SpaceClaim?

Anybody got anything to say about SpaceClaim? Was it Ed Eaton who had seen
some new technology a few months ago that he found promising? Is this it?

Review here:
http://mcadonline.com/index.php?option=com_content&task=view&id=377&Itemid=73

Website here:
http://www.spaceclaim.com/

From the review and website, it appears not to be aimed at mechanical
designers. Is this just a marketing slant to keep from competing directly
against the entrenched CAD systems, or is the system too limited in what it
can do?

Jerry Steiger
Tripod Data Systems
"take the garbage out, dear"

 

These guys have been working on this product for a while. I haven't seen
the actual software, but my impression is that it's a non- or
semi-parametric modeler, avoiding constraints, letting you edit the
model directly by moving faces around. Kind of what I remember HP Solid
Designer being. They seem to be focusing on editing geometry
non-parametrically.

Another blow to the Mac/Linux contingent - a brand new CAD system that
only runs on XP and Vista.

As I remember, JB was the first to post about these guys.

http://groups.google.com/group/comp...st&q=spaceclaim&rnum=6&hl=en#e508d9d1542f3600

I read a blog entry recently about this by Ralph Grabowski
http://worldcadaccess.typepad.com/blog/2007/03/spaceclaim_make.html

Ralph is unimpressed with their licensing scheme - basically renting the
software, and then it turns into a viewer when you stop paying.

I don't think this is going to make much of a splash with people
designing new products, but it may make sense for the Jon Banquers out
there who think that machinists should change design data. They seem to
be targeting consumers of 3D data, and claim that they won't be
replacing SW or Autocad.

 

If that's the case, this is more of a challenge to programs like
SurfCAM rather than SolidWorks.

Matt
http://sw.fcsuper.com

 

It's hard to make scrap when you change the design to suit later, right?
After all, it was just one hole; why should it matter how big, how deep
or where?

Anyone else missing jb (now rumored to be gracing S. California just North
of San Diego with his ads & humorous sayings)?

 

Nope, this is not the technology I was hoping for -but I am astounded
by your memory Jerry, and it turns out that my contact did end up with
this company.

Through my contact I saw a demo of SpaceClaim a while back. There was
some neat, promising stuff. I remember liking how their sketcher
operated/looked, and some other nice interface stuff like their take
on 'select other' which I instantly wished I had in SWx. But beyond
some good interface improvements, what I saw was a package that does a
lot of stuff I don't really want to do.

To be fair, however, it was still early, and their sample model was
just... trying to find a better word here but I can't... awful. I
gave them my honest opinion (about the hot interface stuff AND the
poor samples and how it failed to communicate a case for portability
to the problems I run into), and have not heard a peep from them
since.

I do wish them luck, and maybe if I later see some applications to
real problems I and my cohorts run into I will change or open my mind;
I have to think there is some sort of application in industry that I
am blind to or they wouldn't be able to corral the personnel that they
have. But based on what little I know right now, I will not watch
them closely.

At SWx World I did get a whiff of another initiative in another
company that has some promise along the lines I would like to see, but
lets face history - don't they all seem cool until they come out?

Side point - Contrary to the article you shared, I think it's good and
appropriate that folks keep trying. This is not a mature market - it
<feels> like one on the verge of another big paradigm shift. For
instance, looking at how I and my coworkers have to work with CAD, we
spend wayyyyyyyyyyyy too much time fumbling within the limits of the
BREP (face and edge ID issues, and dealing with limits of faces and
boundaries relating to the underlying processes used to create them).
The CAD guys keep adding patches, and maybe this will bring salvation,
but my designer 'spidey sense' says that there is a better way, just
like when I go to a factory and watch folks use old/existing tools and
through that observation I see opportunities to improve those tools
(aside: I learn a lot more by watching folks, especially if they are
trying out one of my designs, than I ever do by asking them questions
because most folks are just plain nice and don't see or wan't to call
out the bad stuff to a questioner)
The troubling bit is that in my experience, the CAD developers don't
do the Jane Goodall thing and go observe the monkeys - they (sometimes
but not all the time) go ask the monkeys what they want, then try to
give the monkeys what they think the monkeys want. It is a rare
monkey that has the ability to really stand back and look at what they
do dispassionately - for instance, the monkey writing this post
thought that multi-body solids would introduce a lot more problems
than it solved, and was I EVER WRONG about that one! If a developer
sat and dispassionately watched me or one of you work for two-three
weeks they would learn sooo much more about the real problems we run
into - not just the ones we are aware of, but the bizarre stuff we
accept that we just shouldn't.
The example that jumps to my mind is when McEleny said that he saw SWx
on a TV report one day and saw the operator had to move the old, huge
'view orientation' box around the screen every time they tried another
operation and was chagrined that SWx didn't notice the problem before
(to be honest, I as a user didn't key into it either). They obviously
made a change based on that observation of monkeys. At SWx world in
february, watching other folks work during their presentation, I
remembered seeing more 'duh' samples like that 'view orientation'
sample as I watched the monkeys work - and I betcha anything that
folks saw cumbersome stuff that i did during the 'driving swx'
portions of my presentation that this monkey is not attuned into.
But those are simple interface patches - there is an underlying
inefficiency/deficiency in the current CAD process that, in my gut, is
ripe for an overhaul.
I would bet that most of us would agree that we will not be designing
products in 2057 fundementally the same way that we design products in
2007. It will be interesting to see if someone in 2008/9/10 has the
vision and courage to take the first steps toward that 2057 CAD
standard, or if they will continue to... trying to think of a better
way of saying this but I can't... put another layer of polish on the
turd.

 

Preach it, brother, preach it!

SolidWorks was fresh and new and exciting when I first starting using it in
98+. Now it makes me think of the inner tube from some 1930s cartoon, where
you can no longer see the tube itself, just the patches on the patches.

On the other hand, this monkey thought that multi-body parts would make my
job much easier, although I wasn't able to articulate why very coherently.
But a lot of the problems we have now seem to be due to multi-body parts.
Great idea, poor implementation. I think they fought against it so hard
because to do it right meant wholesale rewriting of the code. It feels like
they opted to put patches on patches, rather than start from scratch.

Absolutely. One of the problems with talking about what we do is that we
quickly learn workarounds when things don't work the way we want, then we
get so used to them that we don't remember that there was ever a problem.

Lord, I hope not. SolidWorks will drive me nuts long before then. Although,
barring some tremendous improvements in health and longevity, I'm pretty
sure that I won't be designing at all in 2057. I'll be pushing 110!

Jerry Steiger
Tripod Data Systems
"take the garbage out, dear"

 

That's the way it looked to me. If it gave me SolidDesigner capability with
a better user interface, it would be worth a look.

I'm not thrilled about that aspect, either. Although I don't see a
tremendous difference between renting the software and buying it and being
forced to pay extremely high property taxes to keep it usable.

It seems a strange way to pitch their product, but my marketing skills are
even worse than my programming skills.

Jerry Steiger
Tripod Data Systems
"take the garbage out, dear"

 

Sometimes CAM programmers need geometric data for manufaacturing
purposes that's not part of the model but needs to be added.
But it should be added based on the model, not created by altering the
original 3D model.
Often a firm will set layering standards just for this purpose. Or it
could be done in several other ways.

Not that poor clueless ...

 

If you look at their hardware requirements they look like those for
gaming systems. I didn't see Quadro or OpenGL anywhere. This appears
to be a Vista centric direction for the OS.

On the plus side they appear to support CATIA both ways. Remains to be
seen what that means.

The licensing seems to be a big boost for Linux 3D GPLed software
should that ever become available.

As to calling it new, well the arrow pointing to the HQ seems to point
in the same general direction of the building SW is in.

TOP