Autodesk adds Nastran to its CAE offering

EarningsAutodesk confirmed yesterday that it has acquired certain assets of NEi Software, including its flagship Nastran solver. Autodesk isn’t ready to say anything about its plans for Nastran but we can speculate that it wants to add its powerful linear and nonlinear as well as dynamic response capabilities to the solvers  acquired over the years with Algor,  Moldflow and CFDesign.

Solvers are tricky beasts. They’re the workhorses of CAE, solving huge matrices with various approaches for memory management, matrix simplification and rearrangement, convergence tolerances and so on. Without solvers, simulation grinds to a halt.

Why does Autodesk need Nastran? Good question. Algor was always seen as a powerful but midrange product; rebranded as Autodesk Mechanical, it continues to have that image and a relatively small reseller network. NEi, on the other hand, has over the years evolved its version of Nastran to include their own nonlinear capability and aimed it more at the high-end nonlinear user community. NEi also created and marketed the NEi Nastran Editor, a tool that gives non-specialists greater control over their FEA model setup. It’s not immediately clear if Autodesk also acquired this or other NEi products.

Autodesk is going to share more information soon including what, exactly, it acquired, whether any NEi staff joined Autodesk and what it plans to do next in simulation. We can think apps, cloud, on-demand, new user interaction models … ooh.

I think it’s exciting and yet again validates how pervasive simulation will eventually be. Autodesk’s reach, channel and direct, its presence in markets that NEi didn’t address on its own … It’s going to be cool.

UPDATE, 15 May 2014: Several readers wrote in after this post went live to say that NEi Nastran isn’t actually Nastran, as in  based on the original NASA-sponsored code. It is, however a solver that is often benchmarked against those flavors of Nastran (and others, like ANSYS, Abacus, etc.) and adds similar capabilities to Autodesk’s offering. There’s a bit more in today’s post.

I wish I could tell NEi customers what to expect in the near term, but I can’t. Autodesk is releasing very little info right now. I suggest you contact your NEi rep and see what they say.

Print Friendly


5 Responses to Autodesk adds Nastran to its CAE offering

  1. Dennis Nagy says:

    “…the solvers acquired over the years with Algor, Moldflow and CFDesign.”
    And Plassotech.

  2. Dennis Nagy says:

    NEi NASTRAN is a “facade” NASTRAN: it looks like NASTRAN from the outside (input/output formats and terminology) but the solver “inside” is not NASTRAN (in contrast with MSC.NASTRAN, NX.NASTRAN, and the two older “real” NASTRANs — UAI/NASTRAN and CSA/NASTRAN — that were acquired by MSC in the very late 1990s and led to the expensive FTC action against MSC and the birth of NX.NASTRAN).
    So my guess is that the plug-compatible facade is worth something, even on its own. Given the size of NEi (compared to the size of ALGOR when acquired by Autodesk), the facade might even be usable surrounding the ALGOR solver(s).

  3. Katarina says:

    NEi Nastran has been setting the pace for FEA without use of other software; for example, nonlinear in MSC.NASTRAN is done through Marc and NX NASTRAN uses Adina, compared to NEi Nastran that continues to innovate with insight and control of the analysis.

  4. Dennis Nagy says:

    Hi all,

    It has come to my attention that some clarification is clearly in order about my comment above.

    My use of the word “facade” was in no way meant to be critical or negative. I was just trying to point out (to many people with whom I’ve talked over the past decade, in fact) that there is a basic difference (but not that either is better or worse than the other) between NASTRANs based initially (or more recently in the case of Siemens PLM)in some way on the “original” NASA/COSMIC NASTRAN in their inner structure and workings, vs. those based on plug-compatibility in the I/O file structures (NASTRAN Bulk Data)and command language (DMAP), i.e. in the outer shell (batch user interfaces)or “facade”.

    Katarina Weinberg is correct in noting that the other two successful NASTRANs (MSC and Siemens PLM) needed to use other (acquired or third-party) nonlinear simulation capabilities to augment the MSC.NASTRAN core capabilities which were always weaker(er) in nonlinear simulation (one of the reasons I negotiated a 5-year re-seller agreement with David Hibbitt in 1994 for MSC to re-sell ABAQUS, prior to the MSC acquisition of MARC).