I watched Creo presentation on October 28 and I should say that I was a little bit disappointed with the PTC strategy for the next decade (as they announced it). They are promising to revolutionize CAD market with Creo but what they demonstrated is looked more like evolutionary changes to the existing applications. In any case, it can’t be compared with the real 3D modeling revolution initiated by Parametrics 25 years ago.
What I expected from 3D modeling revolution is, first of all, the long waiting integration of 3D parametric modeling and 3D direct modeling approaches in one powerful and easy-to-use application. What PTC is promising us in Creo is the common data model for Pro/E and CoCreate, but not the common workspace for both parametric and direct modeling. Pro/E and CoCreate remains to be separate applications (only names are to be changed) that can only exchange data with each other. Not too much for New CAD Revolution.
As for 3D parametric modeling itself, it remains more or less the same as it was proposed 25 years ago. All “parametrics” is concentrated mainly in 2D sections (as for me, I would like to see a full freedom of parametric constraints in 3D space) and the capabilities of this 2D solver is rather restricted – it can resolve sections with about 2000 entities and fall down for more complicated sections.
To my mind, the poor solver of Pro/E is the main reason why PTC is not able to implement a real integration of Pro/E with CoCreate (instead of this they explain us that it is much more convenient for users to work with Pro/E and CoCreate as separate applications). Actually, this integration should be the main direction for the New CAD Revolution. In order to provide such an integration one should develop a solver that could work with hundreds of thousands of entities and resolve corresponding systems of equations in real time.
Can we expect to have such a solver in Creo? As far as I understand, Creo doesn’t promise any progress in this direction.
Can we expect any other progress in computational aspects of the parametric approach (for instance, using advantages of multi-core processors)? It seems to me that the answer is “no” again…
I’m rather skeptic about the New CAD Revolution if inside Creo will be the ancient parametric engine that was developed for Pro/Engineer 25 years ago.
You have very good points.
I don't know if I understood things incorrectly, but I though the CREO Kernel was written from the ground up and it's not the same Kernel PTC wrote 25 years ago.
I sure hope the CREO Kernel is new and written for Windows. I can't stand the "Designed for Windows" label on Pro/E when it's clearly designed in Unix and ported to windows. I hope the CREO code is all brand new using a new and modern Kernel.
I used Pro/E for the first 2 years of my use ('96-'98) and I can tell you the Unix version was FAR more robust. I hate Windows anything, and woukd tell you that it's a much better platform for serious 3D modeling and rendering.
I am all for Windows 64-bit moving forward. In the end the OS that is supported is the one I like to use.
I used UNIX with Pro/E way back in the day.
Reading my comment above. I wanted to provide an update. Creo uses the same Engine (Kernel) but it's built on a new platform. It's like taking the engine out of a Toyota and putting it into a Lexus. Same great engine as before with a new body, interiror, options and modern look. 🙂
I'm not sure that you would be so happy were you buy today a Lexus, look on its engine and catch out that the engine inside is the Toyota engine produced in 1985. Even in Toyota Corolla 2012 the engine is quite different from Toyota Corolla 1985.
I had a chance to play with Creo Parametric and I didn't notice significant difference between the solver of Pro/ENGINEER and that of Creo Parametric. May be there were some small improvements but, I'm sure, it is the same code written many-many years ago. And it is far from being perfect. Even for sketches that are not so complicated solver is unstable, slow, and requires a lot of memory.
Look, for instance, the video presented on http://www.cloud-invent.com/Cheetah.aspx. I have a feeling that soon parametric CAD will start using quit different parametric "engines" and, if PTC will ignore this challenge, it can loose it's leading positions on the CAD market.
I have seen the idea behind Cheetah and I agree that a new solver is something that is needed by all the major CAD providers. Not sure if this would be possible due to the market share they have and the investment in the current solvers. Having a nice solver is far from having an actual CAD program that works well, but it is a nice tech demo of a solver.
The solver for Pro/ENGINEER and Creo Parametric should not show any difference because they use the same one. Granite is used for both, but they can improve performance by tweaking things like they did with Creo Parametric for Assemblies.
By the way. The Lexus analogy was just that. I do own two Lexus vehicles or "Lexi". They are fantastic vehcles and I am very happy with both of them. 🙂