SW2004, Splines still SUCK!!!

"I don't know the politics that keep others from fully
exploiting parasolid's possibilities, but I have seen quite
well what it can do."

The bottom line is that the *high level routines* that need
to be in Parasolid don't exist ! This makes it extremely
difficult for software developers and it's why no seamless,
unified, hybrid solution has ever appeared that makes use of
Parasolid besides UGS PLM Solutions products. A few years
back I discussed the lack of high level surfacing routines
in Parasolid with Mike Crown who use to work for Varimetrix,
now VX. Mike Crown felt that SolidWorks Corp. would have a
very difficult time making SolidWorks a true seamless,
unified, hybrid modeler. Time has proven Mike Crown to be
one hundred percent correct.

Mike Crown was one of the nicest, sharpest guys in the
CAD/CAM business and he was well liked by several of us on
alt.machines.cnc. Total straight shooter and a long time
employee of Varimetrix (now VX).

Go have a look at the ACIS website. Download the .PDF on the
3D ACIS modeler. Note how Spatial pushes the hybrid
capabilities of ACIS.

"ACIS is ideal for constructing applications with hybrid
modeling features- integrating wireframe, surface, and solid
modeling functionality with both manifold and non-manifold
topology, and a rich set of geometric operations."

Like I've said for many years in this newsgroup, ACIS
contains the high level routines that make it much, much
easier to produce a seamless, unified, hybrid modeler.

With Parasolid, this is obviously a bitch to do and no one
has done it besides UGS PLM Solutions.

I do not believe for one moment that SolidWorks Corp can
pull it off alone. Only if other vendors such as D-Cubed
create more robust products can any real progress be made.

Based on these FACTS, this should give people a very good idea of
why Autodesk hired D-Cubed to work on their kernel !

http://www.caddigest.com/subjects/autocad/select/spatial_dcubed.htm

This may have been the reason why a slowdown occurred in
what D-Cubed has been able to produce with 2D DCM as well as
3D DCM as D-Cubed does what probably is much more lucrative
work for Autodesk.

SolidWorks must make the move to ACIS or it's going to be a
long slow painful road to seamless, unified, hybrid modeling
in SolidWorks.

It's also going to be very embarrassing when Alibre and
IronCAD move past SolidWorks in the second quarter of next
year by utilizing the surfacing routines in the ACIS kernel.

At $995 Concepts already makes SolidWorks look very bad when
it comes to pure modeling capabilities.

www.cadsoft-usa.com

jon

 

Take a good look at UGS PLM Solutions website for Parasolid.
No emphasis whatsoever on building a hybrid modeler with
Parasolid. None... just on how robust Parasolid is and that
Parasolid is the only kernel used in a high-end solution
(theirs)... Unigraphics.

Bottom line.... Parasolid does not contain the high level
surfacing routines and other various tools that developers
badly need to create a seamless, unified, hybrid modeler.

jon

 

Pro/e does not use ACIS (although, it does have ACIS import/export), it
uses the same kernal it always has (granite, just got that name a few
years ago).

Why do other tools, like SW, not take advantage of other parasolid
features, I would think that is a limitation of the developer or the
choices they make in utilizing those features, customer demands, market
direction and of course cost.
Since SW indirectly owns ACIS via it's parent company, Dassault
Systemes, I can guess that the decisions to utilize and promote
parasolid functionality maybe a contradiction to their future interest
or a partial reason or unwillingness for SW Corp to move faster with
implementing more surface tools from parasolid, for SW? And this may
also be a waiting game for SW Corp to see if they can break away from
parasolid, or shifting to ACIS, while maintaining the speed and
functionality the SW users are use too? IMHO, I still think parasolid
kicks ass but I'm sure the line between of speed and functionality is
becoming a blur so... who knows...when the timing is right, and if ACIS
saves SW Corp money and it does not have a adverse effect on it's users
as well as give them what they need...?

It all comes back to their user base and most SW users do not need or
will ever need surfacing... so what is the incentive for SW Corp??
Money and competition are the major driving influences and also new
users requesting more surfacing tools are the only hope for users who
have been requesting more surfacing tools.

Over the years, I just do not see SW Corp as a front runner to
innovation. Yeah, they are better than AutoDe$k, but becoming similar
over the years.. also they're a company they have been true to it's
intention, to be a mainstream design tool and just a bit more at times.
(mainstream = typical = average = general = conservative = the average
levi wearing 9-5 joe...) Unfortunately, SW has been trying to be
everything to everybody and in some ways loosing touch with the
mainstream, which is still a huge market and that is where tools like
Inventor (surfacing is limited but they are trying) have focused and
another reason why (read, no competition) SW has not pushed enough with
surfacing.
Now, with Pro/e adding more ISDX surfacing to it's foundation package
and SolidEdge pushing it BlueDot functionality.. as well as other
inexpensive tools adding parametric and relational surfacing,.. the hope
for adding more parametric curve and surfacing tools into SW will most
likely be shifting towards the needs of SW users in need of those tools.

So, finally, there are other mainstream, inexpensive and evolved
modeling tools which are again being innovative leaders and pushing
relational or parametric driven curves and surfacing with more ease of
use, control and manipulation functionality... so there is hope in the
forseable future...

...

 

"Since SW indirectly owns ACIS via it's parent company,
Dassault Systemes, I can guess that the decisions to utilize
and promote parasolid functionality maybe a contradiction to
their future interest or a partial reason or unwillingness
for SW Corp to move faster with implementing more surface
tools from parasolid, for SW?"

The tools that you think exist in Parasolid are not there.
If the tools were in Parasolid then SolidWorks Corp. would
have implemented them by now. SolidWorks is without a doubt
a leader at early innovation of 3rd party software
components. That you think the proper high-level tools
somehow exist in Parasolid ignores all common sense. Specifically you
ignore why so many companies have chosen the ACIS kernel
and have put up with Spatial when the ACIS kernel left a lot
to be desired, several years ago. Why do you think
innovative companies like IronCAD chose ACIS first ??? Why
was SolidWorks originally built with ACIS ??? The answer is
the it's a more complete solution with higher level tools
but up until a few years ago ACIS was not robust enough.
ACIS is probably still not quite as good as Parasolid at
solid only modeling.

IMO, Solidworks Corp. was caught out when Autodesk hired D-
Cubed to work on ShapeManager and both SolidWorks and D-
Cubed development have suffered.... BADLY !

"but I'm sure the line between of speed and functionality is
becoming a blur so... who knows...when the timing is right,
and if ACIS saves SW Corp money and it does not have a
adverse effect on it's users as well as give them what they
need...?"

For the last year or more Spatial has mainly worked to shore
up ACIS and make it more robust and make it faster. Very
little work seems to be done on new features. IMO, this is
so that installing ACIS in SolidWorks and CATIA will not
cause regression.

"It all comes back to their user base and most SW users do
not need or will ever need surfacing... so what is the
incentive for SW Corp??"

How bad does it look that $995 Concepts is a more versatile
modeler than SolidWorks and surfacing and solid tools work
very well together in Concepts ?

Concepts can be downloaded for a test drive at:

www.cadsoft-usa.com

Why put up with the low level surfacing functionality of
Rhino or the poor surfacing tools in SolidWorks ?

How bad will it look when in the second quarter of next year
Alibre and IronCAD implement more of what's in the ACIS
kernel ? Hopefully PTC will keep losing their ass and find
away to include ISDX II with Pro/E Wildfire at no charge.
If this happens the shit will really hit the fan because
Pro/E will suddenly become a major bargain. If PTC
manglement had any brains they would already be doing this
instead of continuing to loose their ass and sink in red
ink. What a mess.

"SolidEdge pushing it BlueDot functionality..."

There is a new release of SolidEdge do out soon and it looks
to have some very strong mold design tools and can do stuff
that SolidWorks can't.

See:

http://www.cadcamnet.com/

This new version of SolidEdge will be targeted toward mold
designers / mold shops because it now has *more* innovative
features that most other products don't have. IMO, UGS PLM
Solutions knows that the game is over and that SolidEdge is
going to have to be a real bread winner for them and compete
and win against SolidWorks, Inventor, etc. in more
situations.


"So, finally, there are other mainstream, inexpensive and
evolved modeling tools which are again being innovative
leaders and pushing relational or parametric driven curves
and surfacing with more ease of use, control and
manipulation functionality... so there is hope in the
forseable future..."

What you wrote above is beyond obvious. What doesn't help is
so few SolidWork users (yourself included) don't understand
what the real issues are with Parasolid and why ACIS needed
to be in SolidWorks yesterday.

Anyone who thinks that SolidWorks Corp. can make the
surfacing that is in SolidWorks work properly without
outside help is in serious fantasy land.



jon