ST in NX: It’s not about Direct Modeling
For me, today started like just about any other day. To get up to speed, I check my stream of tweets, LinkedIn updates and Facebook updates from people I’m following as well as the RSS feed from the blogs I follow. That’s when I saw a blog post that caught my eye, titled summarizing PTC’s Decades of Fervent In-House Innovation (and Acquisitions) – Part 1 from P.J. Jakovljevic. It’s a pretty good overview of PTC’s activities in the product development space since their founding. And that’s when I came across this piece of the post.
Until recently, Siemens was the only major CAD provider with a unique flavor of direct modeling with its Synchronous technology (now in its third version). The Synchronous technology, which manifests itself in both Siemens’ NX and Solid Edge CAD products, combines the convenience of direct modeling with a greater ability to import data from other systems more openly. At its core is an intuitive inference engine that anticipates the intent of the designer, reportedly making design and/or modeling up to 100 times faster in certain situations.
I had to smile to myself. That’s because the points that P.J. makes were exactly the same perceptions I had. After I sat down with Paul Brown, who heads up NX marketing, and Mike Rebrukh, who leads NX Product Management, I realized I had it completely wrong.
NX has actually been around for quite some time. For a relatively complete history, you can check out the NX wikipedia entry. But here’s the quick story. The first aspects of what is now NX was originally created in 1969. After UGS acquired SDRC many years ago, the first started merging I-DEAS into NX in 2002. And then in 2007, they first started integrating Synchronous Technology into NX.
“The first thing you need to think about ST for NX is to forget about history-free modeling”, says Mike Rebrukh.
My first thought is: what? There must be some misunderstanding here. After all, the latest rage in the CAD industry is to get away from history based modeling. It requires a lot of knowledge and skills to manage features in an ordered history. Knowledge and skills that casual users and even expert users struggle with. So I start to ask questions and pose statements testing what exactly Mike means. Fifteen minutes later, I think I have a handle on it. And, to my surprise, Mike is right. ST in NX has little to do with history-free modeling. But the end game is the same. Let me explain.
A More Intuitive Way to Manipulate Existing Features
One means of using ST in NX is to manipulate existing features. It happens like the following.
- You grab or select a piece of geometry like a surface or edge.
- ST is intelligent enough to infer what other associate geometry should be moved with it.
- You can use Direct Modeling modifications like move, rotate, etc. to change that geometry. Additionally, you can add constraints such as making two surfaces co-planar.
- ST, behind the scenes, changes the parametrics and dimensions of the individual feature or even set of features that made up that geometry. If you’ve added constraints, behind the scenes ST will add those constraints to the feature or sets of features that were affected.
What’s this mean? There’s a few implications. In the past, you would have to modify one or a few features through dimensions and then replay the model. And in some cases, depending on how sensitive your model was to change, you would have to modify multiple dimensions in just the right way to avoid a feature failure.
With ST in NX, you’re basically locally modifying a number of feature dynamically as you push, pull or drag geometry. It’s all being verified on the fly.
Less Worrying About Feature History
To make this point, Mike and Paul showed me a quick example that made the point quickly.
- They had a model that had rounds at the bottom of a pocket cutout.
- They selected the side surfaces, the ones that transitioned into the round, and added drafts.
- They then showed me the new draft feature in the history tree.
Now typically, to add drafts to surfaces that already have rounds on them, you would have to roll back the model, add the draft and then un-suppress the rounds. That’s because the draft feature changes the geometric reference, the edge at the bottom of the pocket in this case, that is used to create the round. That’s why the draft feature has always had to come before the round.
With ST in NX however, it automatically uses the reblend functionality from Synchronous Technology to automatically recreate the proper geometry. It is intelligent enough to help the user avoid the manual management of features in this case.
Direct Modeling Operations at Features
There’s one more capability that I found interesting during our discussion. The ST in NX implementation allows users to persist Direct Modeling operations, such as move, remove face, etc. as features in the history tree. This capability isn’t as remarkable as the others I saw during our discussion, but I believe it does provide value. There are instances where it makes sense to have a Direct Modeling operation in a specific place in the overall order of the history tree.
A Point of Clarification
I started this discussion with a quote from Mike saying to forget about history-free modeling when it comes to ST in NX. But to make sure it is clear, the capabilities of ST can be used in the history-free mode of NX as well. It is just that most of the value of ST in NX can be realized in history mode.
Analysis and Commentary
So far, I’ve written a lot about the capabilities of ST in NX. But at the end of the day, what does it all mean? Here are my technical takeaways.
- There is a lower risk for feature failures in history models due to the new means of modifying and adding features.
- There is a way to add greater design intent without the complexity of traditional CAD by adding constraints without redefining features.
- It is far easier and simpler to build models without manual activities like rolling back the model to a certain feature because ST intelligently adds features in the history tree.
- There is an entire new set of geometric controls that adds another layer of intelligence in terms of feature order, offering more options to build intelligence into the model.
And ultimately what does that translate to?
- It’s faster because you don’t have to manipulate the feature order nearly as much.
- It’s simpler and easier because you won’t have as many feature failures.
- You save time and effort because models have more design intent, reacting to change more intelligently.
Going back to the original context of the current rage in the CAD industry, the ultimate driver of history-free modeling is to make CAD easier to use by avoiding the complexities of features in a history based model. But after seeing how ST works in NX, I believe much of the same purpose has been achieved, albeit in a different way. Where history free Direct Modeling lets you do away with any order or history at all, NX uses ST to manage that complexity for the user through a number of powerful capabilities. Different approach. Same end game.
With all that said, however, ST in NX is focused in terms of who it will enable. Many other CAD applications are employing Direct Modeling and history-free modeling capabilities to enable many casual users such as product engineers, manufacturing engineers, service engineers and more who do not have the time, knowledge or skills to become CAD experts. If you’ve listened to Dan Staples talk about ST in Solid Edge, you’ll know that he steadfastly is driving ST towards enabling the CAD expect and not the CAD casual user. The NX folks have a little bit of a different vision. They see ST in NX as enabling both history-free and history-based CAD modeling by any type of user, including engineers. But it is still early on with Siemens PLM still filling out this vision.
Conclusions and Questions
ST in NX isn’t what I expected. And I believe there’s a misconception that ST in NX is just another Direct Modeling tool or history-free modeling tool. But that couldn’t be further from the truth. The most value ST in NX lies in enabling the manipulation and creation of features in a history based model. It also offers new means of adding design intent and new feature types. All that provides the CAD expert with more powerful tools to do their job faster and better.
That’s my take. What’s yours? Do you believe new technologies like ST can be used to address the traditional shortcomings of history based modeling? Do you think there is value in further empowering the CAD expert? Sound off and make your voice heard. I know I’m interested in hearing what you have to say.
Take care. Talk soon. And thanks for reading.