FeatureWorks - Works Even Better.

Very helpful post. Thanks. Glad to see that FeatureWorks has now got this ability.

Now what do we do about dealing with non-native surface geometry and taking it from "dumb" to "smart" like can be done with think3's thinkshape with Global Shape Modeling and to a lessor extent VX's Vision ?

BTW, what are you using for CAM ?

jon

we use sig erson.

Did your doctor approve ?

What about .stl imports. Right now all i can do is look at models. I would liek to be able to do more with them for RP production.

I tried out feature works to see what was different. The dumb solid was a SW generated iges file that came in without errors. FeatureWorks got some of the features but when mapping them there were fewer in the feature tree that there were during recognition. In addition the recognized features all had errors and SW2004 locked up.

Sean-Michael Adams wrote:

Raises the question of whether or not SolidWorks should be *totally dependant* on FeatureWorks to begin with for imported non-native geometry.

jon

No clues what RP or .stl imports are I see.

Well jon why dont you suggest something for SW to use in stead of feature works. Another 3rd party...wait no you said that SW relies too much on 3rd parties. Implement something into the program, well now that is something that has to be done in the future.

I am the first one to complain about somethings that I would like to see implemented, sometimes SW what we ask for gets there, others....well you see where we are. Thing is, jon, you always seem to complain about and pick up on the negative that we are saying. You dont use the program, at least from what i remember you saying. So why do you hang in this UG?

Anyways......Cliff Rp stands for rapid prototyping, and .stl is an extension used to get files from CAD to the different machines

I well know . Stereolithography file; polygon mesh. Ever worked with data from a CAT scan or NMR scan for bone? The human mandible is quiet complex; even worse minus teeth plus bone absesses.

Now try to machine (not conventional RP) something that started out life as 1 mm 3D solid pixels .... the size of a head ... with a 1 mm standard error .... a pixel is either on or off; on if anything crossed it anywhere ... within that 1 mm cube ...

RP has a clear place .. you don't need to try to make nice smooth NURBS (or other) surfaces that fit the data or worry (much) about tolerances as the data can be noisy.

Arthur,

Sure.

Allow a user to work on non-native imported surface geometry directly. Do this in the same way that is done with VX' Vision or even better with think3's Global Shape Modeling. SolidWorks needs to stop treating imported surface geometry like illegal aliens. Imported surface geometry needs to be elevated by SolidWorks to first class citizen status.

Only because some of the 3rd party tools are not robust enough. In some cases the needed tools don't exist at all.

As an example:

The ACIS Deformable Surface Modeler does not have the power of think3's Global Shape Modeling. The ACIS Deformable Surface Modeler is not even close to being a robust tool. In the past when I have pointed this out the comment from someone at SolidWorks Corp. (who I like) was that think3 spends too much money on R&D. This, in my opinion, is a total cop out.

FeatureWorks will *never* be the answer for badly designed software that can't work with non-native imported surface geometry. IMO, it should not be the only answer for dealing with non-native imported prismatic solids either.

I focus on what I know for a fact needs improvement. I have used and seen other systems that handle what I complain about in a much better way than SolidWorks does.

There are some areas I stay away from that in the past others have complained about extensively because I don't have the direct experience. An example would be large scale assemblies. I don't work with them so when others complain I read what they say and don't comment.

At the present I consider it lucky that I don't have to. I hope that in the future that SolidWorks is significantly improved where it so badly lags and if it is I would enjoy going back to using it and teaching it like I did in the past.

I think my posts over many years to this newsgroup speak for themselves about where I stand and why I feel the way I do.

IMO, so does this post. ;>)

jon

A totally clueless idiot sitting on a (claimed) old pile of free canned demo disks?

But he's more than happy to endlessly comment on what he knows nothing about (as long as there are buzzwords and ads involved).

Then we should get only dumb silence from you.

Do you use this in your daily work? No clues here.

This thread is not about making ppl feel like s*it....

Jon, one thing that you will find is that in life there is no 1 stop shop for everything. You do what workd for you. If global modeling works for you then so be it. And I will do what works for me. BUt you keep trying to beat into ppls head that global modeling can do everything and think 3 can do everything.....if that were the case then think 3 would be #1 and we would all be onboard with you. That is not the case, will not be the case...so I think that pretty much everyone has heard it before.

You have to have the right tool for the right job. And like I said before the right tool for the right job. If you havent noticed by now, no one to my knowledge has said "hey, jon is right....I will give up working in (insert program) to go with think3 global shape modeling". If ppl want it they will get it, the way that you come off is like you are a sales man or that you get kick backs for gettign sales of the program.

Arthur,

Right. It's not. It's also not about claiming that there is "no 1 stop shop for everything". Further, it's not about claiming that I'm saying that think3's Global Shape Modeling is the only solution as you are now claiming I have.

What it is about and what I stated above and have now cut and pasted for you to read again below is giving the SolidWorks user the proper tools to work with non-native imported surface geometry.

How about you read what I said again below and see if you can come away with what I really said rather than what is convenient for you to think I said.

"Allow a user to work on non-native imported surface geometry directly. Do this in the same way that is done with VX' Vision or even better with think3's Global Shape Modeling. SolidWorks needs to stop treating imported surface geometry like illegal aliens. Imported surface geometry needs to be elevated by SolidWorks to first class citizen status."

Please stop blaming me, Arthur because the proper tools are not in SolidWorks to properly deal with non-native imported surface geometry.

jon

Main Entry: buzz·word Pronunciation: 'b&z-"w&rd Function: noun Date: 1946

Main Entry: clue·less Pronunciation: 'klü-l&s Function: adjective Date: 1862

HTH

Jon,

What you are saying is that I should go out, purchase think3, start over with learning the program, all so that I can import geometry that is not dumb. UMMMMM, let me think about that, no. If you notice one thing that SW as a company good or bad does is incorporate better functionality with each new version. IMO that is what is setting it apart from the rest of the industry. Others programs have done this in the past, but none quite like the SW monster blob that we have here before us. Track record shows, like I said for good or bad, that the company is not content with standing still. Are they perfect, not by any means, is there room to grow, yes.

This brings me to what has happened with you, not just now, but in the past about what you have done. Someone will make a thread about something that they complain about with the SW program. Sometimes it is a gripe from waking up on the wrong side of the bed. Other times it is real and something that SW needs to really address and when that will happen is anyones guess. Jon I have seen you post about how, if SW cant do it then think3 global shape modeling can. AND...... That is what I am talking about, Jon. I am really not intrested in what that program can do. I have demoed it, had a bad experience with the company and personally I am not intrested. I have 3 tool sets that do me fine. There is NO more room in my head for anything else....litterally....if another program were to try and enter into my brain, it would explode.

There are certain reasons why SW imports somethings as dumb geometry. It would be nice for it to come in already parametric and everything recognized. Right now we have feature works which is there was of saying "hey there not perfect, but here is a tool that can help you get to your goal". Thats all.

Hi There.

I would not be shocked if this happened during an "automatic recognition" which in my unprofessional opinion is GREAT if you are recognizing cylinders and simpler things like that, but stumbles on more complex things.

The trouble I see with this add-in (when workiing automatically) is that it often gets in trouble when it has to decide what type of feature something is.

But, it would be unfair for me to say that I was "disappointed" with the automatic recognition because I understand its intrinsic inferiority of the algorithm when compared to the human mind, hence the real and practical need for an "interactive" mode. Some consideration has to be put into the "best" construction for a given feature, which the program can only do on a limited basis.

Since (to my knowledge) these dumb formats do not have a design intent tree built into them, the problem lies not just in the decrypting (our input), but in the creating (their output). Perhaps in the future we will all fly to work in our hovercraft and be able to have a neutral format that has an imbedded feature tree, paraametrically reconstructable by any system.

Kind Regards,

SMA