[Beowulf] Re: Beowulf Digest, Vol 15, Issue 16

[Jim Lux]

----- Original Message -----
From: "Lombard, David N" <david.n.lombard at intel.com>
To: "Jim Lux" <James.P.Lux at jpl.nasa.gov>; "beowulf" <beowulf at beowulf.org>
Sent: Tuesday, May 10, 2005 3:51 PM
Subject: RE: [Beowulf] Re: Beowulf Digest, Vol 15, Issue 16


> From: Jim Lux on Tuesday, May 10, 2005 3:45 PM
> >
> > At 03:37 PM 5/10/2005, Lombard, David N wrote:
> > >Hmm, too "big" or "flimsy" have very precise definition based on
> exactly
> > >the measures you describe.  Such goal functions have been in use in
> the
> > >various structural optimizers for years now, e.g., weight, stress
> > >distributions, deflections, modal responses, &etc.
> >
> >
> > True, when you have hard specifications.  The hard part is negotiating
> a
> > relative weight for each of these factors when doing the tradeoff, as
> is
> > often the case in early trade studies.
> >
> > That's where the "skilled evaluator" comes in.  A person with some
> years
> > of
> > experience can  integrate all those different factors fairly easily.
>
> Fair comment, preliminary design has always been a tough analysis area.

And the one where "interactive" modeling is most valuable, I think.  Once
you've got the conceptual design philsophy worked out, done all the costing,
the detailed modeling is more to make sure you meet the cast in concrete
requirements (or whether you need to write a waiver and have it approved).
There's lots of ways to get the details, once I've figured out how many 6-32
screws I'm going to need and what kind of paint I'm going to put on the darn
thing.  It's the "what if we make it 10x15 meters instead of 12x12 meters?"
kind of questions you want to have those "quick and dirty" answers to.

And, this is where cheap computational horsepower is most valuable. Of
course, it also needs some clever software developers to make decent
"interactive" front ends to allow defining a structure from top down, with
reasonable assumptions from the bottom up.  It seems that these days (at
least from my viewpoint, which is limited) you can define the structure top
down (in a drawing tree sense), but you have to define the details bottom
up.  There's no "generic fastener" block that I can use in a top down model
that embodies a generic structural (or electromagnetic) properties.  Modern
CAD tools seem not to do well with "rules of thumb".