22 May 2008
Just one year after Siemens acquired UGS, Siemens PLM Software is firing on all cylinders
I returned from the Siemens PLM Analyst conference today and wanted to report to you that Siemens PLM finally announced, among other things, Solid Edge with synchronous technology and NX6, also with synchronous technology. Solid Edge is due for release to customers at the end of the Summer. While prices were not announced, I do not expect any price increase and it seems likely that existing customers will get the upgrade at no charge.
In the press release for Solid Edge you might notice a prominent quote from me. As reported in other articles on this site, I remain convinced that this modeling technology is a paradigm shift. Not needing a feature tree and being independent of modeling history makes models faster and easier to use. I spent about 8 hours with SE exercising the system in early April.
The base technology revolves around a concept of working with what we used to call “dumb models.” We called them “dumb” because they were devoid of their construction history. Yet, such models are actually very smart, and that is what SE with ST takes advantage of. SE with ST examines the model for “discovered” geometric constraints such as tangency, concentricity, parallelism, and the presence of holes, patterns and fillets. Holes, patterns and thin-walls are called procedural features. To control changes to the model, users may add driving or static dimensions in 3D and if desired, relate them using a variable table or Excel. In a history based system thin-walls are highly dependent on when they are computed so they achieve the desired result. Using SE with ST, thin walls are controlled by adding or deleting faces for the thin-wall command to compute when executing.
A “Live Rules” dialog window allows selecting the discovered constraints that can be changed during an edit. For instance, when selecting a planar face, all faces that lie on the same plane can be moved en masse, the same distance. A “steering wheel” with a directional axle controls the direction of movement in 3D. Sets of adjacent faces can be grouped and moved or copied to replicate existing geometry such as ribs. If they have fillets, the fillets will try to re-attach themselves to new tangent faces. To add or subtract material, users sketch on a face and, by controlling the directional axle of the steering wheel, can add or subtract material. Interestingly, models also appear to be self healing when deleting or moving features.
Significant changes can be made to the model topology when dragging and stretching faces. During my time with the system, the cases we examined worked. However, I did not have the time to deep dive into other than a few unplanned (by SE) changes. Until the system gets extended use by customers and more analysts, we cannot be certain whether there are unknown restrictions with this modeling technology. The paradigm appears to be sound, and as in the early days of history based modeling, there will assuredly be bugs to be solved and extensions to be added that will improve the technology even more.
If this turns out to be the paradigm change that I predict, users can be expected to accelerate SE and NX’s sales curve. Forget the 100X speed increase that SE promotes (it’s just for regen, which SE ST no longer needs), imagine how happy, and more productive users will be if they no longer need to model to anticipate future changes, and not needing to know how a model was constructed to be able to modify it.
What about NX 6 modeling technology?
Not having spent any time with NX, and seeing only a few short demos during the conference, I can only assume that it has the same abilities as SE. The only exception I observed, is that NX can fence off (using a bounding box) portions of the model to limit the change.
Associativity still maintained
Key downstream applications such as CAM and CAE use the b-rep of the model to perform their magic. This does not change. Frequently, however, both CAM (mold design) and CAE often simplify the model by eliminating certain types and size of features. ST’s ability to automatically heal faces should prove to be a huge benefit.
Importing data no longer an issue
Exchanging history data between history based systems has always been a problem. No commonly used standards have evolved fast enough to keep pace with the disparate history used by the CAD software vendors. On the other hand, the ability to import and even correct boundary representations (b-rep) of solid models has improved greatly. During the past several years, vendors such as Proficiency, have emerged to fill the gap. My contention is that the exchange of model data no longer requires history, even if users plan on changing the model! It may even be easier to change many models from within SE with ST than in their originating system.
About the competition
As similar as they may seem at first glance, neither SpaceClaim nor CoCreate can match the modeling prowess displayed by SE with ST. While both are non-history based, they lack the ability to edit models with the same alacrity. SpaceClaim, the newer of the two, can edit faces and sketch on faces, it lacks procedural features, the discovery of geometric constraints, and control of those constraints during edits.
---
|
