Re: [Man.Distr] Dual Processors for CAD Use

From: <Christopher>
Date: Wed Nov 19 2003 - 22:08:00 EST

>But this is justified because it's so easy to fix things later. And why
>bother hiring expensive, experienced people when it's so easy to correct
>initial mistakes?

It's only easy to fix things later if you realize you've made the mistake and the mistake isn't something that doesn't get someone hurt or waste time or money. I claim it's almost trivial for someone to make a very attractive drawing package that looks extremely competent but in fact is poorly thought out and costly to make. I've seen plenty, just as I've seen plenty of finite element analysis done by people who haven't a clue what the work actually means.

>Isn't the design process more iterative than linear, given the inevitable
>changes along the way? I'm talking about clean-sheet designs, not modified
>template work.

There are all kinds of iterations. When the iterations converge on a goal, it's called design. Other kinds of iterations involve clueless guesswork, and the result is the same whether they're automated or not.

The point is that CAD and FEA are tools. Good tools help an artisan produce quality work, but they won't turn hackers into artisans.

Christopher Wright P.E. |"They couldn't hit an elephant at <a href="/group/PipingDesign/post?postID=OrSxUlTYq3JrkUHiN2-52fYS0alHm_tBLPuSjBC-7FfMmcBIlnO6rxi4Luojq3mSVV-ojtfCzaOsavI">chrisw@skypoint.com</a> | this distance" (last words of Gen.

___________________________| John Sedgwick, Spotsylvania 1864)

<a href="http://www.skypoint.com/~chrisw">http://www.skypoint.com/~chrisw</a> Received on Wed Nov 19 22:08:00 2003