Why Solid Edge matters, part 1: A little history
Siemens PLM Software has announced that Solid Edge ST5 will be available soon.
Is this just another update of another mainstream CAD package? Yes. But Solid Edge matters more than you may realize.
Let me start with a little history of Solid Edge. The product started out as a part of the Jupiter project, a next-gen CAD initiative at Intergraph Corp, in the early 1990s. If you’re old, like me, you may remember that as the time when Microsoft was working on Windows NT, their 32-bit version of Windows. NT promised to be a game changer (and it was.) For CAD vendors, it heralded a shift from Unix-based engineering workstations to Windows-based PCs.
The Jupiter project was ambitious. The goal was to develop a platform that could be used to create a number of graphics applications, including 2D CAD, AEC CAD, mapping, P&ID, and more. To me, the jewel of Jupiter project was Solid Edge: a 3D feature-based solid modeling MCAD system. I was one of a number of industry writers who was flown out to Intergraph’s Huntsville headquarters to get an up-close look at Solid Edge.
My sense, at the time, was that Solid Edge was pretty impressive. Yet, Intergraph wasn’t the only company working on a Windows-based MCAD program. SolidWorks was also hard at work on a similar product.
The SolidWorks versus Solid Edge competition was shaping up to be a major brawl. Both were clean-sheet products that promised to challenge the hegemony of PTC’s seemingly unassailable Pro/Engineer CAD program.
Unfortunately, Intergraph made a number of missteps with Solid Edge:
- They built Jupiter to be a universal CAD platform. This added a lot of program overhead that made Solid Edge fatter and slower than it would have been were it to be fine-tuned for MCAD.
- In the middle of development, they went back and re-worked the core program code to use the (newly available) Microsoft Foundation Class (MFC) object model, instead of their internally developed object model. This caused a year’s delay. As a result, Solid Edge didn’t ship until about 6 months after SolidWorks.
- They used the ACIS geometric modeling kernel, which, at the time, was not a match for Parasolid.
- They didn’t have their sales channel strategy together.
The competition between SolidWorks and Solid Edge was more of a blowout than a brawl. Watching SolidWorks take on the competition was like watching Mike Tyson (in his prime) take on other boxers. Not only did SolidWorks soundly trounce Solid Edge, they pummeled PTC, and bloodied Autodesk.
Ultimately, three major changes saved Solid Edge: the developers pulled out all the extraneous Jupiter overhead that wasn’t necessary for an MCAD program, and changed from ACIS to the Parasolid geometric modeling kernel. And they sold the product, and its entire development organization, to EDS Unigraphics (now known as Siemens PLM.)
For many years, the Solid Edge team (which Unigraphics seemed to have left largely intact) continued to develop new and improved versions of the software. Over time, Solid Edge matured into a capable mainstream CAD system. Able to compete with SolidWorks, Pro/E, and Autodesk Inventor.
In 2008, the game changed for Solid Edge, with the introduction of Synchronous Technology: a system supporting both traditional history-based (ordered) modeling, and history-free direct (unordered) modeling. I was (and still am) quoted on the Siemens PLM website:
Evan Yares, CAD Industry Analyst
“Siemens’ synchronous solver overcomes the order dependencies that have plagued history-based CAD programs by solving for the explicit and inferred constraints at the same time. The synchronous solver doesn’t use a history tree, but rather holds user-defined constraints in groups associated with the surfaces to which they apply…Ultimately, though, I believe this to be a transformative technology — one that represents an important inflection point in the CAD industry. If you hear someone say ‘that’s nothing new,’ don’t believe them. Synchronous Technology is a big deal.”
I’ve been guilty, more than once in my career, of seeing what’s possible, rather than what’s immediate.* I was, possibly, guilty of this with Synchronous Technology. The first version of Synchronous Technology had some limitations and usability problems. It took until version 3 (ST3) for it to really cross the threshold of “good enough.” This week, Siemens has announced Solid Edge ST5, the fifth version of Solid Edge incorporating Synchronous Technology.
Don’t count on me reciting all the changes that are in ST5: There are over 1,300 of them. While there are some great improvements in the software, my take is that the most important thing about ST5 is the continued refinement of Synchronous Technology.
With ST5, I reiterate my statement that Synchronous Technology is a “big deal.”
In the next article in this series, I’ll talk about something that’s happening today that could make Solid Edge matter even more than it has in the past.
*I did this back in 1996, when I said that I thought Trispectives (the predecessor to IronCAD) could be “the Pro/E killer.” The Trispectives folks put that quote in all their advertising (I think they may have even put it on their packaging.) I got a lot of ribbing about it too. Years later, Buzz Kross (the head guy at Autodesk’s Mechanical division) told me that quote made him think I was… I don’t remember the word he used, but it was probably interchangable with “nuts.”
Actually, I wasn’t saying Trispectives was likely to be a match for Pro/E in capabilities. I was trying to get across how its modeling paradigm could ultimately challenge the then-dominant parametric feature-based method, used by Pro/E—the undisputed market leader at the time.
Turns out, I wasn’t nuts. The Trispectives modeling paradigm was an early version of what we now call history-free direct modeling. And direct modeling did ultimately kill Pro/E, though not in the way you’d expect: In 2010, PTC, recognizing the significance of direct modeling, “killed” the Pro/E brand, and launched Creo, a new series of CAD programs that incorporate both feature-based parametric modeling, and direct modeling.