Chapter 4 - Solid Edge versus SolidWorks: Two Different (but similar) Paths to Parasolid

Another fascinating aspect is the parallel development of Intergraph Solid Edge and SolidWorks in the late 90s. Both companies saw the massive potential of the more powerful PCs and the Microsoft Windows NT operating system as ways of breaking into the mid-market space where UNIX machines were just simply too expensive and cumbersome. They would eventually settled on Parasolid for similar reasons, but their journeys from there reveal starkly different corporate cultures and decision-making processes.

Solid Edge: From ACIS to Parasolid

Intergraph initially launched Solid Edge with the ACIS geometric kernel, developed by Spatial Technology. However, as the product matured, Intergraph's engineering team encountered scalability and adaptability challenges that threatened the platform's future growth. This is how Bill McClure, head of Solid Edge at the time, described the situation to me:

Initially, Solid Edge faced significant performance and reliability issues with some of the key functions of the ACIS kernel.  To address this, I initiated a clandestine "skunkworks" project, known only to myself and three other team members.  We even rented an apartment to work in secret, away from the office.  Our rapid evaluation of Parasolid quickly revealed its superior performance and reliability, solidifying our decision that we needed to switch Solid Edge to Parasolid. While the team continued to work on the Parasolid implementation offsite, I faced a challenging review meeting with the sales management team. The VP of Global Sales was highly concerned about the numerous customer and Application Engineer complaints regarding Solid Edge's modeling problems. The next day, Jim Meadlock (Intergraph CEO) demanded a solution. I revealed our secret project, explaining that our Parasolid implementation showed excellent test results.  I emphasized that switching to Parasolid was crucial for our survival in the Mechanical CAD market. Jim not only endorsed the move but also suggested we broaden our discussions with Unigraphics to explore a potential joint venture. This pivotal decision ultimately led to the acquisition of Intergraph's Mechanical Software Division by Unigraphics Solutions.

This transition occurred before Unigraphics acquired Solid Edge, setting the stage for the product's integration into what would become the Siemens PLM portfolio, a fact that might be surprising to those that assumed falsely that the move to Unigraphics was responsible.

SolidWorks: Granite Denied, Parasolid Adopted

SolidWorks' kernel story reveals the sometimes-personal nature of enterprise software decisions. There was initially a prototype built on ACIS, but due to similar issues that Solid Edge has seen in their experience, they tested Parasolid as well. SolidWorks also approached PTC's CEO Dick Harrison in 1995 with a request to license PTC's proprietary geometry engine—the same geometric engine powering Pro/ENGINEER. Harrison declined to commercialize it. Mike Payne told me the story this way,

We started SolidWorks using a trial copy of ACIS, but it was full of bugs. Can you imagine a graphics kernel at the heart of your code bleeding memory like a stuck pig? I reached out to Dick [Harrison, CEO of PTC at the time] and asked him if we could license the code for the geometry engine from Pro/ENGINEER. He gave me the side-eye and said, ‘But we don’t do that, sell the engine, I mean.’ I countered, ‘That doesn’t mean you can’t start doing it now, though.’ He just stood there and after a beat said, ‘But we don’t have a model for selling it.’ So, that wasn't going to happen. As it turns out, I had already created libraries in parallel for plugging either ACIS or Parasolid into SolidWorks and found that Parasolid fixed most of our bugs and was much faster, so the decision to switch to Parasolid was easy. As time went on, Pro/ENGINEER would refuse to benchmark against us, so I guess that tells you how it worked out for SolidWorks at the end!

Harrison's refusal would prove consequential. PTC never built a model for commercializing their graphics engine because they didn't want to become an OEM for software. They preferred to focus on their core products and offer APIs for partners and customers to build on top of them (see the PTC Granite chapter above). Faced with this rejection by Dick to license the PTC graphics engine, SolidWorks signed a contract with Chuck Gridstaff at Unigraphics and adopted Parasolid, joining what would become a growing ecosystem of Parasolid-powered applications.

Author's Note: When I was working on this article and gathering these testimonials, it turns out that although Mike and Bill knew each other, they had no idea that each had struggled with simular ACIS problems (bugs and performance issues) and reached the same conclusion (they went to talk to Tony Affuso and settled on Parasolid). Both were surprised and amused when we talked about it. It is a small, weird world, the CAD/PLM world, for sure!

2025 Update: The Pattern Continues

In the opening months of 2025, ANSYS migrated SpaceClaim—originally developed by Mike Payne in 2005 and acquired by ANSYS in 2014—to theParasolid kernel, following in the footsteps of SolidWorks and Solid Edge in moving away from ACIS for improved robustness and interoperability. The product has since been rebranded as ANSYS Discovery.

The Great Divergence: Sales Strategy as Destiny

While both products shared similar technical foundations and target markets and released within a few months of each other, their sales strategies created vastly different trajectories. This divergence would ultimately determine which company would capture the larger share of the exploding mid-market CAD opportunity.

SolidWorks made a bet that would define its success: a channel-centric sales model. Rather than building a large direct sales force, the company partnered with regional resellers who could provide local support and relationships. This strategy proved remarkably effective, enabling rapid geographic expansion and customer acquisition at a fraction of the cost of direct sales.

The results were spectacular. SolidWorks' growth rate outpaced Solid Edge, establishing market momentum that would prove difficult to reverse. By 1997, just a few years after launch, SolidWorks had attracted the attention of Dassault Systèmes, which acquired the company in a deal that would transform both organizations.

In contrast, Solid Edge used a traditional direct sales model and a smaller channel model and never achieved the explosive growth that SolidWorks experienced. Its acquisition by Unigraphics (later acquired by Siemens) positioned the product within a comprehensive PLM portfolio. While it never achieved SolidWorks' market dominance, Solid Edge found its niche as part of Siemens' broader industrial software strategy, particularly in manufacturing and engineering workflows.