Mario wrote:
>> I'm eager to compare it with SolidWorks.
They are quite different. Solidworks is history based. Create a cube,
punch a hole and add some fillets to all the corners. It remembers all the
steps and you can go back and change the size of the cube and it will
rebuild. But if you move the hole where it splits the edge of the fillet,
it fails to rebuild and you start solving errors. This is common to all
history-based parametric CAD systems. It's easy to build parts, then you
spend all your time trying to figure out why it blew up when you changed a
dimension.
Rhino doesn't use a history tree. Grab a surface and push or pull it, move
a hole, or chop the cube in half. Rhino lets you do anything. It's a lot
like AutoCAD in that regard. It doesn't matter how built the solid model.
When I can't get SolidWorks to properly build the model (fillet issues are
real common on plastic parts), I export the model into Rhino, change it,
then import it back into SolidWorks.
If you need to drive a model with dimensions to create a family of parts,
SolidWorks and other like it are the way to go. But if you are designing a
single part with complex geometry, Rhino is about as good as it gets. They
are both CAD systems, but the history based systems give up geometry control
in exchange for the ability to somewhat automate design changes. I have a
design that is spreadsheet drive to produce 20 different version simply by
altering the values in Excel. Rhino can't do that.
Ken Elliott
=====================
-----Original Message-----
From: slug@nks.net [mailto:slug@nks.net] On Behalf Of Mario Lombardo
Sent: Saturday, June 09, 2007 1:27 PM
To: slug@nks.net
Subject: Re: [SLUG] Real World Class 3D modeler for Linux
On 6/9/07, Ken Elliott <kelliott11@cfl.rr.com> wrote:
> Ken Elliott wrote:
> > Compare the Stealth Figher vs. the B2 Stealth Bommer.
>
> Chuck Hast wrote:
> > The flat surfaces on the stealth fighter are set in such a way as to
> > scatter radar and reduce the reflectivity of the fighter, that is
> > why it is like that not due to the design limitations.
>
> I suggested this example because BOTH are designed to scatter radar.
> But the flat surfaces tend to create a more focused reflection than
> the non-uniform curved surfaces of the B2 Bommer. The figher has flat
> surfaces because of the limitations of the CSG modeler and lack of
processor power.
> This is well known among CAD developers.
>
> CSG modelers usually describe non-uniform curved surfaces as a series
> of connected triangles. The smoother the surface, the more triangles.
> Trying to model a water hose can take millions (or more) of triangles,
> while a NURBS modeler only needs 2 entities: a cross section curve
> (like a circle) and a path curve. NURBS curves are very lightweight
> in data, but it takes lots of number crunching to process it.
>
> In the past, NURBS were not practical for simple geometry (tables,
> houses,
> tanks) because of the overhead of caluculations. CSG is super fast,
> and as long as the geometry didn't get too complex, it was the tool of
choice.
> Most CAD systems today started off as CSG modelers. But once the CPUs
> got fast, NURBS became possible, then practical, and now there is
> little reason not to use them. This development occurred in the late
> 90's, and the first affordable modeler (under $30,000) appeared on the
> market (Rhino in 98, I think).
>
> So here's the big deal about Rhino. All other CAD packages started
> off as CSG, and added some NURBS entities. So your model is a mixture
> of CSG and NURBS. Rhino is 100% NURBS, except for text and
> dimensions. It can simply do things that no other modeler can do.
> Just having this package on Linux would be a wonderful thing. But
> McNeel has opened the source code and specs of the file format. No
> other CAD company does this - they all hold your design data hostage.
> McNeel is by far the most open minded of all the CAD vendors. The tool is
fantastic and he thinks like open source people think.
>
> OpenNURBS greatly lowers the difficulty of bring high-end design tools
> to Linux, and we badly need this sort of thing for Linux to be taken
> serious in engineering circles. Here's a thermal analysis program that
supports Linux.
> http://www2.rhino3d.com/resources/display.asp?language=en&listing=107
>
> This is all good for Linux. Engineering and design tools were one of
> the early target markets Microsoft went after with Windows NT. The
> field was dominated by Unix workstations, and MS went after them with
> a lower cost platform. If anyone in a large company will be
> pro-Linux, it would be IT people first, then engineers. I watched MS
> work hard to woo the CAD developers in the early 90's and it paid off big
time.
>
> Vote here:
> http://offbroadway.blogspot.com/
>
> OpenNURBS
> http://opennurbs.org/
>
> CSG
> http://en.wikipedia.org/wiki/Constructive_solid_geometry
>
> NURBS
> http://en.wikipedia.org/wiki/NURBS
>
> I'll shut up about this for now. As you can tell, I'm passionate
> about getting good engineering tools ported to Linux.
>
> Ken Elliott
>
> =====================
> -----Original Message-----
> From: slug@nks.net [mailto:slug@nks.net] On Behalf Of Chuck Hast
> Sent: Friday, June 08, 2007 2:25 PM
> To: slug@nks.net
> Subject: Re: [SLUG] Real World Class 3D modeler for Linux
>
> On 6/8/07, Ken Elliott <kelliott11@cfl.rr.com> wrote:
> > >> If you can design tanks with it you can design anything with it.
> >
> > Far from it. Tanks are simple geometry. Streight edges, flat sides,
etc.
> > Items with smooth curved surfaces are much more difficult. Think
> > car bodies, boats, consumer goods.
> >
> > Compare the Stealth Figher vs. the B2 Stealth Bommer. The figher is
> > all flat sides due to the limitations of the software and cpu
> > horsepower used at the time. NURBS and fast computers allowed for
> > the complex calculations required for a complex curved surfaces.
> >
> > But if you disagree, that's fine. I've only been using CAD systems
> > for 25 years and I could be wrong...
> >
>
> The flat surfaces on the stealth fighter are set in such a way as to
> scatter radar and reduce the reflectivity of the fighter, that is why
> it is like that not due to the design limitations. It is all about
> radar signatures and those flat faces are set in such a way to make it
> hard to get a good primary return off of the aircraft.
>
> --
> Chuck Hast -- KP4DJT --
> To paraphrase my flight instructor;
> "the only dumb question is the one you DID NOT ask resulting in my
> going out and having to identify your bits and pieces in the midst of
> torn and twisted metal."
> ----------------------------------------------------------------------
> -
Not at all. The info is much appreciated. As I mentioned before, I've been
on the lookout for such. Passion is important as it drives one to learn
more and more about a complex piece of software.
Obviously it must be something amazing to impress your experience level.
I'm eager to compare it with SolidWorks. I'm keeping a copy of this thread
to aid me.
-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages posted
are those of the author and do not necessarily reflect the official policy
or position of NKS or any of its employees.
-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages
posted are those of the author and do not necessarily reflect the
official policy or position of NKS or any of its employees.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 17:07:05 EDT