching : CTS  Quest For a New Method
Quote:
Originally Posted by bewitched
No, we have a paid professional tactician who does that for the more serious races, and he looks like he's doing fine.
But I can report that he has never used a single CTS to get anywhere quickly. Quite the opposite in fact. The tablet computer that is strapped to his left hand for most of the race runs very sophisticated mathematical models and these require very accurate measured and predicted data. The software we use is Expedition.
For an example of its output, take a look at the 'positions' section of the Vende Globe site to get an idea of how their ground tracks look. They to will be using routing software and it's obvious they are not utilising a single CTS.

You are talking about a completely different type of routing. Expedition (and other routing programs) uses predicted wind (GRIBs) and current data (mathematical models), and the boat's sailing polars, to calculate a course. The wind data is relatively finegrained, often using points on a 0.2degree grid, and very frequent simulationstep intervals. The current data depends on the models being used. The main differences are:
1) You're sailing.
2) The wind is constantly changing, and is certainly not uniform across the playing field.
3) The current's are nonuniform as well.
4) Sometimes you are covering other boats (racing tactics) although this isn't something that Expedition is going to drive.
Of course you're not going to get a constant CTS!
What SWL and the others are doing is solving a *much* simpler problem. It's been simplified so that mere humans can quickly solve it given the basic minimum amount of data. The goal is to get "close enough", with the understanding that it is the navigator's responsibility to know when these simplifications are appropriate (and when they're not).
Expedition (I have it and use it to help route me on ocean races), performs literally thousands of short projections (and millions of calculations) to arrive at it's best guess. And "garbage in, garbage out" still applies. If I do just a minor tweak on my polar performance (because of crew fatigue, gear breakage, or sea state) then the predicted fastest route can change wildly. It's actually a pretty chaotic system when you add in the wind patterns.
__________________
__________________
Paul Elliott, S/V VALIS  Pacific Seacraft 44 #16  Friday Harbor, WA
www.sailvalis.com





Thread Tools 

Display Modes 
Rate This Thread 
Linear Mode


Posting Rules

You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off




