67 68 69 70 71 72 1,042 Replies Latest reply on Mar 23, 2016 2:39 PM by David Tiefenbrunn Go to original post
      • 1,020. Re: SOLIDWORKS 2016 User Interface
        Ryan McVay

        Matthew Peterson I knew I was going to get a bunch of flak on that posting. You are asking all the right question! Thank you, finally!

         

        Yes, I totally agree that there are a LOT of buzzwords out there and I feel they were used to confuse the users in the industry. Honestly. Take a look at all the blogs around 2012-3 time period! Siemens was touting synchronous technology, PTC was introducing CREO, DS was introducing their new "CATIA-lite" (now 3DEXPERICENE) and Autodesk was showing off Fusion360. They were all based on different technologies and different levels of "direct editing".  The biggest offender was the term direct editing or direct-editor. All these organizations has different solutions but lumped them together to basically confuse you and say, "We have direct-editing" capabilities.

         

        When it comes to the last 20 years of CAD, yes we all grew up with promises of parametrics and found that promise came with the caveat of managing parent-child relationships and linear solving (aka history-tree) and first building a strategy on how to build a part so that it reacts to a possible future change- which may or may not ever occur. Linear solver works by first solving the base feature and works its way down the history-tree- solving one at a time. There was CPU processing/performance issues that drove this technology. This is all ancient technology based on slower CPUs. We've come a long way since 1998 in those regards!

         

        So, to say that any useful CAD system has to have these functions doesn't hold much validity these days.

         

        See most people don't know the newer systems are based on simultaneous solving of the geometry (history-free) and the ability to send your parameters directly to the model faces (legacy -child) instead of residing in the 2D sketch (legacy- which is the parent). In the newer systems you can change the models by direct editing (pulling on controlled geometry and the 2D/3D solvers are solving the geometry constraints- ether implied or applied- simultaneously or make planned and precise changes by changing the parameters (dimensions) on the faces. The newer systems have broken the parent-child relationships between the 3D faces and the 2D geometry hidden away in sketches. You even have systems that allow you to store your solid definition into a database and then build a configuration that pulls different "elements" of the model together and solves to create your new 3D configure part. Other systems allow you to apply or work you data in both a history-base/ordered and history-free/direct editing environments- IMHO the best of both worlds.

         

        I hope you can see that parametrics does NOT equal parent-child relationship. Parametrics means a dimension or rule driven value assigned to geometry- now it can be 2D or 3Dgeometry. History-based systems are built on old technology constraints and rely on parent-child relations to break the solving of the geometry into "manageable" chunks- solve 2D sketch constraints then update 3D faces separate functions separate steps.

         

        I truly hope this helps and thank you for your questions.

        • 1,021. Re: SOLIDWORKS 2016 User Interface
          Matthew Peterson

          OK thanks that helps a lot.

           

           

          I hope you can see that parametrics does NOT equal parent-child relationship. Parametrics means a dimension or rule driven value assigned to geometry- now it can be 2D or 3Dgeometry. History-based systems are built on old technology constraints and rely on parent-child relations to break the solving of the geometry into "manageable" chunks- solve 2D sketch constraints then update 3D faces separate functions separate steps.

           

           

          I am not giving up on the parent-child == parametric though. You have changed the definition of parent-child to parent = sketch child = 3d geometry. That is not my understanding.

          My take on what  "direct-editors" have done is moved the "parent" to "other 3d geometry".

          A "rule", as you put it, by its nature requires inputs (parents) that must be "solved" before the "result of that rule" (child) can be resolved. This is also linear in nature.

          So maybe what so called revolutionary direct editors have tried to do is to understand what is and what is not necessarily required to be a child and thereby take advantage of parallel processing?

          • 1,022. Re: SOLIDWORKS 2016 User Interface
            Chris Clouser

            I'm not sure why SWX is stuck.  Why can't it become a hybrid somehow?

             

            They must realize that they must move forward, and they must also realize that we need to be able to retain usage and functionality of the billions and billions of dollars of legacy data throughout the entire user base.  Unless we just jettison all of our investment in SWX like we did with AutoCAD.  But leaving 2D behind was an easy decision.  Moving on to a new 3D modeling platform will be painful.

            • 1,023. Re: SOLIDWORKS 2016 User Interface
              Ryan McVay

              Matthew Peterson The "rules" on how to solve changes and how the data is stored and "rebuild" is what makes the difference in all these different tools. There are many rules that get applied and how they are applied (manually or automagically) is what makes these new tools "modern" and allow for simultaneous solving. It's how the systems are solving and identifying geometry and the time of editing that is the true magic and behind a bunch of technology patents.

              Sketches are used to define the base geometry. Dimensions and "geometric constraints" are then passed directly to the 3D geometry. Once the 3D data is generated you can either toss the sketch or keep it and reuse it to create other geometry. Now- that is how the "direct edit" functionality works. Most people will find they need both history and direct edit functions to control things like the order of blends, etc. That is next level of these tools! It's all there, now.

              • 1,024. Re: SOLIDWORKS 2016 User Interface
                Ryan McVay

                Chris Clouser That is exactly what DS is telling you. If you don't know the stories behind CATIA V4, V5 and V6 you might want to read them! Very enlightening! DS has moved forward with a new products on the 3DEXPERINCE platform.

                • 1,025. Re: SOLIDWORKS 2016 User Interface
                  Matthew Peterson

                  EXACTLY!!!

                  There are billions of dollars into these "history trees".

                  And in MOST industries that I have been a part of.
                       "Managing parent-child relationships and linear solving (aka history-tree) and first building a strategy on how to build a part so that it reacts to a possible future change- which may or may not ever occur."

                  should be:

                       "Which more often than not DO OCCUR."

                  In many ways the design intent is much more important than the geometry itself.

                  • 1,026. Re: SOLIDWORKS 2016 User Interface
                    Al Griego

                    You're right. Moving to a new platform is a painful process. It wasn't so bad going from AutoCAD R12 to Unigraphics version 10.4, but it would be a lot harder to go to another solid modeling package. Probably the worst part about changing over is having to remodel standard parts. Part families for fasteners and such make it easier, but it's still a massive pain.

                    • 1,027. Re: SOLIDWORKS 2016 User Interface
                      Matthew Peterson

                      The multi-billion dollar question for Dassault is:

                      10 years from now, will a part I draw today in SolidWorks (a part whose design intent is VERY important) have my same design built into it, and if so, will I be editing it with a fully supported and cutting edge piece of software?

                      I don't care if the software is not SolidWorks, but if the geometry is a dumb solid then Dassault has failed miserably.

                      • 1,028. Re: SOLIDWORKS 2016 User Interface
                        Ryan McVay

                        I can't answer that. The only thing I can do is to point you to the past actions/decisions that DS has done. And then let you extrapolate how you think they would move forward. Read up on CATIA V4, 5 and 6 issues.

                        • 1,029. Re: SOLIDWORKS 2016 User Interface
                          Matthew Peterson

                          I sincerely hope that the engineers at Dassault and SolidWorks are working their a-words off for a full history tree conversion tool for SolidWorks legacy data to the next true DS platform.
                          Because if they leave us stranded on a dying platform I will do my absolute best to ensure that I and any company I ever work for never does business with them again.

                          • 1,031. Re: SOLIDWORKS 2016 User Interface
                            Matthew Peterson

                            The multi-billion dollar question for Dassault is:

                            10 years from now, will a part I draw today in SolidWorks (a part whose design intent is VERY important) have my same design built into it, and if so, will I be editing it with a fully supported and cutting edge piece of software?

                            I don't care if the software is not SolidWorks, but if the geometry is a dumb solid then Dassault has failed miserably.

                            It really is a multibillion dollar question. We'll look at some of Dassault's past experience in Kernel upgrades. As suggested I've been reading up on Catia's history. From Wikipedia:

                                 Conversion from CATIA Version 4 to Version 5 created construction problems for the Airbus A380 which caused up to $6.1B in additional costs due to years of project delays when aircraft wiring was too short to make connections.[3]

                            • 1,032. Re: SOLIDWORKS 2016 User Interface
                              Chris Clouser

                              HA!  I swore I'd never do business with Autodesk ever again in the late 90's and then they bought the version of NASTRAN I'm using, decreased the support, increased the difficulty to get it installed and running, and doubled the annual maintenance from $800 to $1700.

                               

                              Also, they make SWX's full-backdating racket look pleasant because now with Autodesk products, if you are late on your maintenance (even a day), you lose your perpetual license.  It will still work, but you can never again upgrade, you have to transfer to the subscription-based plan at a cost factor of about 1.7 times the over inflated annual maintenance cost.

                               

                              Dassault, I don't know what this 3DExperience mess is.  I don't care.  My 3D experience is SolidWorks.  FOCUS ON THAT!  Do whatever it takes to keep SWX state-of-the-art.

                               

                              Quit spending my money on stuff that doesn't help me!

                               

                              I don't know about Catia either.  I learned it in '05 and it was a complete piece of CRAP.  SWX ran circles around it.

                               

                              You're probably making way more money off of SWX, so make that your flagship and Catia can be for those unfortunate souls that need to work on a regular basis of over 50k parts.  Or fix the few issues SWX has and s-can Catia all together.

                              • 1,033. Re: SOLIDWORKS 2016 User Interface
                                Paul Salvador

                                well,.. since it's Friday.. thought I'd give it a go,.. and, I used my HolyCubes,....

                                holycubes-zxys-preview.pngholycubes-zxys.pngholycubes-zxys-a.png

                                • 1,034. Re: SOLIDWORKS 2016 User Interface
                                  Ivar Kjelberg

                                  I can only agree

                                   

                                  In your early days you were chasing the clicks by adapting the menus to the minimize the average path we users did, but in the later versions, probably with new programmers, it is gently but surely going the other way ...

                                   

                                  I dropped HP ME30 and Unigraphics too heavy, a few decades ago, because SW came with fresh air easy quick access ...

                                  Today I have the feeling you have become the dinosaurs, and the newcomers are the new SW ...

                                  67 68 69 70 71 72