74 replies to this topic

[Estar]

The racers started on the 25th, but the 'rally/cruisers' start was delayed until today (27th) due to weather.

There looks to be one essential weather feature that will dominate the routing decisions. There is forecast to be a low developing right in the middle of the course and then drifting north. The fast boats will want to stay north and go over the low to keep strong favorable winds. The slow boats will not be able to make it over the top and thus will want to go south to stay clear of the headwinds and muddled winds below it.

Here is the fast boat course (on 12/3)

 1203.jpg   275.4K   96 downloads

And the slow boat course (on 12/5). You can see the low has now drifted quite north and a weak high/muddled region has filled in its path - need to keep south of all this

 1205.jpg   249.95K   94 downloads

Note: a cruiser will want to go more downwind at the start than these routes show. Routing programs hate downwind, as it is 'slow', but for cruisers in a strong breeze (as there is right now) its way 'fast enough' and easier and more comfortable. For the 'real' cruiser, the classic 'southwest until the butter melts and then due west' looks like it will work very well this year.

[Ajax]

So a program chose these routes, or you did? I really need to download and try these out this week.

What do you define as "fast" for the fast boats? Pushing 10 knots? Less?

[Soñadora]

This is a pretty good video of why faster is better. Kind of cheesy, but after watching these (and later his videos aboard a Westsail 42), I give him props.

[Estar]

So a program chose these routes, or you did? I really need to download and try these out this week.

What do you define as "fast" for the fast boats? Pushing 10 knots? Less?

Ajax - For definition: 'route' = manually setting up any old waypoints and courses between them; 'routing' = creating optimal route based on weather and current.

To do 'routing' most of us use programs, that create 'optimal routes', with human input/adjustment. You can make the adjustments by #1 playing with the polars to prefer one sort (speed/direction) of wind over others, or #2 play with the grib file (if you don't think its correct, or want to block off or make more attractive some areas), or #3 use features in the programs to set max wind or waves or exclusion areas or marks to be left to port or starboard, etc.

It is helpful to understand what the 'normal' best routes were along these passages in the days before everyone had sat weather on board. That is the 'default' route and you need to see a good reason in the gribs that you really believe will happen to go very far off that default. In Evenstar's case, the default route is to go SE out to 65w/28N (or there abouts) and then south to the BVI. In the ARC the default is SW until about the latitude of the Cape Verde and then due east. And in the south atlantic Vendee the default is to leave Fernando de Noronha and Gough island medium close to starboard.

In this case, all I did is play with the polars. The fast line represents 100% of Hawk's polars, while the slow line represents 60%. We actually cruise at about 85%.

And you were precisely correct that 100% of the polars is pushing 10kts in 20kts of breeze at 120 degrees.

I normally use some of the expensive racing routing programs, but I am having a lot of fun learning this free one and it is quite good.

The main thing to realize is that after 3 days the grib accuracy is low and after 5 days is about nil . . . but to do an effective routing you need to look out at a long way (at least 5 days) to be sure you don't run yourself into a cul de sac. And you have to make sure you are not going out of the way to chase after some wind in the model that does not actually appear in the real world (one of the vendee boats appears to have made exactly this mistake). Some experience and judgement comparing model output with the real world is needed to handle both of those - one of the reasons I like to keep my hand in looking at weather when I am 'on the beach' like this winter. But you need actual feedback from the boats out there to make it worthwhile (I have friends on two ARC boats), otherwise you just look at the model output in isolation and start believing they are correct.

[us7070]

I've done the ARC

My bet is that not many boats will want to sail DDW right out of the gate - it's rolly and uncomfortable.

I would guess that a lot of boats will head SSW on port tack, maybe to 20N, or even further S before heading W

[us7070]

here is the expedition routing for a generic 45ft IRC racer cruiser

Attached Files

•  arc.JPG   253.77K   71 downloads

[Joli]

That matches or routing using max sea and a GFS forecast.

here is the expedition routing for a generic 45ft IRC racer cruiser

oz

[Estar]

Yes, thats consistent with my 'slow' boat route.

You may not realize that most of the fast boats left 2 day before (the 25th not the 27th).

Today it looks like that low is going to develop/form further north and not create as much blockage underneath it. But there are 2nd and 3rd lows shown developing (off Venezuela) and spinning NE. Too far away in time for forecast accuracy but I would keep an eye on them.

12/05:
 car.jpg   288.53K   10 downloads

[Presuming Ed]

FWIW, fleet tracking is here: http://www.worldcrui...leetviewer.aspx

[Ajax]

Estar-

Great explanation, thank you.

[Estar]

From the position of a mid-fleet vessel (Puddle jumper), and looking at weather and routing from there.

Excellent sailing, stongish winds behind the beam, all the way until 12/07.

The prior forecast low, is still expected to materialize, but now centered further north and not causing quite as much disruption to the winds across the route. You can see the low here (middle upper edge) on 12/05
 1205.jpg   90.43K   33 downloads

But the fleet will probably see an area of light and mixed direction winds starting 12/07.
 1207.jpg   90.68K   41 downloads

That will continue thru 12/09, when a high pressure system pushes across, way north of the track, but pushing nice NE winds down along the track.
 1209.jpg   92.77K   28 downloads

[MR.CLEAN]

Good info Evans; thanks much. Will help me track a couple of pals and give me something to rib them with when i call.

[us7070]

I enjoyed the ARC when I did it, and I'm pretty sure I'm doing it again next year.

Hanging in Las Palmas before the start is fun, because there is a generally high level of excitement and anticipation amongst the participants.

The "official" parties have a commercial feel about them, and are somewhat lame, but there is plenty of unofficial socializing going on.

Once you are on the ocean, it's pretty much like any other crossing.

I didn't hang out in Rodney Bay much afterwards, instead I rented a car and spent a week or so exploring, as I hadn't been to St Lucia before. Rodney Bay is touristy, and not so nice, but elsewhere it's quite beautiful, and worth seeing. There is good kiting and windsurfing at the southern end of the island, near the bigger airport. That week or two before christmas is pretty quiet, and you can just go where you want, not worrying about reservations.

[Soñadora]

cool stuff Evans. Forgive me if I sound like a total noob, but I'm assuming the ARC is starting east and heading west.

btw, if anyone's wondering why I posted those videos, he talked about using Herb for weather routing. There was a low approaching and Herb wanted them to get west from where they were to miss it. Herb started to get concerned that their Wetsnail wasn't moving fast enough. At about 5:12 in the second video, you can see that they didn't get out of the way and had to deal with some crap conditions.

[Estar]

I'm assuming the ARC is starting east and heading west.
Correct. Usually a great downwind passage, but can be a bit rolly.

Wetsnail wasn't moving fast enough.
That's one of the main challenges with routing cruising boats - they are not fast enough to get out of the way of approaching systems - and you have to look way out into the future (when the forecasts are very inaccurate) to try to determine where you should be and what to avoid. You have to develop the skill of reading 7 day away weather forecasts and separate the signal for the noise.

[olaf hart]

So, what sort of speed do you need to keep ahead of approaching systems?
I realise this is a very general question, but what sort of speed would you want a boat to have in this situation?

[rattus32]

Evans, I have a question - how do you close (finish) the routes using Virtual Wind/qtVlm?

For example, if I do a "straight" (i.e. unobstructed) route from, say, Newport to Bermuda, download the gribs and choose routing directly between the endpoints it works great, producing "optimized" routes based on grib and polar inputs.

If there's land in the way, I need to set waypoints (this is annoying - in semiconductor routing, automatically working around obstructions is a matter of course). I create waypoints by placing them and adding them to a Route.. which is fine until I need to close the route, by connecting the final waypoint to the final mark - which I haven't figured out and the 333 page forum is overtaxing my conversational French.

If I add the final mark to the Route, all kinds of strange virtual routes emanate from each of the waypoints to the final mark ?! If I take it off the Route, despite naming it the endpoint in routing, no closure.

I would be delighted to post a tutorial on setting this up once the details are straight n my mind. Got Tips?

Thanks,

Mike

[Estar]

So, what sort of speed do you need to keep ahead of approaching systems?
I realise this is a very general question, but what sort of speed would you want a boat to have in this situation?

Too fast for pretty much any cruiser. A very typical weather system speed is 12kts = 288nm/day. So you really need the ability to do 300nm /day.

Evans, I have a question - how do you close (finish) the routes using Virtual Wind/qtVlm?

If there's land in the way, I need to set waypoints (this is annoying - in semiconductor routing, automatically working around obstructions is a matter of course). I create waypoints by placing them and adding them to a Route.. which is fine until I need to close the route, by connecting the final waypoint to the final mark - which I haven't figured out and the 333 page forum is overtaxing my conversational French.

mmmmm . . . sounds like you are trying to create a route first. I am doing it the otherway around . . . I set a starting and finish point and ask it to create a routing (Which btw it will automatically avoid land if you click the check box) . . . and then convert that to a route. So, no, I have not tried to set up the route first. There is an english help wiki. It is not very complete but might answer your question.

[rattus32]

Evans, I have a question - how do you close (finish) the routes using Virtual Wind/qtVlm?

If there's land in the way, I need to set waypoints (this is annoying - in semiconductor routing, automatically working around obstructions is a matter of course). I create waypoints by placing them and adding them to a Route.. which is fine until I need to close the route, by connecting the final waypoint to the final mark - which I haven't figured out and the 333 page forum is overtaxing my conversational French.

mmmmm . . . sounds like you are trying to create a route first. I am doing it the otherway around . . . I set a starting and finish point and ask it to create a routing (Which btw it will automatically avoid land if you click the check box) . . . and then convert that to a route. So, no, I have not tried to set up the route first. There is an english help wiki. It is not very complete but might answer your question.

I've tried that (end-end), and for some paths it works fine. When I try to produce a routing between, say, New Haven CT and Portsmouth NH, I get a message "Sorry, impossible to reach WP Try with another grib". This is with an up-to-the-minute grib file for the region from SailDocs. When I do the same for New Haven to Bermuda, no problem. Both direct paths cross land. The Bermuda route shows route trees being pruned as it maneuvers around the end of Long Island. Same grib, same boat starting point, different destination. The English wiki is useless. I also can't find a logfile to try to see what's wrong... guess I'll have to wade into the French forum!

Yeah, I need to HTFU.

[olaf hart]

Interesting about the 12 knot average...
That is the max cruising speed the Dashews use for their power boats.
Makes sense now.

[Estar]

Good sailing in the ARC for the next few days. Hope they enjoy it because . . .

Then an area of light winds to duck south of on 12/4

12/04  arc1204.jpg   109.2K   6 downloads

And a low pressure system to also duck on 12/5 - probably get about 12hrs of headwinds out of this unless you go a long way south.

12/05  arc1205.jpg   112.59K   8 downloads

Then a big area of lightish winds in an unorganized low pressure area - not much way to duck this.

12/07  arc1207.jpg   110.95K   6 downloads

[Joli]

Anyone else seeing this? The routing is bouncing around like a beach ball.

I'm using the morning position of an Oyster 62 currently sailing the ARC. They probably have polars similar to ours and that I'm basing the routing on.

11/27/12




11/28/12



11/29/12



11/30/12

[ice9a]

Anyone else seeing this? The routing is bouncing around like a beach ball.

I'm using the morning position of an Oyster 62 currently sailing the ARC. They probably have polars similar to ours and that I'm basing the routing on.

11/27/12
11/28/12
11/29/12
11/30/12

There are a couple big feature the week of 12/05 that the gfs has been moving around a lot. For a cruiser, I would be going for a southern routing. Thats going to be safe and comfortable in all the scenarios For a racer it's a tricky one....but I would right now be going down the GC track and looking for the forecasts to stabilize before making any big deviations.

[Joli]

todays....... I wouldn't venture far from rum

[us7070]

Joli - which program is that?

[Estar]

Joli - which program is that?

looks like maxsea.

My friends are both further back in the pack, and slower, than the boat joli is looking at and 'the answer' is different, and starting to be more like what I would have expected as the 'standard cruising solution. There has been a consistent forecast for some sort of mess along the direct route about 12-06, but the details have moved all over the place.

12-06

 1206.jpg   390.85K   5 downloads

This shows the wind field (wind strength when the boat actually gets to that point). You can see the big dead spot (dark blue) in the middle. This routing just ducks under the lightist bit.

 wind.jpg   258.24K   4 downloads

[Beau.Vrolyk]

Joli,

In listening in on this discussion I came to a conclusion very similar to yours - stay near the Rum line. At the time of the early forecast, I must say I was quite surprised that the routes weren't attempting to get south to the center of the normal trade-wind band as soon as reasonably possible. I suppose it's always a percentage thing, and perhaps the routing software doesn't have the concept of "when in doubt go to where there is usually more wind". But this routing puzzle reminds me a bit of what we face going from the mainland to Hawaii. There's a rum line and then there's the area that usually has more wind. You bias your thinking towards the historically windier path.

All of that is by way of a question: does the tool you're using have a "usually more wind" sort of a bias? Or, does it simply use the polars and GFS?

BV

[Estar]

All of that is by way of a question: does the tool you're using have a "usually more wind" sort of a bias? Or, does it simply use the polars and GFS?

I have used all three of the major packages (adrena*, expedition and maxsea) and none had a specific 'error toward the pilot chart route' feature. They all use basically the same "optimal isochrone" calculation, which is driven by the gribs (Wind and current and waves - from any source not just gfs). But they do all give you manual ways to look at weather risk and to pull the route toward where you think it should be. And you can do a routing just using a grib of the pilot chart winds, and you can also run the routing with the actual winds for the same time frame each of the past 10 years (Steve Dashew, when he was the maxsea distributor, had a cd with both those data on it) as two ways to see the 'statistical routes'.

In the end you can spend 'too much' time on this and the weather will still surprise you.

*just as an aside, I believe Adrena is now the 'premier' offshore routing program. It's used by the french from the figaro thru mini to vendee. Its been slow to catch on in the English world (but has a fine English language version) and my impression is that the developers don't think we English speakers represent a significant enough market. to bother with.

[us7070]

I've played a bit with Adrena - if you buy all the modules, it is substantially more expensive than Expedition - I think 2 or 3 times as much.

I didn't see much to make me think there was any reason to spend the money.

As someone who mostly navigates on American-owned boats, there is an absolute expectation that I will be familiar with Expedition, so I own a copy. Mostly, the boats I sail on have a dedicated PC running Expedition. I've never been on a boat running Adrena, so there is no real need for me to buy it.

Route optimization is only one of many great features in Expedition, an not the one I use the most.

[Joli]

I'm looking at max sea. I've always believed distance races are won on VMC more then any other factor. It stuns me to see boats sailing perpendicular to the course. VMC = 0

[Estar]

I've played a bit with Adrena - if you buy all the modules, it is substantially more expensive than Expedition - I think 2 or 3 times as much.

Hard to know exactly what is apples and oranges, but I would have said the comparison was: "Adrena offshore package" = 1511 euro= $1964, Expedition = $US1295. So that would suggest Adrena is 1.5x. However, there are versions of Adrena that do offshore routing that are half expedition's price. It seems that generally Adrena has a somewhat broader mission than Expedition, just for instance to include some AIS functions , radar overlays and 'dual speed' (useful for delivery's and cruisers). So, it would seem to depend on what your purpose/needs are.

Personally the thing I like most about Expedition is its clean user interface. Both Adrena and maxsea have more cluttered interfaces

It would be fascinating for someone to do a direct comparison of the core routing engines.

Route optimization is only one of many great features in Expedition, an not the one I use the most.

What do you use most?

[us7070]

I've played a bit with Adrena - if you buy all the modules, it is substantially more expensive than Expedition - I think 2 or 3 times as much.

Hard to know exactly what is apples and oranges, but I would have said the comparison was: "Adrena offshore package" = 1511 euro= $1964, Expedition = $US1295. So that would suggest Adrena is 1.5x. However, there are versions of Adrena that do offshore routing that are half expedition's price. It seems that generally Adrena has a somewhat broader mission than Expedition, just for instance to include some AIS functions , radar overlays and 'dual speed' (useful for delivery's and cruisers). So, it would seem to depend on what your purpose/needs are.

Personally the thing I like most about Expedition is its clean user interface. Both Adrena and maxsea have more cluttered interfaces

It would be fascinating for someone to do a direct comparison of the core routing engines.

Route optimization is only one of many great features in Expedition, an not the one I use the most.

What do you use most?

A brief glance at the Adrena web site shows that the Expedition has quite a few of the features in the Pro version of Adrena

Expedition has AIS and radar display.

pre-start..., tide/current in layline calculations...

I do a lot more round-the-cans than distance racing, so things like the start and race functions get used more.

As I mentioned, I've been using the polar editing functions a lot recently.

I think Expedition is the best GRIB viewer I have seen, and I use it to the exclusion of all others - but I haven't really explored Adrena's GRIB capabilities. I would like to see Nick add CAPE, and I believe it is coming.

Does Adrena dislplay Lambert Conformable GRIB's? I couldn't tell.

Expedition has pretty good networking capabilities - it sends things like time-to-layline, and many other numbers, calculated by expedition, to pretty much any racing instrument system, so they can be displayed on deck. I would guess that Adrena can do this too.

one thing I perpetually have trouble with in expedition, is getting the "Sched" function to read position reports and display other boats during a distance race. sometimes it works, but sometimes i can't get it, and give up. I can do it manually if i really care, but just looking at the Adrena screen shots made it look a bit easier in this regard.

[Joli]

Here's today. As Estar mentioned the problem is 12-5 and 12-6, there is a no wind zone 600 miles from the finish that is 500 miles across. If you're racing I guess you pick a side and go for it, the program can't decide, as shown by the day to day history. Cruising? Well, plenty of battery charging can be had. : ) We be cruisin mon.

[Estar]

Here's today. As Estar mentioned the problem is 12-5 and 12-6, there is a no wind zone 600 miles from the finish that is 500 miles across. If you're racing I guess you pick a side and go for it, the program can't decide, as shown by the day to day history. Cruising? Well, plenty of battery charging can be had. : ) We be cruisin mon.

This is a fascinating case example - a well predicted barrier of light and turblent winds but where the forecast can't settle on exactly what/where the winds will be, so the routing programs can't decide where to go - quite like a doldums crossing.

Seems like some sort of analytical process for this sort of barrier might be useful in the routing programs. It would (probably) need to be analytically different from the normal isochrone method. I am not sure what it should be - I look at barrier thickness, cloud cover, and cape when figuring doldrum crossings. With the doldrums you pick a point "X" as the best looking crossing point, and then do the isochrone runs before and after X to figure out how best to get there.

[dinghydoc]

Very interesting thread for me. I'm no stranger to the software biz but I did not expect to see Expedition costing $1,295. I guess I won't be learning to use that one anytime soon. My boat probably doesn't have the right compliment of instrumentation anyway. MaxSea is less of an investment but still $700 with the routing module. These will have to go on the big ticket list. I've been playing with qtVlm but it has a very crude interface (it's Open Source so low expectations). Still very cool to see how these tools are used and get a sense of their limitations.

[Estar]

Very interesting thread for me. I'm no stranger to the software biz but I did not expect to see Expedition costing $1,295. I guess I won't be learning to use that one anytime soon. My boat probably doesn't have the right compliment of instrumentation anyway. MaxSea is less of an investment but still $700 with the routing module. These will have to go on the big ticket list. I've been playing with qtVlm but it has a very crude interface (it's Open Source so low expectations). Still very cool to see how these tools are used and get a sense of their limitations.

You don't need to interface to instruments to do routing, but you do to use many of the other features of these program. About the only thing other than routing that I actually use are the laylines, to tell me when I can tack/gybe and be clear of a feature. For our ocasionally start line I just use the count down timer on my timex and some reference waypoints on the plotter with set distances from the line. 7070 obviously uses much more of the program.

Historically the market has been tiny and not very price sensitive (pro offshore navs and a tiny fraction of the blue water cruisers like us and steve dashew). However, the new virtual sailing games may have changed that . . . but the needs are somewhat different there than for 'real' passages.

I find qtVlm to be surprisingly good for free! It does pretty much everything a blue water cruiser would want.

[us7070]

I should say, my use of Exp in round the cans is pretty casual - it's not like I'm sitting on deck with a tablet..., like some kind of pro

we always turn it on, ping the start, and enter the course as indicated on the RC course board.

But how much we use it after that varies - some times we use time to layline, but sometimes not.

I really like the TWD stripchart, and have been displaying that on my iphone with a VNC app

targets are done through the instrument processor, not Exp, although all the polars on which they are based were made/refined in Exp

[Joli]

one of the guys in sa developed an app for cell phones for starts. you ping the pin and boat, it tells you how far, favored end, time to the line and other features. seems pretty cool for $5 bucks. i think the app is startline. ill get it before summer.

[dinghydoc]

one of the guys in sa developed an app for cell phones for starts. you ping the pin and boat, it tells you how far, favored end, time to the line and other features. seems pretty cool for $5 bucks. i think the app is startline. ill get it before summer.

Yep, Startline, an Android app in Google Play. I like it. I'm going to try it as well after I make the switch to an Android phone.

[dinghydoc]

Very interesting thread for me. I'm no stranger to the software biz but I did not expect to see Expedition costing $1,295. I guess I won't be learning to use that one anytime soon. My boat probably doesn't have the right compliment of instrumentation anyway. MaxSea is less of an investment but still $700 with the routing module. These will have to go on the big ticket list. I've been playing with qtVlm but it has a very crude interface (it's Open Source so low expectations). Still very cool to see how these tools are used and get a sense of their limitations.

You don't need to interface to instruments to do routing, but you do to use many of the other features of these program. About the only thing other than routing that I actually use are the laylines, to tell me when I can tack/gybe and be clear of a feature. For our ocasionally start line I just use the count down timer on my timex and some reference waypoints on the plotter with set distances from the line. 7070 obviously uses much more of the program.

Historically the market has been tiny and not very price sensitive (pro offshore navs and a tiny fraction of the blue water cruisers like us and steve dashew). However, the new virtual sailing games may have changed that . . . but the needs are somewhat different there than for 'real' passages.

I find qtVlm to be surprisingly good for free! It does pretty much everything a blue water cruiser would want.

I don't think I'd spend that kind of change if I didn't plan to go all the way with instrumentation. Good to hear you find qtVlm is useful. It looks to be written for gaming not real life but it doesn't sound like that's a concern. I suppose as long as you apply common sense to the results what's diff.

[Joli]

Todays........ the router flopped it again. Although it looks like the bulk of the fleet has decided a southern route is the best option.

[us7070]

in expedition, there is an option to turn off all the "tracks", but leave the optimal route.

can you do that in maxsea?

[Joli]

Are you asking if the isochrones can be removed? Sure, just check or uncheck the data you want displayed.

in expedition, there is an option to turn off all the "tracks", but leave the optimal route.

can you do that in maxsea?

[us7070]

Are you asking if the isochrones can be removed? Sure, just check or uncheck the data you want displayed.

in expedition, there is an option to turn off all the "tracks", but leave the optimal route.

can you do that in maxsea?

no - not asking about isochrones.

in addition to the optimal route, your screen is also displaying all the non-optimal routes that were evaluated.

for some reason exp calls those "tracks", and i usually turn them off.

typically, i am comparing several (5-10) optimal routes, with slightly different inputs, so the non-optimal tracks make it too busy, and harder to see the wind.

i can always turn them on, if i want to see them.

[Estar]

Good to hear you find qtVlm is useful.

I just did a comparison of the optimum Adrena route and the VW route, using the exact same grib and polars for BP's position to the crozet ice gate. They are essentially identical.

Adrena  BPADRANA.jpg   231.07K   11 downloads

VW  BPvw.jpg   88.23K   11 downloads

[Joli]

Mutliple scenarios can be run, lots of variable can be changed to see what's best. Old tracks can be saved or deleted, deleting a track is done by right clicking the track, seeing it again is by opening the saved track. Is that what you are asking?

Are you asking if the isochrones can be removed? Sure, just check or uncheck the data you want displayed.

in expedition, there is an option to turn off all the "tracks", but leave the optimal route.

can you do that in maxsea?

no - not asking about isochrones.

in addition to the optimal route, your screen is also displaying all the non-optimal routes that were evaluated.

for some reason exp calls those "tracks", and i usually turn them off.

typically, i am comparing several (5-10) optimal routes, with slightly different inputs, so the non-optimal tracks make it too busy, and harder to see the wind.

i can always turn them on, if i want to see them.

[Estar]

On your screen shot, you have the route (dark bold red), the GC track ( lighter red), the isochrones (yellow) and the alternate tracks used to create the isochrones (black or blue?). He is asking if you can turn the later off. My memory of maxsea is that they go on/off together with the isochrones - that is you can not have the yellow isochones without the black/blue radial alternative tracks, but I could be wrong about that as I never dug down that deep in the display options.

Mutliple scenarios can be run, lots of variable can be changed to see what's best. Old tracks can be saved or deleted, deleting a track is done by right clicking the track, seeing it again is by opening the saved track. Is that what you are asking?

Are you asking if the isochrones can be removed? Sure, just check or uncheck the data you want displayed.

in expedition, there is an option to turn off all the "tracks", but leave the optimal route.

can you do that in maxsea?

no - not asking about isochrones.

in addition to the optimal route, your screen is also displaying all the non-optimal routes that were evaluated.

for some reason exp calls those "tracks", and i usually turn them off.

typically, i am comparing several (5-10) optimal routes, with slightly different inputs, so the non-optimal tracks make it too busy, and harder to see the wind.

i can always turn them on, if i want to see them.

[us7070]

On your screen shot, you have the route (dark bold red), the GC track ( lighter red), the isochrones (yellow) and the alternate tracks used to create the isochrones (black or blue?). He is asking if you can turn the later off. My memory of maxsea is that they go on/off together with the isochrones - that is you can not have the yellow isochones without the black/blue radial alternative tracks, but I could be wrong about that as I never dug down that deep in the display options.

right - that's what i'm asking about.

[WHL]

Selecting "reverse" isochrones in Expedition can be occasiuonally useful to see the sensitivity of the route and the degree of latititude you have to be high or low of the suggested optimal route
With respect to an earlier question about risk in a routing that takes you away from historic norms, you can you the Race Notes feature of Expedition to define an area to avoid in the routing and in that way. you can force it to route more towwards a less risky conclusion

[Estar]

Selecting "reverse" isochrones in Expedition can be occasiuonally useful to see the sensitivity of the route and the degree of latititude you have to be high or low of the suggested optimal route

So what specifically do you look for , or look to avoid, with the reverse isochones?

I see them in all the packages, and I have read about them, but have not found much use for them in actual real world practice. . . .perhaps because I don't correctly know what to be looking for? I mostly look at the forward isochones, and the wind field, and at alternate routes.

[Estar]

As a specific example to discuss related to the ARC.

Here are some screen shots . . . what useful information do you all see in them?

Here are the (adrena) forward isochrones. Obviously you want to avoid the area I have circled in red - naything else useful here?

 forwardiso.jpg   226.94K   21 downloads

Here are the Adrena reverse isochrones. I see the same area to avoid. But is there some additional information here beyond teh forward isos that I do not see?

 reverseiso.jpg   249.88K   18 downloads

Here is the wind field from VW's. Personally I find this is the most intuitive tool - you want to stay in the wind and usually there is a 'river' on this chart you can follow.

 windfield.jpg   80.4K   19 downloads

Here is the VW reverse isochones where they add the grey area - I am not sure exactly how its calculated but wider is supposes to be less sensitive to weather forecast changes and narrower is more sensitive/more risk.

 reverse.jpg   235.21K   16 downloads

Now in all thses pics, I think both programs are making a bad (risky) choice, and the better choice is to go south of the 'bad area'. Do any of you see information in any of these pics that support that? (I base it on looking at the forecast inconsistency several days in a row, and the actual weather several years in a row).

[us7070]

Estar - the ARC right now is a particularly pathological situation for an optimal routing program.

also, i think you are plotting too many isochrones and reverse isochrones - it's hard to see what's going on.

i think the expedition default setting for isochrones is 6hrs.

an isochrone is a line comprising points that are an equal sailing time from the start

a reverse isochrone is a line of points that are an equal sailing time to the finish.

the optimal course is defined by tangency between isochrones and reverse isochrones.- although in general you won't see where they are tangent, because you would need to plot too many isochrones and reverse isochrones.

if the tangency is very "acute", so that the tangent isochrone and reverse isochrone diverge away from each other sharply adjacent to the place where they are tangent, then the optimal route is well defined, and moving off of it will result in much greater elapsed times.

on the other hand, if the isochrones and reverse isochrones are tangent, or nearly tangent, over a wide area, then the optimal route is not so well defined, and one can sail off of it without greatly increasing the elapsed time.

it's hard to see it, but in your problem area, the isochrones and reverse isochrones are more parallel than in other areas of the route - thus they are alerting us to an area where the program can't really optimize well.

[WHL]

Estar - the ARC right now is a particularly pathological situation for an optimal routing program.

also, i think you are plotting too many isochrones and reverse isochrones - it's hard to see what's going on.

i think the expedition default setting for isochrones is 6hrs.

an isochrone is a line comprising points that are an equal sailing time from the start

a reverse isochrone is a line of points that are an equal sailing time to the finish.

the optimal course is defined by tangency between isochrones and reverse isochrones.- although in general you won't see where they are tangent, because you would need to plot too many isochrones and reverse isochrones.

if the tangency is very "acute", so that the tangent isochrone and reverse isochrone diverge away from each other sharply adjacent to the place where they are tangent, then the optimal route is well defined, and moving off of it will result in much greater elapsed times.

on the other hand, if the isochrones and reverse isochrones are tangent, or nearly tangent, over a wide area, then the optimal route is not so well defined, and one can sail off of it without greatly increasing the elapsed time.

it's hard to see it, but in your problem area, the isochrones and reverse isochrones are more parallel than in other areas of the route - thus they are alerting us to an area where the program can't really optimize well.

Exactly. Well said.

[Joli]

Here's today, it looks like pressure will be building from the north. The only one that bit hard and early was the Alikaria.

[Joli]

Never knew they were called tracks. I always thought of them as alternate vmc's if you forced the route to another area. I don't know how to remove them without also removing the isochrones, maybe it can't be done with max sea?

When you reverse the isochrones are you simply sailing the course in reverse? If so, what value is it, the boat is sailing different angles in different places at different times?

[us7070]

Never knew they were called tracks. I always thought of them as alternate vmc's if you forced the route to another area. I don't know how to remove them without also removing the isochrones, maybe it can't be done with max sea?

When you reverse the isochrones are you simply sailing the course in reverse? If so, what value is it, the boat is sailing different angles in different places at different times?

no, reverse isochrones are not the isochrones of the reverse course - that would not be of much use.

imagine a one-design race, where all the boats have the same polars.

boats anywhere on the same reverse isochrone are tied - they are an equal sailing time from the finish.

boats on the same isochrone are _not_ tied - they are all an equal sailing time from the start - but who cares about that? some are going the right way, and some are going the wrong way!

[us7070]

the use of "track" for those other lines is not in any way a convention - it's not what I would call them.

It's just the term expedition uses in the settings dialog box to identify them. this is where you turn them on and off.

in a way, you can think of those lines as how the program has found the isochrones - sail along each alternate track for an equal time..., make a mark.., connect the marks - that's an isochrone.

[Joli]

So it's not really reverse, it's a mirror. If run backwards you only measuring time not distance? Right? Running start to finish you're measuring distance along a common time line, the isochrone.

Never knew they were called tracks. I always thought of them as alternate vmc's if you forced the route to another area. I don't know how to remove them without also removing the isochrones, maybe it can't be done with max sea?

When you reverse the isochrones are you simply sailing the course in reverse? If so, what value is it, the boat is sailing different angles in different places at different times?

no, reverse isochrones are not the isochrones of the reverse course - that would not be of much use.

imagine a one-design race, where all the boats have the same polars.

boats anywhere on the same reverse isochrone are tied - they are an equal sailing time from the finish.

boats on the same isochrone are _not_ tied - they are all an equal sailing time from the start - but who cares about that? some are going the right way, and some are going the wrong way!

[WHL]

Another way I like to take a peek at reverse isochrones: One of the aspects missing in the routing is how a boat performs in certain sea states. You can approximate that by refining polars with your experience on different wind angles and sea states, but reality can be very different e.g. If we run normal and reverse isochrones and it shows that we can be 5 degrees hotter, wiith little risk on time to the finish, then those 5 degrees might allow for higher average speeds that aren't in the polars, simply because the sea state you are actually experiencing, allows you to surf more often or have better action with less rudder movement.

[Joli]

I look at the "tracks" as alternate vmc's (velocity made to the course) but there can only be one optimum vmc. If a skipper believes there will be better opportunities in some other location and chooses to not sail the optimum vmc then certainly he can deviate to an alternate "track". Easy to understand, tough to know sometimes if the choice is correct.

It's amazing how the complex has become simple with a $300 laptop, gribs and data transmission.

Sailing in the great lakes where current is not much of a consideration and very unstable weather is the norm I've almost always found it best to sail the favored board to the best vmc (ie rocking the vmg polars). Sometimes you'll get burned by someone banging the corner but mostly you'll win with thios philosphy.

the use of "track" for those other lines is not in any way a convention - it's not what I would call them.

It's just the term expedition uses in the settings dialog box to identify them. this is where you turn them on and off.

in a way, you can think of those lines as how the program has found the isochrones - sail along each alternate track for an equal time..., make a mark.., connect the marks - that's an isochrone.

[Joli]

Bow out, making trees... that's always the goal. If you're not making your polars then sure, deviate.

Another way I like to take a peek at reverse isochrones: One of the aspects missing in the routing is how a boat performs in certain sea states. You can approximate that by refining polars with your experience on different wind angles and sea states, but reality can be very different e.g. If we run normal and reverse isochrones and it shows that we can be 5 degrees hotter, wiith little risk on time to the finish, then those 5 degrees might allow for higher average speeds that aren't in the polars, simply because the sea state you are actually experiencing, allows you to surf more often or have better action with less rudder movement.

[us7070]

Bow out, making trees... that's always the goal. If you're not making your polars then sure, deviate.

Another way I like to take a peek at reverse isochrones: One of the aspects missing in the routing is how a boat performs in certain sea states. You can approximate that by refining polars with your experience on different wind angles and sea states, but reality can be very different e.g. If we run normal and reverse isochrones and it shows that we can be 5 degrees hotter, wiith little risk on time to the finish, then those 5 degrees might allow for higher average speeds that aren't in the polars, simply because the sea state you are actually experiencing, allows you to surf more often or have better action with less rudder movement.

Most boats don't have good offshore polars.

I typically find that on Newport-Bermuda we can be more than 10% below polar speed - especially when the course is ~upwind.

This w.r.t. polars that are known to work well for inshore round-the-cans racing - not just some random file from the internet.

It's partly that the sea state is different, but also because the boat is a lot heavier - water, gear, food, fuel, sails, liferaft etc...

One option is just to slow the boat down - run optimizations at 90% of polar speed - and I do this.

But, what this approach doesn't capture, is that the target TWA's are different too - upwind your tacking angles might be 10-15deg wider.

Downwind, if there are waves to surf, you may COG a deeper average course than your target angles, or maybe not if the waves are not from a good angle.

The main thing is to realize that optimal routing is just an analysis tool - it helps me think about routing in a quantitative way. I do lots of experiments.

[Joli]

At the end of the day, I think you guys would be fun to race with. I'll play the part of snactition.

[Estar]

the optimal course is defined by tangency between isochrones and reverse isochrones.- although in general you won't see where they are tangent, because you would need to plot too many isochrones and reverse isochrones.

if the tangency is very "acute", so that the tangent isochrone and reverse isochrone diverge away from each other sharply adjacent to the place where they are tangent, then the optimal route is well defined, and moving off of it will result in much greater elapsed times.

on the other hand, if the isochrones and reverse isochrones are tangent, or nearly tangent, over a wide area, then the optimal route is not so well defined, and one can sail off of it without greatly increasing the elapsed time.

Ah, so I presume the area that VW shades grey is where they are 'close' to tangent - that grey shading does make it rather more obvious.

[WHL]

At the end of the day, I think you guys would be fun to race with. I'll play the part of snactition.

LOL. I like bacon sammies

[us7070]

the optimal course is defined by tangency between isochrones and reverse isochrones.- although in general you won't see where they are tangent, because you would need to plot too many isochrones and reverse isochrones.

if the tangency is very "acute", so that the tangent isochrone and reverse isochrone diverge away from each other sharply adjacent to the place where they are tangent, then the optimal route is well defined, and moving off of it will result in much greater elapsed times.

on the other hand, if the isochrones and reverse isochrones are tangent, or nearly tangent, over a wide area, then the optimal route is not so well defined, and one can sail off of it without greatly increasing the elapsed time.

Ah, so I presume the area that VW shades grey is where they are 'close' to tangent - that grey shading does make it rather more obvious.

can you post an example of the grey shading?

expedition has a shading feature, which will shade an area to either side of the optimal course.
the width of this area corresponds to a time differential -time in excess of that of the optimal course - that is configurable by the user.

where isochrones and reverse isochrones are more parallel, the area is wider.

[us7070]

So it's not really reverse, it's a mirror. If run backwards you only measuring time not distance? Right? Running start to finish you're measuring distance along a common time line, the isochrone.

Never knew they were called tracks. I always thought of them as alternate vmc's if you forced the route to another area. I don't know how to remove them without also removing the isochrones, maybe it can't be done with max sea?

When you reverse the isochrones are you simply sailing the course in reverse? If so, what value is it, the boat is sailing different angles in different places at different times?

no, reverse isochrones are not the isochrones of the reverse course - that would not be of much use.

imagine a one-design race, where all the boats have the same polars.

boats anywhere on the same reverse isochrone are tied - they are an equal sailing time from the finish.

boats on the same isochrone are _not_ tied - they are all an equal sailing time from the start - but who cares about that? some are going the right way, and some are going the wrong way!

I'm not totally sure what you are trying to say, but it doesn't sound quite right.

imagine that the program computed 1000 possible courses, one of which is the optimal course.

somewhere on each course, there is a point, that is a certain sailing time - t - from the finish.

for any value of t, draw a line connecting the points, one from each route, that is t hrs from the finish - that line is a reverse isochrone

[Estar]

can you post an example of the grey shading?

Post #51 above, last pic.

I am always interested in how spread out fleets get. I would intuitively think they would group into several logical routing clumps, but usually (and even at the level of the Bermuda fleet) it just looks random, like someone threw darts.

 arc.jpg   161.9K   19 downloads

Here are the two routing options today. Adrena is picking the north today by 9hrs. I would still go south, or as Joli mentioned, just plan on motoring down the middle for a while

 arcroute.jpg   215.4K   17 downloads

[Joli]

It looks like new pressure will come from the north. If your rum or north and a few days out a northerly route could work.

[Joli]

Heres today. Pressure is building from the north as the program indicated for 3 days now. The Akilaria is now in 20 knots of breeze and 100 miles ahead of the Swan 80, the Swan will suffer through light stuff for another day.

[Estar]

Heres today. Pressure is building from the north as the program indicated for 3 days now. The Akilaria is now in 20 knots of breeze and 100 miles ahead of the Swan 80, the Swan will suffer through light stuff for another day.

My friends both started in the second start - 2 days behind the Akilaria - so my routing is from much further back. And their boats are much slower than the Akilaria or a swan 80. They both are thinking they will go a little south, but also use a little motor.

That later start, with 20/20 hindsight looks like a mistake, but at the time it looked like a good thing to do.

[forss]

Looks like Andreas Hanakamp (AUT 40 Vaquita Class40) is going to win this race.
Poor Swan 80..

But volvo ocean race skipper is from another league.

[Estar]

I think Zen deserves great credit (so far). Its a swan 48 up running with the really fast boys . . . you could actually cruise on a swan 48!

[Roleur]

Can we go back to the subject of isochrones and reverse isochrones. I plotted this is VM with 3 hour isochrones



It seems clear that the grey area highlights were the isos and reverse isos are more or less tangent. What I'm wondering is how do you adjust the sensitivity (the width of the grey area). i get that the width is an indication of the sensitivity of the route, but relative to who? Is that relative to a route that a Volvo navigator would want or that a casual cruiser would want. Is there something I could change to make the grey area generally narrower? Or does that question make no sense?

[Estar]

Can we go back to the subject of isochrones and reverse isochrones. I plotted this is VM with 3 hour isochrones



It seems clear that the grey area highlights were the isos and reverse isos are more or less tangent. What I'm wondering is how do you adjust the sensitivity (the width of the grey area). i get that the width is an indication of the sensitivity of the route, but relative to who? Is that relative to a route that a Volvo navigator would want or that a casual cruiser would want. Is there something I could change to make the grey area generally narrower? Or does that question make no sense?

I don't know but . . . One of the developers is over on cruisersforum and happy to answer questions, and there is also a 'power user' there who is writing a better English help manual.

I understand the reverse isochrones but still don't really find them all that practically helpful. They can show you points where the route is sensitive, but practically speaking I don't really find that all that helpful in figuring out where to go. I much prefer running different model and grib perturbations thru the routing and seeing the sensitivity and routing alternatives that way.