What ? You can't spend 100.000 on exotic tornado snuffers and sail development if you tried ?

There is no reason why you can spend the entire fortune of Bill Gates on a single beach catamaran, but I really don't believe that that will lead to any significant (= measureable) performance gain over say reasonably well designed a 25.000 boat when limiting yourself to say a ruleset like the F18's.




Too many people believe that this is a magic thing. That one only has to put a computer to work on it and that machine will figure out for you what is best. I do alot of computer simulation on noise disturbed data (= real life data) and it simply is not that simple or even that accurate. Alot of people also believe that they are skilled designers when they know how to operate a finite element software package but in reality they are simply not. A very large portion of the designing is to be found in intepreting the computer generated results and quite often you'll see numerical glitches. It take understanding of the algorithms used to make sound intepretations on accuracy and dependability of the generated data. Simply generating a plot with some colourings in it is not accuracy or dependability. Even plots with margins of error are often better at producing a false sense of accuracy and dependability.

An to make things worse. The behaviour of a hull when sailed over a disturbed watersurface is too complex too model accurately in all its details. That means that stochastic analysis (random variables) must be applied to even be able to "run the numbers" on a computer. Stochastic analysis is what you do to predict the outcomes in a Casino and we all know that we can only assign probability levels to those outcomes and thus never be sure. In my experience, probability analysis is something that is widely misunderstood even by well educated engineers. And I must admit it is a very tricky subject that requires you utmost focus to not overlook details that can change the whole model or your results.

To give an example :

Proof which situation has the highest probability of being truthful or else proof that all situations are just as likely the happen :

-1- it is easiest to throw at least the number 6 ones when throwing the dice 6 times
-2- it is easiest to throw at least the number 6 twice when throwing the dice 12 times
-3- it is easiest to throw at least the number 6 three times when throwing the dice 18 times


And then the best part of stochastic processes it that you need to be able to produce a component far more accurately then the difference that was designed into the boat from a prior design. Else the natural variance in production overpowers your designed in difference. Then all that work is just pointless as you will have to produce 10's of boats and test them all to see which boat is actually reflecting what you designed. Much like how the top Laser sailors go through scores of boats and mast to find the best combination. That is OD design for you !

As a matter of fact just producing a [censored] load of models and find the best one by testing them all may actually be the most inexpensive option as you won't need highly accurate (expensive) production tooling and you can still sell the lesser products to part time racers and recreational sailor, thus getting a pretty sizeable return on investment.

Indeed, one can easily spend 100.000 on a golden doornob with platinum and diamond decorations, but in the end of the day it is just a doornob whose function is implemented just as well by a 1 dollar plastic handle we all buy at wallmart.

Wouter