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 Discussion » Performance loss

Page: First Previous 6 7 8 9 10 11 12 Next

Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2014-07-18 15:02:50 - Re: Performance loss
	rod, Starting at 100%, a loss of 8% is achieved by With tack or gybe: Tack to an angle that gives sog=16 kts, more speed is more performance loss. (the angle depends on polar and tws) Without: Turn 110 degrees. Greater angle is more performance loss. For more complicated things (not starting with 100%, or doing more than one manouvre) please refer to the PDF I posted earlier in this thread. --- Last Edited by kroppyer at 2014-07-18 15:04:28 ---
Posted by dtayls Posts 17 Registered at 2014-06-23 03:34:16	2014-08-15 04:30:08 - Re: Performance loss
	Wow! Great thread. I've raced for almost 40 years and I'm an engineer & a computer scientist, so this one's right up my alley. Having just read through the whole thread here are a few thoughts from me is not particular order: 1. SOL started with a realistic model and that should probably remain. Anything that breaks with realism should be explained, at the very least. 2. There are good real-world reasons for deaparting from the realistic model. Server load and reasonable sample rates need to be considered. 3. Different yachts recover differently so how you should sail each yacht should be different. 4. The world is changing quickly. Downwind-flyers come on and off the plane and their performance changes massively when this happens. If a boat is only just on the plane then a gybe will be costly. 5. And don't even talk about foiling and whether you can keep it up during a tack/gybe!!! I suspect the SOL models are based on displacement vessels. It's probably time for an update particularly given that we're playing with some pretty impressive craft these days.
Posted by dtayls Posts 17 Registered at 2014-06-23 03:34:16	2014-08-15 06:24:08 - Re: Performance loss
	I should also mention that I was impressed by some of the analytical work. Excellent stuff. This reminds me of a marketing subject I did around 20 years ago at university. A key component was a computerized marketing strategy game. In our group we analyzed the 'black box' of the game so well that after the results were published the lecturer bailed me up and accused me of cheating ... Let me just say "I like your work."
Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2014-08-15 09:45:59 - Re: Performance loss
	I made this yesterday: Sailonline Performance Calculator http://sol.kroppyer.nl/performance/ It's about the same as the spreadsheet I posted earlier, but this one is more precise. Only time lost to next DC is not always accurate (because there is no clear definition for "time lost"). dtayls, funny that you read and posted on this thread just after I finished the calculator. Of course a new model would still be simplified reality, heavily simplified. But it's not hard to make a model that's more realistic than the current one, with the current model I fail to see the relation with reality. You might see some worked out propositions for new models over here fairly soon.
Posted by dtayls Posts 17 Registered at 2014-06-23 03:34:16	2014-08-16 01:24:32 - Re: Performance loss
	Kroppy, Some SOLers are sailors while others are not. It is probably true to say that most sailors want the model to be ultra-realistic while SOLers just want a model (aka. rules of the game) that they can understand and play against. I personally think there is a middle ground that will probably work for everyone ... if implemented and explained properly. The 15 second sample/update interval is an interesting aspect of the game. I am sure there are many reasons for this - server load and UI development requirements are just two that come to mind. But let's face it, this is not an online shoot-em-up game and people should probably not be trying to sail these things in real time. And if you want to play with a high performance yacht and really know how to tack/gybe on the plane/foil then get out on the water and do it IRL! But that said, I think there is a lot of value to be had in changing some parts of the model and most of these are probably quite straight forward to implement. Dtayls.
Posted by HappyHour Posts 19 Registered at 2009-01-08 06:48:51	2014-10-02 01:47:42 - Performance loss with high speed boats
	A performance model hasnt to be realistic but it should work. Our current model works with slow boats ( max speed 15kn ), of course far from perfect but it works. It doesnt work well with faster boats (max speed 15+ kn). With a AC72 it just gets crazy. For example: Gybe sailing downwind with 40kn speed at 92% performance, hurrah, no performance lost. Doing the same at 94% performance, oops, performance drops to 74%. Thats not just unrealistic. thats stupid! Isnt the 93% rule outdated, a rule from history when the max speed was around 14kn? Why dont we drop the 93% rule, let all performance calculations as it is and set hard limits, like max perf loss 90%. Maybe even better: Divide our fleet into "slow" and "fast" and set differrent max performance lost limits, like 88% for the fast boats and 93% for the slow boats. IMHO it should be possilbe to implement these changes to our server software. Just an idea.
Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2014-10-02 08:48:53 - Re: Performance loss
	I fully agree that performance loss needs an update. To say the least I really disagree on the idea that performance should be different across the fleet. First of all I like the beauty of everyone is exactly equal too much. Second, why would the people with less practise be faster in manoeuvres than the people with more practise? [edit: I misunderstood the idea] --- If it's going to happen, should we try a "do it once and to it good" approach, or should we spend some time on an intermediate performance system as "quick fix"? --- Last Edited by kroppyer at 2014-10-02 15:55:39 ---
Posted by HappyHour Posts 19 Registered at 2009-01-08 06:48:51	2014-10-02 12:43:18 - Re: Performance loss
	when i was talking about fleet, i didnt mean the boats in a single race. I meant the diff between a "slow" yacht like first 40.7 and "Racing machines" like VO4 or our latest baby AC72. Our fleet of available boats. From 6.5m upto 90m. --- Last Edited by HappyHour at 2014-10-02 12:48:49 --- --- Last Edited by HappyHour at 2014-10-02 14:04:02 --- --- Last Edited by HappyHour at 2014-10-02 14:06:05 ---
Posted by kroppyer Posts 244 Registered at 2010-02-05 15:25:55	2014-10-02 15:54:33 - Re: Performance loss
	Ah! I misunderstood. Now I fully agree: the performance system should depend on the boat type. A very simple solution would be as I described a few posts back and use the weight of a boat for the performance calculation.
Posted by HappyHour Posts 19 Registered at 2009-01-08 06:48:51	2014-10-07 15:56:46 - Re: Performance loss
	"or should we spend some time on an intermediate performance system as "quick fix"?" If some time means some hours, we should do the "quick and dirty fix". As i said in my previous posting, let all the calculations as it is and set a hard limit for the max. performance drop. that would at least eliminate the worst impacts of our current perf system. Today i sailed 2 runs with a fast boat in the Atlantic Circle and tried to beat the performance drop system and it seems to work. A 3 step tacking, like from 45 twa to -5 twa, -20 twa and -45 twa may be faster then a normal tacking from 45 twa to -45 twa. The same with gybing. Gybe to the slowest possible speed to minimize the performance loss. Anything will be better than a 34kn speed to 34kn speed gybe with a 17% perf loss at 94% performance. A nice tactical aspect, but im sure this isnt what the developers had in mind when they wrote our current code. --- Last Edited by HappyHour at 2014-10-07 16:23:31 ---

Page: First Previous 6 7 8 9 10 11 12 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: 七月 13th 12:00 Registration will open soon

Facebook

Login

Support Sailonline

Races

SYC Ranking

Series

Mobile Client