COFES-Russia Seminar – Plenary Session Speech

At the plenary session of the COFES-Russia Seminar I delivered a short speech about the round table discussion that I would be moderating later in the day. The purpose of the speech was to shed light on the topic of discussion of the round table which was “Intelligent Methods of 3D Modeling in CAD“. For those who are interested here is the speech:

“We have been using history based parametric modeling for the past 25 years. The thing with this way of modeling is that even before you start modeling a part or an assembly you need to create a strategy for it. This means that before you start creating geometry on the computer you need to have done most of it in your head. You need to plan how you are going to create your sketches, set up constraints, add relationships, etc. Then you need to start creating features, set up their parameters and figure out how the features will depend on each other. How much of all this really adds value? In the end only a few parameters really matter. But you still need to carry out the entire exercise whether you want to or not.

The same thing happens for geometry editing as well. If someone hands you a model, you look at it and you decide what you wish to change in it. For example, say you want to move a face. You are looking at the face and the face is looking back at you. But you can’t do anything unless you pick apart the feature tree and figure out what it is exactly that you need to edit so that the face moves like you want it to.

Some are of the opinion that history free direct modeling is the answer to this problem. But are the current solutions good enough? We have Inventor Fusion from Autodesk, the V6 CAD platform from Dassault Systemes, Synchronous Technology from Siemens PLM. And now on October 28th PTC is going to announce something called Project Lightning.

At the same time Direct Modelers are getting more intelligent. They are adding capabilities similar to the ones in the history based parametric modelers. So where is all this headed?

After all these years, it appears that we are still stuck with a geometry problem. So the question that I will be asking in the round table discussion is whether you think geometry is a solved problem? If yes, how good is the solution? If no, then how far do you think are we from the solution? But most important of all. Do you think we are going to hand this down to our kids? I mean will our kids be modeling like how we do today or are things going to change?”


  • Did you finish the speech before the hourglass ran out?

  • I believe I did. 😉

  • I don’t have a problem with History based feature modeling… It seems logical to me (Perhaps I don’t create complicated enough parts!). All my headaches stem from Assembly models and making sure that parts that are supposed to fit together do.

  • Vladimir Malukh

    Yes, he managed that 🙂

  • ksudavid

    When or how can we see the results of your round table discussion? Is this something that will be publicly available?

  • It was an interesting discussion. I’m busy traveling now and hope to write about it when I get the time.