Bertrand Sicot On SolidWorks V6, CGM And Parasolid

Yesterday I blogged about my thoughts on SolidWorks V6, CGM and Parasolid. That led to an interesting discussion, which at the time of this writing, sits at 28 comments. One of my readers asked me whether it was premature of me to talk about the architecture of a future version of SolidWorks and wondered whether I had any inside information on this topic. Although my company is a SolidWorks Solution Partner, at this time I do not have any inside information on what SolidWorks V6 will be like or how the current Windows version of SolidWorks is going to evolve. These are merely my thoughts on the subject. Time will tell whether I am right or wrong.

But you see, I don’t have the patience to wait for time. So I decided to ask the one person who would be able to tell me right now – SolidWorks’ new CEO Bertrand Sicot. This is what he wrote to me:

“We are developing SolidWorks V6 to be a next-generation design application that has truly innovative solutions to problems that are faced by designers every day. We have already demonstrated many of our ideas in the design preview shown at SolidWorks World 2010 in Anaheim. You have hit the nail on the headโ€”this is our opportunity to really get it right for our users, with new products built for online, desktop and mobile access. The Windows-based version SolidWorks that our customers use today will continue to be improved and enhanced for years to come.”

So there you have it. Straight from the horse’s mouth. The current Windows-based version of SolidWorks will be improved and enhanced for years to come. How much it will be enhanced and for how long is anybody’s guess. I don’t think even the folks at SolidWorks know today. I guess they will let the circumstances decide what happens to the Windows-based version of SolidWorks.

As regards the Parasolid modeling kernel, the huge royalties that SolidWorks pays Siemens PLM today will dwindle as users start adopting SolidWorks V6, which we now know is powered by CGM. At some point SolidWorks will merely be paying Siemens PLM royalties for Parasolid Communicator, the technology used to access Parasolid data, just like how every other CAD vendor is doing today.

This, I’m certain, is going to happen.

  • ron

    “…truly innovative solutions to problems that are faced by designers every day.” Reads like what PTC said before Creo was launched.

  • JoeAggie

    Thank you for sharing Mondemoiseau Sicot’s email. It appears that the folks at SW understand the “problems that are faced by designers every day” and I am very interested to know what those are !
    Deelip, perhaps you and your readers can share thoughts on such problems faced by designers (focusing on mechanical design only).

    • If you know what I do for a living, you will understand that the most common problems I see are related to data exchange.

      But like I say to my customers, I am just the person providing the tools. You are the people doing the work and I want to know what drives you nuts. Because I stand to make a lot of money doing my bit helping you maintain your sanity. ๐Ÿ˜‰

      • anonymous

        I enjoy your blog, and your true technical ability and experience add weight to it vs. other CAD blogs out there.

        But, I find it striking that someone with formal “partner” arrangements with the CAD companies would be willing to expound and offer forth conjecture on product futures…I don’t mean to imply that you have disclosed information that you should not be…but aren’t you worried at all about doing something like that, even if it happened accidentally?

        I really don’t mind either way, as I don’t bear the risk and benefit from whatever you post…but how to you straddle this fine line?

        • Oh yes. I worry about it a lot. That’s why I need to go through whatever I write and check that I’m about to do something stupid. But in this business, ignorance can be bliss. If things don’t affect me in the immediate future, I prefer not knowing them. That way I can speculate and write the things I do.

          This is an opinion blog and it would be pretty boring if I didn’t actually get the chance to express my opinion from time to time, now wouldn’t it? ๐Ÿ˜‰

        • JoeAggie

          If you follow any of the MSDN forums (Microsoft Developer Network), you will see that it very common for software developer partners to offer their perspective on what future would/should be. Their latest Azure platform is benefitting from such forums. Why should it be aby different for CAD software.

      • JoeAggie

        In my short experience in design world both as a designer and the one leading a team of designers, the most frustrating problem is making a small change in a part and messing up the entire assembly ! Don’t laugh, but I have seen it so many times when a small change is done on a part and the entire assembly goes phhtttppp. This problem is often attributed, by CAD tech support folks, to user error or that the user needs training on how to use tool correctly. *When a small change in the design that is expected to take less than 2 hours takes 2 days, it is a problem. *
        And when I say short experience, I started on 3D CAD right about the time SW came out first.
        And problem I am describing is not just in SW or ProE or NX. I have used (and use) all three tools directly and indirectly through my team.

        • Davide

          agree 100%
          the biggest problem for any MCAD user is to make changes that appear “logic & easy” to our mind and face huge unexpected problems due to how the changes are propagating in the assembly structure.
          This is an old story and I do not see big improvements since PTC era. Direct editing can relief the pain but, in my opinion, it is not the final solution because it does not convey design intent.
          There is a long way to go…

  • Anonymous

    I think I am going to like this guy. He is clear and says the things that make me feel better about Solidworks. The cloud sucks, but I am the only designer who works remotely with no, slow or unreliable intenet access. This is to be expected when I am cruising on my boat among desert islands. Then again, my office access is not so good.

    All designers want control of the geometry of a part. The relational attributes of Solidworks are most helpful to allow a top down design where high level changes can propagate to many parts. Prismatic machined parts seem to be adequately addressed by most 3D CAD programs. These are typically defined by fully defined sketches mostly on principle planes. We also need some easy fast threaded holes and rods.

    Surfaced parts are the big challenge. These parts depend on Solidworks doing something reasonable between defined profiles or guides. It is essential that lofts, sweeps, and boundary surfaces make perfect smooth and continuous contact with the defining curves. If a loft can be a ruled surface make it ruled. If a loft has sections with parallel edges create the shape that would maintain that property in intermediate stations. Edge conditions of tangency and continuous curvature can be visually sensitive and are fussy. Curves or sketches that are projected on surfaces should maintain all of the geometric properties and functions. Intersecting curves must intersect. Sketched tangent curves should retain the directional sense of the tangent, if it is drawn as a radiused corner it will remain a radiused corner not change into a cusp. Surface trims need to reliably trim the correct part of the correct surface. It would be best if all dimensions are stored exactly as entered. Surfaces that trim each oher should be intersecting at a curve. Do not pretend that accuracy is 10uM when I get 1/4 inch errors. Shell and offset surface need to be sensible and accurate where features intersect.

    All designers want control of their CAD data. I often save a part as it progresses in files with distinct names. Sometimes I go back to a previously rejected idea. Sometimes a crash corrupts the file. I want potential damage limited to the one active file. Sometimes an entire subassembly must be duplicated as parts are revised. I do not want the previous version at risk. Solidworks files are a big fat mess. The essential data to define most of my parts is 1000X smaller. The rebuild time is not very long. Let me have the local compact form.

    Most designers want to work with a known good tool. I just hate surprises that come with SP0. Let me choose which version that I want to use. Do not force “upgrades”. We know that all software has many bugs, the bigger the code the more bugs. Solidworks currently can be provoked to fail with almost any kind of feature, users learn this and avoid the trouble spots. Fortunately there is more than one way to define a geometry.

  • Guest

    JoeAggie, You need to have a look at Synchronous Technology, NOW.
    This was the precise reason i have lately switched to it.

    • JoeAggie

      Guest, yes NX 7.5 is promising ….

  • Anonymous

    You are describing issues that are inherent with History based CAD systems. Why should your CAD system have to keep a history of how the part is built? This is a throwback to a time when computers were slow and had little memory. This is in the DNA of History-based CAD systems.

    Don’t you really just need to have the ability to recognize and change the geometry? Or Maybe you want to see how the geometry,PMI or assembly has been changed (ECO) automatically? Technology is available today to do this.

    Small changes or large changes to parts are not a problem in Direct Modeling systems. Designs are always going to change, people need the ability to make changes without having programmed the potential change in advance. Why are users so tolerant of these issues? There are many good solutions available today.

  • Jon Hirschtick

    Deelip you mention “huge royalties” as a factor — not so. Not a cost issue. Keep up the great writing. I don’t always agree with what you write, but I always respect you for taking opinions and backing them up. See you at SolidWorks World in San Antonio!

    • Jon, my reference to “huge royalties” was not related to it being a factor that drove SolidWorks V6. I was merely mentioning a substantial loss of business for Siemens PLM

      • Neil

        Any ball park idea how much they would pay to use Parasolid, Deelip?
        Would they have to license it say for 5 years at a time?
        Because DS have now stated SW will be “improved and enhanced for years to comeโ€ I guess they are stuck with paying for licensing for a while yet and regardless of how many or few users there are.
        I did actually ask on the SW forums where the same email quote appeared how many years the ‘years’ referred to by Bertrand is, and what level of improvements to expect, but of course you never get answers to questions like that ๐Ÿ˜‰

        • I wouldn’t know the numbers even vaguely. Anyways, royalties are usually paid per seat. At least that’s what I do. SolidWorks and Siemens PLM may have a different arrangement. So if the number of active seats of SolidWorks 20xx go down the royalties will do down as wll, irrespective of the number of years the contract has ben signed for.

          • Neil

            Yeah ok, thanks. I was thinking Siemens would probably make DS pay a good premium for renewal knowing they need it to keep customers happy in the meantime but also that its being eventually terminated. You would want to extract what you could I think.
            BTW I also asked on the forums if they had reassigned their coders so that both SW and SWV6 were getting the same attention rather than concentrating on the cloud. Its another of those searching questions that get conveniently lost on corporate ears…
            Probably Jon reads great writing there too and knows the answer. Perhaps you can corner him at SWW and get a SW in retrospect piece out of him.

          • You make it sound like I am a badass interrogator. ๐Ÿ˜‰

          • Neil

            Well I cant be there myself…

  • Everyone one is talking about CGM and Cloud. Why SolidWork is not talking about how new product would be different in modeling and design terms from SolidWork 2010.

    I think the reason is if they do that then it would be just show what is wrong in current SolidWorks and which of their competitor is doing it in right way !!!

    New SolidWork V6 is no way going to be CATIA light. As ex-developer of SolidWorks I know SW team would definitely not like to have that. Ex-CEO John Mceleney always said how they don’t want to go CATIA way.