The CAD Insider

COMSOL Who? The Smartest Guys Get the Least Attention

COMSOL fills the room. The company’s annual North America conference packs in between 200 to 300 people in Newton, Mass., near Boston. COMSOL holds regional conferences around the world, including one in Bangalore in August. Upcoming conferences will be in Switzerland, Shanghai, Taiwan, Seoul and Tokyo. Conspicuously absent is COMSOL’s birthplace and global HQ, Stockholm, Sweden.

If ever there was a technologically gifted company that needed a sales and marketing boost, it would be COMSOL.

I’m in Boston—again—this time by invitation from COMSOL to attend my first COMSOL user conference. The night before, I had dinner with a Boston-based friend and mentioned the reason I was in town.

“COMSOL who?” he asked. Bostonians are not known for their subtlety. My friend’s whole career has been around design and engineering software.

“COMSOL… They do simulation,” I said. “They have an office nearby [Burlington], and they’ve had their annual conference down the street [Newton] for a couple of years.”

COMSOL CEO Svante Littmarck addresses the multi-discipliniary crowd of mostly engineers but also physicists and chemists. The big news was the COMSOL Compiler and v5.4.

Nope, not ringing any bells. Clearly he has some marketing work to do. Despite a force in the U.S. now equal to its original office in Stockholm, Sweden, and moving its CEO to America, it still lags far behind industry juggernaut ANSYS in revenue, seats and all-around impact and visibility. Both companies do Multiphysics, the term for applying numerical methods to solve a variety of 2D and 3D field problems, from the way molecules behave in a nuclear collider, to blood cells in an artificial heart and to space capsules re-entering Earth’s orbit—physics, chemistry, magnetics, stress, electrodynamics and so on.

Seriously, what can’t COMSOL solve?

“We don’t do crash analysis,” said Bjorn Sjodin, vice president of Product Development, conceding only one discipline to ANSYS.

Otherwise, COMSOL holds its own. ANSYS may be the first simulation company to broach a billion dollars in revenue—they’ve been around longer, Bjorn said—but seem to be neck and neck in technical abilities and scope of solutions.

COMSOL’s advantage, according to Bjorn, is the integrations of solutions. You do fluid flow (CFD) and apply loads to your FEA seamlessly with fluid structure interaction (FSI). The acronyms are in common use in the presentations and hallways like they were common English. It is one interface, regardless of discipline, solution or numerical method, whether CFD, FEA or another. COMSOL is even looking at quantum mechanics to solve what happens at the nanometer scale (10-9m) when continuous fluid flow math gives way to particle collisions and avoidance.

By comparison, ANSYS, with its bigger market share and marketing department, has made itself a leader in Multiphysics. It has done so with acquisitions. Its multi-discipline solutions are used with different interfaces and data files, Bjorn said.

COMSOL seems determined to not give any ground technically to ANSYS. Its crack team of in-house content creators, led by Dr. Valerio Marra, may be the only marketing team to be led by a PhD in engineering. Valerio and his team detail the triumphs of many COMSOL customers to all the rest. It’s no wonder every version of COMSOL News, the glossy print publication the company takes great pains to produce, is the most technical of any in-house publication we’ve seen, bordering on academic journals.

Nature Can Adapt Internal Structure – Why Can’t We?

(picture courtesy of www.digitalefolien.de) 

If you study the internal structure inside the head of a human femur, cutaway above, you can see the variations of the internal structure. The structure is very dense inside the surface of the ball, where most of the stress would be from contact with socket. The fibers of the structure get long as it delivers the load to the long part of the bone. The structure is sparse in the lower left of the ball, where there would be little stress.

You could not have designed it any better. Because you don’t have the software tools.

Image from Autodesk’s Netfabb product page shows and internal structure that is of uniform size and shape in all three directions.

So as we are used to making parts solid, but then asked to make them lighter and/or use material, we should consider an internal structure. CAD programs, if they offer that option, will give you a few choices of shapes and structures. They all seem to use the 3D lattice approach, which means geometrically perfect, made of uniformly sized cells or members and extending linearly and unwaveringly in 3 dimensions.

Not at all like our femur.

Wouldn’t it be nice if the design software would have enough insight to “grow” the internal structure, varying the cell size for example, decreasing it when stresses were high, increasing it when stresses are low? And, while we are at it, can we position the structure so that it lines up along the stress path? Once it is done, it can be subjected to another load case, which adds material to compensate for a stress flow – without taking away from the first solution. Each successive load case builds up the part so, in the end, it has grown up exactly as needed. It is no more and no less of the part it needs to be.

In the world of engineering and design software, we are increasingly seeing “upfront optimization.” Programs such as Dassault System’s Tosca and Hyperworks Optistruct do exactly this – but only for solid parts. I’ve not seen anything that can optimize shells with internal structures.

Should any vendor be able to create an internal structure that can accommodate various load cases with an adaptive internal structure, it would certainly have our vote for software of the year.

Simulating the Unsolvable: The Story of Alain de Rouvray

Alain de Rouvray, CEO and founder of ESI Group and winner of the Spencer Tracy lookalike contest, at ESI Summit in Milpitas, Calif.

All those who are shocked by the results of the recent presidential election should have listened to Alain de Rouvray.

The president and founder of ESI Group knew something was amiss with the polling that was forecasting Hillary Clinton as the winner. De Rouvray noticed that the polling relied on an unrepresentative sample size, mostly the urbane. The great majority of the U.S. was not being accounted for. When you corrected the predictions for the missing data, the result was Donald Trump. De Rouvray recognized this two days before the election.

It’s not the first time de Rouvray has found an answer that has eluded others. Another, the answer to how materials behave in previously unpredictable situations, led to the company he founded and still leads, ESI Group.

ESI was able to prove that it could successfully simulate seemingly insolvable situations. An early example was protection against solar wind with particles traveling an order of magnitude faster than previously analyzed—a situation that remained unsolvable even by the best minds at Los Alamos National Laboratory.

Now a 1,100-employee, publicly traded firm headquartered in Paris (de Rouvray is French, though he now lives in San Francisco), ESI Group has been anything but a smooth-sailing venture.

“It was a bit rough in the beginning,” said Amy de Rouvray, his French-born, UC Berkeley–educated daughter who heads ESI’s marketing.

De Rouvray had to fight off threats from newcomers whose methods, de Rouvray decided, were a little too similar to his own. After a seven-year legal battle, the courts agreed that this was unfair competition.

The short, 73-year-old de Rouvray will remind you of Spencer Tracy—with an accent. He, along with his family, retains 30 percent of ESI.

ESI is best known for PAM-CRASH, a specialized CAE program that lets auto makers predict what happens to cars—and their passengers—as they meet immovable objects.

“It was a problem nobody else was solving,” said de Rouvray. Or perhaps they could not. The way that sheet metal crumpled and parts cracked and how pieces went flying—it was all beyond the limits of small-deflection, stability and elasticity theories, around which all finite element analysis codes are built.

Classical solutions rely on continuum mechanics, said de Rouvray. That doesn't apply across discontinuities—like across cracks or when pieces fly off.

That was the sort of problem that no one deemed possible of solving. Except de Rouvray.

Once metal analysis was tamed, de Rouvray decided he would solve fabric behavior, specifically how a car’s airbags react. Think of the analysis of fabric that is crumpled into a wad then literally explodes onto a human. How do you solve for that? Impossible, said the entire auto industry, which was spending hundreds of thousands of dollars testing with real bags and fake humans.

“We solved the airbag problem,” said de Rouvray, with characteristic finality.

ESI IC.IDO virtual reality program immerses users into potentially large data files, such as aircraft, trains or assembly lines. (Image courtesy of ESI Group.)

“We were so good at it that Audi stopped testing [for airbags] and relied on ESI simulation,” said de Rouvray. It can run thousands of simulations in the time it would take just to get instrumented for a single crash test, according to de Rouvray.

ESI seems to have built on its expertise with crash test simulations and added a variety of other analyses, including modeling and viewing solutions.

ESI MINESET using a combination of machine learning and visualization to explore simulation results. (Image courtesy of ESI Group.)

De Rouvray is not done solving puzzles that others cannot. At ESI’s latest summit, held in the company’s recently opened Silicon Valley office, I saw several technologies being nurtured or in full bloom. ESI’s IC.IDO (get it?) is VR technology for full 3D immersion into the large models and simulation results. Its MINESET program will let engineers explore large datasets, such as those resulting from multiple simulation runs, to detect patterns and relationships. Last year, ESI acquired CIVITEC for its sensor simulation platform, Pro-SIVIC, which will simulate self-driving cars before or instead of road or track testing.

By incorporating new technologies to solve new problems, ESI Group seems to be continuing to keep up with its overall goal of supplying “virtual prototyping” solutions.

Is This the Fittest CEO in the CAD,CAM and CAE Industry?

MSC's Dominic Gallello swims and runs for distance

Dominic Gallello, CEO, of MSC Software, is 60 years old but is not acting his age. He has just swum across the Bosphorus Straits. There’s 4 miles separating Europe and Asia near Istanbul and every year, about 2,000 swimmers take to the water to cross it. Dominic covered the distance in an impressive hour and 34 minutes, finishing 7^th in his age category.  

Dominic swims the 4 mile wide Bosphurus Straits in an hour and 34 minutes, finishing 7th in his age category

I’ve known Dominic since his days with Autodesk, where he created and ran their mechanical division. Back then he was defending Mechanical Desktop and creating Inventor – and running long distances.  Close to home it was The Dipsea, one of the oldest foot races in America - with a start that has 688 stairs to climb and then it is uphill from there. Since then, he’s gone longer, higher. There were marathons. 

Dominic summits on the Alps on a marathon

Here is Dominic summiting in the Swiss Alps in Zermatt, looking none too worse for the wear.

But ground is hard and running on it grinds away hip cartilage. Undaunted, Dominic jumped in the pool. Marathons were replaced by open water swims. He participates each year in a number of ocean swim races in Southern California, near MSC headquarters in Newport Beach, CA.

 

Runs in the family. Dominic before the big swim with wife, Naoe, who is also a marathoner.

 

Crashing a Perfectly Good Car

LAS VEGAS (3DEXPERIENCE 2014) — Does it bother anyone that a brand new car has to be run into a concrete barrier? I am watching an Acura SUV ramming into a barrier in an offset frontal crash test. In slow motion, carefully-styled fenders, bumpers, never-used wheels, brakes and other parts become an explosion of debris. There goes a $40,000 vehicle.

 

The dummy is barely hurt. What really hurts is the actual cost of such a test. Dassault puts the cost of a crash test at one million dollars, give or take — a cost they claim to be able to cut by seventy-five percent with their simulation software.

Physical crashes may never go away, as we seem to be most assured with a real car crash, rather than electrons. Electrons can lie. But simulating even a few crashes could save a lot of money.

Dassault has succeeded in making crash testing about as real looking as you can get — or better. Watching a sequence onscreen at the show, I had to constantly ask if it was real or simulation. I am told to look for debris flying around, which is a telltale sign of the real thing. The RTT software (acquired by Dassault, now renamed 3DEXCITE) and the underlying analysis software (LS-DYNA) is not able to model disintegrating parts. But it is amazing what it can do. RTT used to create pretty pictures that auto companies could put on their brochures and websites. You’ve seen them. Super sexy stuff. Now it has been pressed into service for crash tests and has transformed how crash tests can be seen, in ways you could never do with simulation tools and certainly in ways you could never do in real life.

Simulation software allows you to see what is happening to the frame during a crash, unobscured by the sheet metal exterior.

For example, you can remove the concrete barrier for an unobscured view of the front end crumpling. You can remove the fenders and hood to see how the engine and components move during a crash. You can isolate the fuel system to see how it is being compromised. From automotive safety engineers to automotive executives and even the general public, 3DEXCITE offers a fascinating look at what is happening during the 150 milliseconds of a crash. It is clear and obvious and needs no explanation, unlike FEA tools.

Can You Believe What You See?

The graphics are scarily good. However, we should be on guard, for visual realism is no substitute for physics. It may look real, but it may not necessarily be true behavior. Analysts know how wrong analyses can go. A wrong assumption, an extra zero and incorrect material may still yield believable graphics but will not correlate with physical results. This is of particular importance in cases where our intuition is of no help. We have no frame of reference for materials and systems during impact. We are likely to believe what we see.

From Their Cold, Dead Hands: How to Pry Users from their CAE Programs

From Their Cold, Dead Hands: How to Pry Users from their CAE Programs

LAKE OSWEGO, OR - Autodesk’s NEi Nastran acquisition gives the company instant credibility in the CAE (computer aided engineering) community. No longer can CAE users dismiss Autodesk as that pretentious CAD company that did little more than assemble a rag-tag collection of little-known analysis applications.

Nastran is serious CAE. It's what the big boys use. It puts planes in the sky and cars on race tracks. I'm not sure I would have boarded the Boeing 737 on which I am writing this had I know Algor¹ was used in its analysis.

Nastran NEi used for Cessna. From NEi website.

But now the acquisition of NEi Nastran changes the game. Your PhD friends will not laugh at you for using Autodesk software. You can just tell them "It's Nastran, dammit!"

CAE applications garner loyalty beyond that of CAD users, owing to the fact that CAE applications take much longer to learn, are harder to use, require higher educational qualifications, and cost a lot more money. These factors combine to create an atmosphere of exclusivity and clubby camaraderie. No serious CAE user is going to switch CAE programs without a fight just to save a few bucks.

Autodesk admits it will not be able to pry current CAE, except from "cold dead hands," but also shows no panic at the prospect. Autodesk can afford to be patient, and wait, because CAE is not their core business. They can sell their CAE products all around the entrenched CAE users and to CAD users. But the lower price, when coupled a deep and protracted commitment to the market, can't help but over time erode the hold of CAE companies of even their most stalwart supporters.

Commitment is really is the key; that low initial cost of the software is secondary. Autodesk has not yet shown it can support CAE users in the manner to which they hard core analysts are accustomed. Will users be able to pick up the phone and get a reseller who has a PhD in mechanics and so can tell users when to use Von Mises failure theory -- from years of experience? Both ANSYS and MSC provide this level of support.

Should you have been paying attention in your Mechanics of Material classes, you may not need support for basic questions, but how many would you build the next Boeing aircraft without an advanced level of backup?

Can Autodesk fly his plane? David Weinberg, founder of NEi Nastran, and Lt Colonel, Aircraft Commander at USAFR. From LinkedIn.

CAE companies instilled and maintained a culture that worships theory, academics with advanced degrees, deep knowledge, years of experience, cultivation of engineering judgment ... qualities that narrow the field of prospective support personnel. CAD companies, on the other hand, set the opposite tone: ease of use, democratization, ... everything that CAE is not.

As the products of CAD companies and CAE companies converge (see ANSYS Acquires SpaceClaim...), note that eventual success depends not only on product portfolios or the pricing (no matter how low they go), but on something that is much harder to establish: credibility, built on trust.

Market leaders establish trust after providing years of excellent service and great products. And so if indeed an analyst can nail that new composite tailfin panel analysis, if he or she really can do it in less time, or can run it from an iPad, if I can be cool and still hold my head up among my peers, or we can be sure the application has already kept planes in the air... yeah, then, maybe we will try it.

See Also

Autodesk Nastran - Autodesk product page

---

Footnote

1. Algor was a midrange analysis program purchased by Autodesk in 2009. Although a general purpose solver, it never achieved the status of Nastran or ANSYS. It is currently called Autodesk Simulation.

Autodesk Flow Design the Easiest CFD Program Ever

At an Autodesk press event for another purpose (see Autodesk Buys..NEi Nastran...), it was Flow Design that stole the show.

CFD has never been so easy. Watch airflow separation increase with speed over the back of a Porche in this YouTube video of Project Falcon, now available as Audodesk Flow Design.

Announced last January with little fanfare (Flow was mentioned in an Autodesk blog post in January, and then picked up the same day in TenLinks Daily), there never an official press release, no phone calls or webinars for the media. Even Autodesk's own product site does not do it justice. Just another Autodesk Labs product done good.

The existence of a bona fide CFD (computer fluid dynamics) program for wind-tunnel simulation -- push-button easy and available at a rock-bottom price -- dropped like a bomb in the room of industry analysts and journalists. Our existence depends on knowing everything, but, well, we got caught flat-footed. A possible game changer, and we missed it!

How Much?

The fact that it is available for a steal, $210 a year, was by itself a shocker. One industry analyst remarked that a low price reflects badly on the product. After all, a competitor’s product sells for $80,000. At 0.25% of the price, no one was going to take Flow seriously.

History

Based on CFD technology from the Blue Ridge Numerics acquisition, Autodesk’s easy flow simulation was first shown at Autodesk Labs under the name of Project Falcon, back in September, 2012. Today, Autodesk demonstrates the technology in its Gallery on San Francisco's Market Street. There you stand in front of a Xbox Kinect scanner, push a button, and in a few seconds, it shows you your 2D profile with colored streamlines moving over it -- for what that's worth.

A little more useful demo was when they had a bicycle where you could hone your aero tuck. (See this YouTube video.) Most of a bike rider’s energy on level ground is spent overcoming air resistance.

One Trick Pony

But I digress. Did I remember to say it takes only seconds to solve problems? 

More importantly, it takes no time to set up. Flow takes shapes from a variety of objects, photographs, drawings, scans. If you are used to constructing a mesh of volume elements painstakingly, you now have a lot more time on your hands; with the speed and simplicity of Flow, you might consider a hobby -- or retirement, because you might be out of a job.

Oops: I got a bit carried away. Flow is not a full-fledged CFD code. It handles only wind tunnels. And it makes a lot of assumptions. While air speed can be adjusted, an arbitrary speed is assigned initially.

But supersonic flow? Careful boundary layer planning? I don't think so. Coupled heat transfer? Combustion? Nope. I guess some of you CFD eggheads can keep your jobs.

Get Ready to Justify Your CFD Program

But if you just got your boss to sign a purchase order for a new multiuser license for Fluent, Exa, COMSOL, and the like for who knows how many tens of thousands of dollars or Euros, you better have some pretty good answers as to just how is your favorite software so much better as he shows you what he can do on with Flow on his laptop. 

Not Just for Planes and Autos

 
How would a new structure affect wind patterns in London's financial district? Even architects can now use CFD to find out.

The simulation group at Autodesk is keen on making Flow Design available not just to traditional users (mainly in aerospace and automotive) but also to other industries that Autodesk serves. Architects, not typically CFD users, and certainly not mandated to use high-end simulation software in school or in practice, would now be able to easily run air flow studies, such as the effect of wind around a new high rise.

See Also

Project Falcon Graduates from Autodesk Labs to Autodesk Flow Design - announcement that Flow was now a commercial Autodesk product, Scott Sheppard, It's Alive in the Lab, Jan 15, 2014

Flow Design - Autodesk product site

 

Autodesk Buys Instant CAE Credibility with NEi Nastran Acquisition

LAKE OSWEGO, OR -  Autodesk made it quite clear this week they are serious about simulation. At a small press event, they revealed Nastran-based products. See news release. One is a standalone FEA (finite element analysis} solver, called Autodesk Nastran. The other is in-CAD, which will work inside Inventor and in SOLIDWORKS. 

How Not to Handle a Leak

The products are a result of Autodesk's acquisition of NEi Software. It would have been news delivered with a thunderclap -- if it had not already leaked months in mid-May. It seems NEi customers and resellers mentioned "certain assets" of NEi being acquired by Autodesk. Further reports mentioned Autodesk obtaining NEi Nastran code but nothing else. Since the news was out, Autodesk did post about they would be incorporating NEi code but said nothing about other NEi assets. The fate of NEi Software, the company, was unclear.

Oh, Yeah. We Bought NEi

And so it was on an uncharacteristically hot day (90 degrees in Portland!) Autodesk assembled a half-dozen simulation-savvy but befuddled analysts and journalists to its Lake Oswego offices for the official announcement. Autodesk had indeed purchased the entire company.

The acquisition cost was not revealed. NEi employees (numbering in "double digits") are now Autodesk employees. NEi will continue to operate in their Westminster, California offices -- not far from archrival MSC Software. NEi Software founder CEO David Weinberg was not seen at the media event, although I was told he is still with the group, and was very active during the four months it took to create in-CAD for Inventor.

We Told You We Were Serious

More importantly, Autodesk enters the big leagues of CAE, such is the cachet of Nastran. With the acquisition of NEi comes customers of high-end FEA software, something Autodesk may had trouble acquiring on its own. NEi uses a hybrid direct/reseller sales network, which Autodesk will take over.

(Nastran is one of the classic FEA solvers, in use by every major aerospace firm, big auto, big enterprise -- and those who wish they were. Originally developed as a NASA project, it was picked up by a number of private companies and made commercial. MSC Software is currently the leading incarnation of Nastran with MSC.Nastran, both in terms of market share and robustness. NEi may be in the top five.)

Why NEi?

Though a general purpose solver, NEi has tried to build a reputation for composite material analysis for aerospace (ridiculously more complicated to analyze than homogenous isotropic materials, such as metals) and non-linear analysis (a specialized area in which materials get weird like with viscoelastic materials such as rubber and certain plastics, or even ordinary materials when taken to extremes in temperature or load.)

Autodesk assured the skeptics among us that it will continue to support and develop in-CAD for SOLIDWORKS, which was previously named NEiWorks. Autodesk Nastran products will sell for "around $10K."

See Also

Autodesk in the Big Leagues with Nastran - Monica Schnitger, Aug 12, 2014

Autodesk Releases Official Statement about NEi Nastran Purchase - John Evans, Design & Motion, Aug 13, 2014

Give Up Making CFD Meshes - Converge is Better and Faster

COLORADO SPRINGS, CO (CIC/NAFEMS 2104) - The mesh is critical to the speed and accuracy of a CFD analysis. Make too coarse of a mesh of your control volume and results near abrupt shape changes will not be accurate. Make too fine a mesh over the whole volume and the analysis will take forever. You could do a couple of runs to determine zones each corresponding to to different sizes of elements. Or you could get Converge to refine the mesh for you -- and I'll bet Converge will be faster.

Converge solves combustion problems with CFD using a mesh that refines itself on the fly
(picture from Convergent Science site)

It's so fast that it can solve rapidly changing volumes such as those occurring in combustion chambers of engines, where the dyamimic shape of the flame must be modeled and accounted for. If that's not impressive enough, it can calculate and adapt meshes in supersonic shock waves from jets.

Converge utilizes what it calls Adaptive Mesh Refinement (AMR) to reduce the size of elements in critical areas. Because AMR is coupled with the solver, it refines the mesh during the analysis. You don't have to wait a couple hours to see problem in the results and rerun it. You set the level of refinement before the run. One level means the elements are reduced in length by a half, two levels and the elements are halved again, and so on.

On the Fly Mesh Refinement

You start with a coarse mesh:

Coase mesh to start... (picture from Convergent Science site)

Converge takes one look at your rather amateurish mesh and decides to start splitting up elements in areas that seem to need it, either around abrupt changes in geometry or where there is a large delta in your critical variable, in this case velocity of gas. The color changes indicate analyis has already begun.

Converge is already starting to refine the mesh in critical areas...(picture from Convergent Science site)

The control volume mesh will keep refining itself after successive runs until it reaches a preset limit. Here is a highly refined mesh after multiple runs. Note the relatively small size of elements in the narrow passage around the valve and a velocity profile suggests a more accurate solution.

Converge will keep refinining the mesh (picture from Convergent Science site)

Founded in Caterpillar

Converge had its beginnings in Caterpillar, where it was used by the construction vehicle maker to model the chemical reactions in their engines. Converge spun out of Caterpillar and can do general purpose CFD analysis, where there is non-reactive fluids and external volumes (example: flow over a wing section).

"AMR is most suited to transient solutions," says Rob Kaczmarek, Converge's director of sales and marketing, meaning the types of volumes that would change quickly. "We're seeing about 4 times faster than our competition in run time alone -- and that's not even including the time saved by not having to create a mesh!" He names ANSYS and CD-adapco as competitors. Rob claims Converge holds its own with most steady state problems.

Easy to Use Templates

You can set up Converge with templates so users who may not be savvy with CFD can answer a few questions and get results. Converge can be 'bought' as needed, for a project, for example. It will work on the Rescale platform, which does its calculations on HPC, so you don't even have to have a monster workstation.

Clearly, this is not your father's CFD.

For more information

Convergent Scienes website http://www.convergecfd.com/

ANSYS Buys SpaceClaim. What Were They Thinking?

The CAD world expected a lot from SpaceClaim. A promising startup founded in 2005 by none other than Mike Payne, whose previous companies, PTC and SolidWorks forever changed mechanical design. SpaceClaim, the product, looked good, it was easy to use and with direct modeling the trending topic in MCAD, it seemed to be in the right place at the right time. On the product side, CAD insiders gave it a thumbs up.

Despite all its promise, however, it failed to take the world by storm.

Some would say it faltered out of the gate. Its subscription pricing came was ahead of its time, and was eventually abandoned for a more conventional perpetual license model. Then it proved to be difficult to unseat SolidWorks and Pro/E users. SpaceClaim's "adjacent seat" tactic really meant you had to to have two CAD systems. That's one more than the number CAD users want. Then they were a front end to CAE software. A front end to CAM systems. Even a ECAD (electrical CAD) system. Maybe users would use it to read, to convert from one CAD system to another, be a super viewer of CAD data... the tactics seemed to vary with the weather.

As late as December of last year, SpaceClaim claimed to have 40,000 seats. Not bad. But not enough to change the big picture. Even perennial contender Solid Edge had risen to an order of magnitude larger.

Co-founder Leaves

It doesn't look good when a founder leaves for a day job with another company in the same space. Co-founder Blake Coulter left SpaceClaim in early 2013 and surfaced at GrabCAD.  Marketing had been lackluster, consisting mostly of Blake tirelessly tweeting and talking his online friends into reviews of the product, plus the occasional press release about key customer wins. Bernie Buelow, VP of Marketing left not long after Blake. The position remained unfilled.

ANSYS the White Knight

Then ANSYS buys them. $85 million may seem like a lot of dough at first, but compare that what Dassault paid for SOLIDWORKS, over $310 million SolidWorks 17 years ago (1). At the time SolidWorks had only 6,000 installed seats. That's over 3.6 times as much as SpaceClaim -- for less than one sixth the number of seats.

Was it the channel? CAD insiders say Dassault bought SOLIDWORKS for the reseller channel they had set up. But also, SOLIDWORKS was already perceived as a company on the rise.

Already in Bed

SpaceClaim was already in bed with ANSYS. SpaceClaim had signed up with ANSYS as their modeling front-end. It made sense, as modeling was hardly the strength of CAE programs, whereas it is a SpaceClaim specialty. But why buy a capability you already have access to? And why pay $85 million for it? I doubt it would be justifiable in savings of license fees.

An Answer to Autodesk

The answer of why ANSYS bought SpaceClaim may have more to do with Autodesk's actions of late. Autodesk, mostly a CAD company, has been buying up CAE companies to the tune of $300M. CAE companies collectively have been shrugging this off. The seemed content, almost complacent, failing to notice rising temperatures, making excuses for inactions: What does a CAD company know about analysis? Who would believe them? Their channel cannot support it. Not like us, who know what we are doing. We have real solutions. Our results are precise. We have PhDs writing our software. And PhDs using it. What does Autodesk have?

Such an encroachment did not go unnoticed by CAE market leader ANSYS, who saw Autodesk enjoying an unopposed march deep into its homeland. If Autodesk was going to buy up companies in the CAE space, well, they would buy up a company in the CAD space. Just as Autodesk could hope to migrate their CAD users to their newly acquired CAE tools, why couldn't ANSYS buy CAD users to convert them to ANSYS? Two could play this game.

The most likely CAD users to be interested in what ANSYS offers are MCAD users, many of them are already mechanical engineers, in which case their education would involve FEA and other analysis tools. No doubt ANSYS went shopping for MCAD users -- and had to look no further than SpaceClaim, a company they had already partnered with.

See Also

• 3 Key Questions as ANSYS Acquires SpaceClaim - tighter integrations to enable engineers, full-blown CAD offering, parametric driven design studies, Chad Jackson, Lifecycle Insights, May 6, 2014
• ANSYS Buys SpaceClaim (Finally) - for $85 million in cash, expects SpaceClaim to have revenue of about $14 million in 2014, Monica Schnitger, Schnitger Corp Hot Topics, May 1, 2014
• ANSYS Acquires SpaceClaim: May the Force(s) Be With Them, Jeffrey Rowe, MCADCafe, May 1, 2014

---

(1) Dassault Systemes Signs Definitive Agreement to Acquire SolidWorks, June 24, 1997