You got it man, I mean, take the Sunday for example. A bike at first developed on "paper", using all of the aplied science and math I am so fond of, then tested for over a year in the real world. The beauty of the development cycle that I used is that every time I talked with Sam, Bryn, Matheiu etc... about performance, the performance traits always matched expectations. This gave the riders a lot of confidence to suggest performance changes that they knew they wanted, and it let me deliver new prototypes that reflected those changes almost immediately. We went through 3 years of what I would be satisfied with of "normal" development in a little over a year and 3 frame revisions. It really let us hone in on and modify exactly the performance traits that we wanted to affect, without casuing any detriment to traits that we really loved. Thats pretty awesome IMO, and thats the real power and efficiency of all this mathematical work.Kornphlake said:Right, you run numbers to make sure things will be strong enough, but you get to a point where you don't really know what the numbers mean untill you tinker around with it. For pedal kickback for example, you can calculate how much pedal kickback you'll have in some quantifiable units but you don't know how much is acceptable until you build up a bike and ride it to see if you can notice it or not, then it's an optomization game from there. The same could be said for structural analyisis, there's a point where you just don't really know what the variables are, I can calculate the impact force of a 200lb rider hucking off a 8' roof to flat but I don't know how that will affect the structure of the frame if the rider lands rear wheel first, how far on the rear wheel, tweaked out a little bit, and such. Hopefully the safety factor would account for variables like that but you really don't know until you have done some testing. Like DW points out the two really do go hand in hand, I'd consider a design that was built in CAD and used some FEA and such along with thourough testing to be superior to the homegrown tinker toy bike but computers and analysis can only take you so far.
dw