Sailonline.org - Online Sailing

Facebook

Login

Support Sailonline

If you haven't already - join the SAILONLINE YACHT CLUB!

Please also consider making a donation - all amounts are greatly appreciated!

Tweets by @sailonline

Board » Technical Support » qTVlm suggestions

Page: 1 2 Next

Posted by JB Posts 129 Registered at 2013-03-07 08:54:34	2016-02-24 14:17:04 - qTVlm suggestions
	In spite of a great work from qT Team and, now that maitai is around SOL, some suggestions from here: 1 - Routing side: 1.1 - There’s a disconformity between the SOL (TWD, TWS) live data and the one qT reads (GPS on) and routes latter (is it the interpolating algorithms at play and/or some problem/s with the “geographical centering” of the used Grid, the SOL vs. the qT one?); 1.2 - The routing inside the TWS under 6 kts - 7 Kts is over optimistic, the BS is over calculated; 1.3 - The minimum qT Isochrones Time Step calculation (5 minutes) with those TWS ranges (under 6 kts - 7 Kts), is not enough for coping with the wind vorticity associated with those conditions. Near/inside HP systems routing and getting a decent (TWA, BS) is almost a lottery; 1.4 - The Boat Performance recover option after tacking/gybing maneuvers should be more customizable, at least assuming a time linear, or not, recovery during a certain period of time that you chose; 1.5 - The “Simplification” process, even using the “optimal” option, is very radical, especially before/after maneuvers routed points. A second “by hand” optimization, normally associated with adding/moving extra route waypoints gets easily better results; 1.6 - Customizable Alarms; 1.7 - “Barrier” polylines file saving/importing option; 2 - Graphical side: 2.1 - The chart graphical scale disappears behind the digital dials when they are “on” (there was once a qT version where you could move it - the scale - to wherever you want); 2.2 - IMHO the analogical displays gimmick is dispensable; 2.3 - I would prefer to have: 2.3.1 - Only digital displays; 2.3.2 - Customizable locations on the qT chartplotter; 2.3.3 - With #3 decimal digits displays. 2.4 - ZyGrib has a functionality (menu ”Weather map/Options/Show Grib grid”) where you can see the GRIB nodes. That would also be good to have in qT; 2.5 - The former option for recording “screen shots” was better, now you can’t chose the graphic file format (JPG or PNG) nor the folder for record it (by default on the qT Program Files folder, not good); 2.6 - Isotachs lines would be welcomed. Thank you qT. Fair winds to all.
Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2016-02-24 15:45:06 - Re: qTVlm suggestions
	I think there's a forum somewhere where QtVLM devs are active (mostly french forum, but English is fine) I can react to some of the points: 1.1 The disconformity between SOL and QtVLM is mainly due to interpolation. I've search QtVLM's code once, and found QtVLM previously seemed to support multiple interpolation methods (non of which were equal to sol's though). 1.3 I believe that the small scale routing you are talking about is supposed to happen in the route module when you check "use route module between isochrones". 1.4 I agree the performance after tacking/gybing is not really clear or elaborate enough, but you shouldn't hope on it becomming complex enough to be able to mimic sol's interpolation.
Posted by JB Posts 129 Registered at 2013-03-07 08:54:34	2016-02-24 17:15:25 - Re: qTVlm suggestions
	Viva Huib. Hope you're good! A small notice. The graphical scale is ... really the chart graphical scale. Pls, have a look on the pics. a BIG HUG from here! Fair winds to all. Attachments Now_you_have.JPG (25.0 KB) Now_you_dont.JPG (25.2 KB)
Posted by maitai Posts 16 Registered at 2016-02-25 08:15:33	2016-02-25 08:31:56 - Re: qTVlm suggestions
	Hello, my first message here You can move the scale wherever you want, just right-click somewhere and choose "move map scale here"
Posted by Bimmer Posts 68 Registered at 2015-02-23 15:38:27	2016-02-25 12:38:25 - Re: qTVlm suggestions
	maitai Welcome to sailonline. Really nice to have you here
Posted by JB Posts 129 Registered at 2013-03-07 08:54:34	2016-02-25 12:55:26 - Re: qTVlm suggestions
	Thank you maitai for your tip. It works - pls see attached pic! The big qT issue is the Time Step (TS) used for calculating the Isochrones when the routing reaches inside low TWS areas associated with quick TWD rotation. The way to solve this problem is to impose qT a customizable minimum TS for, let’s say, 30 sec - 1 minute, to calculate the Isochrones when reaching a TWS less than a certain value, also customizable (5 Kts - 6 Kts or other value), avoiding the actual minimum TS of 5 minutes which, clearly, isn’t enough! The only drawback will be a longer time calculation, for sure, and the use of some extra RAM on your machine, but it will be up to you to stretch the routing calculation, or not. The other issue is the qT “Simplify route to the optimum” way. Sometimes, qT decimates a large quantity of calculated waypoints and tends to “linearize” the remaining routed path. As said, it gives you latter more work for introducing “by hand” extra waypoints (“marks”) and moving it “by hand” for reaching satisfactory results, which are easily better. A Big Hug from here! Fair winds to all. Attachments Chart_Graphical_Scale_OK.JPG (62.2 KB)
Posted by maitai Posts 16 Registered at 2016-02-25 08:15:33	2016-02-25 15:07:45 - Re: qTVlm suggestions
	Thanks for your input, I'll try to answer point by point: Time Step: Due to the way isochrones are internally calculated, it is not a good idea at all to set it to less than 4 times the vac duration, which is generally 5 minutes and sometimes 1 mn in VLM races, and is forced to 5mn in real boat mode (I could eventually make that a parameter). What is the time step in SOL? You can however deactivate the "use route module between isos" option and opt for a 1 vac time step between isos, but I seriously doubt it will give better results. I also suppose there are differences not only between SOL grib and polar interpolation, but also in the way SOL and VLM are assuming a point is reached. In VLM a point is considered "reached" if the distance to it is less than half of the distance made by the boat during previous vac. What is the rule in SOL? Also when routing qtVlm assumes you are sailing in VBVMG mode, something specific to VLM that probably don't exist in SOL. Simplify/Optimize: Optimum simplification just remove points that do no improve final ETA, while optimisation is there to re-position point "between isos" if needed. You can try to run first optimization and only then simplification to see if that helps. But again I guess the navigation mode in SOL and VLM (vbvmg) are different. You said you can easily improve manually the resulting route, can you provide an example I can reproduce? Note that you can change the navigation mode towards POIs in route (you can choose between VMG, ORTHO or VBVMG which is the default), so maybe you can choose something closer to SOL mode, whatever it is? Boat Speed in small TWS: Boat speeds are just taken from the polar, so if boat speed is too high then just fix the polar. Time to tack/gybe: In VLM races you can tack/gybe as much as you want without any time loss (a pity in my opinion). In qtVlm real mode you can give a penalty up to 1 full vac (5mn). I have already been asked to improve that so I am opened to suggestions. Thanks again for your comments Philippe.
Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2016-02-25 19:10:50 - Re: qTVlm suggestions
	maitai said @ 2016-02-25 15:07:45: (...) What is the time step in SOL? (...) In VLM a point is considered "reached" if the distance to it is less than half of the distance made by the boat during previous vac. What is the rule in SOL? Also when routing qtVlm assumes you are sailing in VBVMG mode, something specific to VLM that probably don't exist in SOL. (...) --- END QUOTE --- Time step in SOL is 10 to 30 seconds. Usually close to 10. In SOL no points are reached. Crossing the finish line is just that: line between previous and next position cross the line, then you're finished (some finish time correction is calculated). For rounding marks the same thing happens with a rounding line. Collision detection (currently) just checks whether the new position is allowed (i.e. not on land, and inside course boundaries). In the future (I'm working on it!) we will check for line intersections here too. Since SOL doesn't work with waypoints nothing else needs to be "reached". Sailonline only has constant heading and constant wind angle, triggered by time only. Routing using the VBVMG mode is probably just fine for sol. Once QtVlm and the user have decided on a route a SOLer needs a list of headings/wind angles. Occasionally (certainly in large wind shifts) these angles shift so much within 5 minutes we need multiple commands on sol, where on VLM a single VBVMG command might suffice. --- Last Edited by kroppyer at 2016-02-25 19:19:27 ---
Posted by maitai Posts 16 Registered at 2016-02-25 08:15:33	2016-02-25 22:18:27 - Re: qTVlm suggestions
	Ah, I understand now the need for more points, and why you feel simplification simplifies too much: that's because qtVlm assumes heading or wind angles will be managed by the server or the pilot (i.e. qtVlm assumes the boat will never go in bad areas of the polar, if a target is just upwind it will tack as many times as necessary "automatically", even if the wind shifts). If it is so, a good advice for Sol'ers would be to convert the route in ortho mode before simp/optim. Export and re-import the route in ortho mode for that, or select navigation mode individually on each POI, and only then run the simp/optim procedure. I will add an option for that in routing module when converting to a route, and also add a general parameter for the "default vac length" in real mode, that will help in your case I think. Cheers Philippe PS: I must try a race on SOL one day
Posted by JB Posts 129 Registered at 2013-03-07 08:54:34	2016-02-26 10:21:00 - Re: qTVlm suggestions
	Viva Philippe, Je te remecercie tes reponses et, surtout, ton intérêt dans ce sujet. Now, I’m going to prepare some files for reinforce my previous comments about: 1 - Time Step calculations; 2 - Simplification/Optimization process, in/with qTVlm router. I’ll put it latter on DropBox for you and/or use a “Something”Transfer to upload those files to you (my e-mail is pertisjm@net.sapo.pt, pls send me yours). Concerning your SOL sailing, feel free to sail with my boat (psail) in any race you like. Just let me know in advance to inform SRC. That one - psail - already sailed in qTVlm. But that’s another story. A bientôt Philippe! Fair winds to all.

Page: 1 2 Next

Please login to post a reply.

Races

Next Race: 00d 00h 00m

Current Races:

Garagiste's Test Lab

The first TIMED race of the northern hemisphere Summer takes us to the Gulf of Finland on a 165nm zig-zagging course beginning and ending in the seacoast village of Haukilahti and circling the island of Suursaari. It’s only appropriate that the boat for this race is Finland’s own Club Swan 50. This is a TIMEDrace so you may RE-REGISTER HEREto try again after finishing a run. You will have 13 days and 11 hours to test your skill and decision making after the race opens.
Race #1928
INFOby brainaid.de
PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking:
TRQ2 - TRCH - SUPSOL - SYC
RACE CLOSE: Saturday,
26 July at 23:00 UTC
Race starts: Jul 13th 12:00 Registration will open soon

Facebook

Login

Support Sailonline

Races

SYC Ranking

Series

Mobile Client