Evan on Engineering Software

Over on WorldCAD Access, a reader commented that he thought Autodesk, to move higher in the MCAD world, might acquire some CAM technology.

It's an interesting thought.  And, given the fierce competition in the mainstream CAD market, it's intriguing to think what Autodesk might do.

But I don't think Autodesk management would be dumb enough to get into CAM.

The CAM market is incredibly fragmented.  I can think of a bunch of significant CAM vendors, including (in no particular order whatsoever) Cimatron, Teksoft, CNC Software, UGS, MCS, Delcam, Gibbs, CGTech, DP Technology, Planit, Vero, Pathtrace, DTL Mori Seiki, PTC, SurfWare, VX, Open Mind Technologies, Auton, Missler, Hitachi Zosen, Dassault, Machineworks, and SolidCAM.  Of course, that's just off the top of my head.

CIMdata lists UGS, a full-line CAD/CAM vendor, as the worldwide largest CAM vendor. But UGS is a rather special case.  They started life as one of the very first CAM vendors, blossomed as a CAD vendor, then grew to become one of the troika of serious PLM vendors. 

IBM/Dassault is listed second on CIMdata's list -- based not purely on their revenue from selling CAM software, but rather on total end-user payments for NC (CAM by any other name) software and services.  Which is to say, IBM is masterful at generating services revenue.

PTC, the other notable PLM giant, is also listed by CIMdata as being one of top CAM vendors, but they're somewhat down the list.

The thing that UGS, IBM/Dassault, and PTC share in the CAM world is an enterprise orientation that's top-heavy with legacy customers.  None of these companies have any presence, to speak of, down at the local machine shop. 

Possibly, the key to understanding the CAM market is that it's inherently legacy oriented.  People who run factories full of multi-axis machining centers tend to use what works.  They care about cycle times, tool life, and finish quality more than software integration.  Consider, for example, the nuclear market.  It's a niche market -- requiring special capabilities that are simply not up for negotiation.  Within this market, you'll find a bunch of hard-core dedicated Anvil 5000 users. 

Now, I may be one of the few guys under 50 who truly appreciates the genius of Dr. Patrick Hanratty, who is arguably the father of both CAD and CAM, and to whose influence (if not direct work) many of today's major CAD/CAM products can be traced. (And, I gotta digress here -- As much as I am in awe of Ivan Sutherland's Sketchpad, Hanratty's work in the field predated that of Sutherland, and unlike Sutherland, Hanratty has spent his entire career in the field, and is still doing remarkable work today.)  In any case (back to my point), people who make things that can destroy a city (or not) if they fail tend to be picky about their CAM software.

Yet, enterprise and exotic markets are not the mainstream for CAM.  Within the mainstream, the most successful CAM products come from CAM companies -- not CAD companies.  Because PLM, by its nature, is a hybrid combination of interlocking processes, the people who write checks for CAM software tend to be different than those who write checks for CAD software.

Alan Christman, chairman of CIMdata, wrote a succinct summary of technology trends in CAM software in MMS Online, and the only real trend towards CAD/CAM synergy he mentioned was in solid based machining.   And, though a technical synergy, it can be accomplished through API integration.  It doesn't benefit from a true integration into CAD any more than, for example, CAE would.

I think it's a fair argument to say that, for an application vendor such as Autodesk, owning their own CAM product would do more to scare off their CAM partner vendors than it would to help them win against their mainstream competitors.

Now, if I'd like to take a complete flyer, and wildly speculate as to which company Autodesk might acquire to move further into the high-end of the CAD (PLM?) market, it would be MSC Software -- though their business partnership relationship with IBM/Dassault would make it rather interesting.