Catia V6, Synchronous Technology and Experience
I was digging around for something else, and found this. It is about two months old, and is speaking in reference to Catia V6.
V6 is a brand new kernel, one that has capabilities that Florack claims to be unrivalled in the industry, being able to resolve Geometry and Topology simultaneously and being able to open files from pretty much any existing CAD system and edit it natively. V6 is claimed to remove many of these painful limitations that the CAD industry has suffered long and hard from.
It comes from this MCAD mag article by Martyn Day. The first part of the article is incomprehensible CAD doublespeak, but one phrase from this quoted block stands out – “and edit it natively”. Who knows what they are actually saying, but what I’m hearing is that the new Catia kernel is also on this direct editing bandwagon.
And of course the next question is why should SW continue to line the pockets of their competitors (UG/Solid Edge) by paying for the Parasolid kernel, when they could just integrate the Catia kernel? I think a lot of people have asked that question for a long time. The Parasolid kernel seems to be very effective, but Synchronous Technology is separate from Parasolid, and the Spaceclaim style direct modeling functionality sounds to be built into the Catia V6 kernel.
I think one of the stumbling blocks of the Spaceclaim/Synchronous Technology paradigm is going to turn out to be that it really only deals with analytical geometry – flat and cylindrical geometry. It doesn’t work on more generalized shapes required by NURBS. When they announce that, well, then I’ll be excited.
If we’re talking about Catia V6, we can’t just talk about direct editing speculation. There are other things as well that are important in V6.
It might have been 5 years ago when we heard Bernard Charles say that CAD interoperability was going to come about by all software working across all hardware and OSes. At the time it was seen as a dodge for getting around the notorious SW/Catia lack of interoperability. SolidWorks had a viewer version that ran on Java, but it was unusably slow. It sounds like part of the vision of V6 on a larger scale is to start to deliver on the interoperability using more modern tools. One way of overcoming interoperability problems is to just ignore the proprietary feature history, and use direct editing. It also solves the marketing dilemma of “why can’t SW and Catia share files?” and another huge one “why can’t SW07 read SW08 files?” One aspect of this direct modeling concept that people aren’t jumping up and down about is that the whole version compatibility issue simply goes away. If you have direct editing software, you are compatible with all versions of other software programs.
SaaS, software as a service, the ominous threat of delivering applications primarily across the web instead of having them installed locally, seems to play an obvious role in all of this. Even SW people have been making sure that those listening are hearing the SaaS message again and again. Do you really believe it? Do you really want to?
Anyway, the SaaS dream is being combined with the old virtual reality dream, and the word “experience” is being abused to the point I expect it to file suit. “If you build it, they will come”. Does this really apply to a “Second Life” sort of application for manufacturers?
SolidWorks Island on Second Life seems to be one of those marketing feelers to see if this was a great idea that no one ever recognized. SolidWorks Island is a lonely place where you can ride surfboards on the sand and watch videos of the real world while you sit by yourself in a virtual world. Anyway, not to say its a bad idea, maybe just that people aren’t quite ready for it yet.
I can see where the immersive “experience” might work out for things like planes, trains and automobiles, Catia’s forte, but probably not so much for the ventillation system, or a potato peeling machine. By now we’ve all seen the DS ad where people in the real world interact directly with virtual people and virtual stuff. Maybe this kind of thing is 50 years off. Maybe for some industries much less. It all seems to fall into the “2D is going to disappear” category of CAD salesmen’s wet dreams.
So, editability with direct editing is great, banishing the version problems forever will be great, but direct editing has a lot of questions to answer. It is certainly not going to replace parametrics, but is it really of much benefit when they are used together (a la Synchronous Technology), or will it be used separately, and side by side (a la Spaceclaim)? And who is ready for that really immersive experience?
Matt: “direct editing has a lot of questions to answer. It is certainly not going to replace parametrics”
There is a lot of work to be done. In my opinion, direct editing has just broken surface. As feature recognition technologies improve and are able to capture design intent, parametric modeling may wey well fade away. CAD vendors which adopt direct editing will have to realise that the one that does a better job at capturing design intent is most likely going to make more money. This will be a good enough fuel to keep innovation in this field going for a long time to come.
****
Well, I don’t believe feature recognition can capture any sort of intent. All it can do is recignize existing conditions and assume you want to maintain existing conditions. Conditions like tangency, concentricity, parallelism… which is really just parametrics applied directly to the faces of a solid model. So I think the ST stuff is going to reinforce parametrics. It might take a bite out of history, but not parametrics.
I’m not really a big believer in “design intent”. Design intent is only good at making the kind of change it was built to make. Any other kind of change requires you to remodel the design intent itself. The concept is too difficult to articulate and define properly to be very valuable, really.
I sure hope they bury the SAAS software as an online service idea. It would ruin my working and cruising lifestyle. Satphone bandwidth is only 4800 baud and costs about $1.19 per minute. Life proceeds at a slower pace when I am cruising but not that much slower. It is possible to work in remote parts of paradise. There is room for more of us to do it.
****
I’m pretty sure SaaS is going to be limited to these little applets that we see. To get websites with dynamic content that stays up to date, they have to use some sort of programming instead of static html writing. I don’t think SaaS is going to go into full blown CAD software. It would solve the platform problem. The browser becomes the OS, and browsers are easier to port to different platforms. Graphics? Who knows. DirectX is MS, and OpenGL is open source. I hope MS doesn’t win the graphics control game.